42TAMS-ULG
BARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTERPLAN PROJECT
P O Box 3830 ADDIS ABABA ETHIOPIA TEL (251) 1 15-93-24 , 15-94-64, FAX (251) 1 61-32-79
---- _ ■■ - — ■  ~=—— _
December 6, 1995
Ain Solomon Kassahun
Baro A kobo Master Plan
Project Manager
Addis Ababa
Dear Sir,
BARO-AKOBO RIVER BASIN INTEGRA TED DEVELOPMENT
MASTER PLAN INCEPTION REPORT
We have pleasure in submitting lhe Inception Report for the above project. The report is in three volumes as follows:*
Volume I ; Main Report Volume 2 
Review or Previous Reports on Resources Development
Volume 3 
Data Volume
Volume 1  is essentially  a self sustained  volume since it contains an  overview of the previous study  findings More details of this review arc given  in  volume 2.  Volume 3  is devoted  to  data on  water resources,  socio  economics and  water resources projects that was too  voluminous to include in Volume 2.
It will be noted  in  section  5  of Volume I that we are recommending  additional data collection early  in  the phase II studies.  We await your reaction  to  this proposal but meanwhile in  section 6  ot Volume 1  we have included  two  scenarios for phase II activities,  one assuming  that the additional data collection  is agreed  and  the other assuming  no  additional data collcciion  is undertaken to that presently defined in lhe scope of services.
We look forward in receiving yuur comments in due course so that we can finalise the Inception Report.
Yours faithfully
TAMS-ULG
Project Manager
TAMS Coniultanu. Inc.
Th* TAMS Building 655 Third Avenue New York. NY 10017CONTENT
VOLUME 1: MAIN REPORT
1.          INTRODUCTION ............................................. ............................................................. M
1.] Background, ..., ,.................................. *.......................................-........................- - I-1
1.2         Project Area. . .,,............... ............................................................... 
2.          OVERVIEW OF PREVIOUS STUDIES............. ........................................................... 2-2
1-1
2.1        The Russian Study ( Selkhozpromexporl, 1990)................................ ................. -.        2-9
2.2        The EVDSA Reconaissancc Study Report. (1990).......................................................  2-9
23        The EVDAS Socio-Economics Report, (1993)..........................................................     2-9
2-4       ARDCO-GEOSERV Study. (1995)...............................................................................  2-9
3.          POL ICY REVIEW AND STRATEGY FORMULATION..................... .........................  3-1
3.1        Introduction. .......... ............ .................. ...........  ........... .. ............................. .. 3-1
3.2        Strategy Formulation. ... ................... .....................................*..................... - -..            3-2
3            3 Water Resources Planning . ......................................... ............... .............. —.           3-3
3.4        Principles for Natural Resources Planning............ .... — .. ............ ...........................  3-4
3.5        Indicative Master Plan...................... ............ .............. - -.............. .. — . ♦---------- 3-5
4.         PHASE II METHODOLOGY............................................................. ................ .............  4-1
4.1        Introduction........ .......................... . ♦ ♦....................... ........................... ..  4-1
4.2        Natural Resources and Agriculture................. .................... .. .................... ............  4-1
43 Water Resourses.,. ................................................ ............... ... .......... ..................... ... +.           4-4
4            4 Socio-Economics. , ,.,, ♦................... ... ................ . _................... ... ..................  4-27
4.5       Environment................................. ............ ................. .......................................  432
5          RECOMMENDATIONS FOR DATA COLLECTION AND PROCESSING.... 
5.1        Data Collections and Processing Needs.... T,................................................  5-1
5.2        Additional Field Data Collection..............***►♦............ ................................. ..« ..
+
5-1
5-1
53 Collection and Processing of Hydrological Data.............................................. ,....................5-3
5.4       Long Term Data Collections Program.................. .. .............. .............................  5-3
5.4       Phase II Dau Collections Program............................. .. . .............. ... ...........................  5-4
6           TASKS AND MANPOWER PLANNING FOR PHASE 11.....................................
6.1 Shedule of Tasks.......................... .......
63 Schedule of Staff...................... ................ ..
BIBLIOGRAPHY
■
6-1
6-1
6-2
VOLUME II: REVIEW OF PREVIOUS REPORTS ON RESOURCES DEVELOPMENT
1 NATURAL RESOURCES................................................................................................................... M 1.1         Introduction..................................................................... ................................. .. .... ..  1-1
TAMS-ULG UARO-AKOBO RJVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN           iCONTENT
1.2       Geology and Minerals.................................................. ............................ ...... I -1
1.3       Soils...................................................................................................................... 1-7
1.4 Land Use, Land Evaluation and Lar.d Use Planning. .. ,.......................... ..  .1-16
1.5      Forests ... ............................................................................................................    1-29
1.6       Wild Life................................................................................................................ I*32
1.7       Tourism............ ...............................................................................................      1*38
1.8       Fisheries...........................................................................................................
1.9       Apiculture............ .. ......................... -...................................................................  '*44
1.10     Mining. . ..................... .............................................  ......... 1-46
1.11     Energy Resources Survey...................................................................................... 1-47
2. AGRICULTURE........................................................................................................................
2.1        Introduction........................................................................ .. ..............................  2-1
2.2       Crops Production....................................... .................................................... ..         2-1
2.3       Livestock...................... .................................................... .................................. 2_14
3, WATER RESOURCES...........................................................................................................  3-1
3 1       Climatology and Hydrology............... ....................................................................  3-1
3.2       Groundwater Resources................ ..........................................................................  3-6
3.3       Flood Protection Projects............... .. ................................................................      3-18
3 4       Irrigation Projects................ ................................................................................    3-20
3.5       Hydropower Projects..... .......................................................................................   3-28
3 6       Waler Supply Projects............................................................................................. 3-37
4. SOCIO-ECONOMICS...........................................................................................................  4-1
4 1       Introduction............... ........................................... .. ................................ ..  4-1
4.2 
Local Government Administrative Structure........................................................... 4-2
4.3 
Ethnics Groups.....................................................................................................     4-4
4.4 
Population ............................................ .. ................................................ .. 4-5
4.5 
Health... ................................................................................................................... .4-6
4.6 
Education.................................................................................................................  4-9
4.7 
Rural Infrastructure...............................................................................................  4-10
4.8        Commerce.............................. .. .............................. ...................................,.. . . .4-13
5          ENVIRONMENT.....................................................................
5.1       Background and Objectives...... ..................................
5.2       Structure and Approachof Environmental Section
5.3        Environmental Perspectives of the Project Area..
5.4       Current Patents of Land Use................... ...................
5.5        Environmental Concerns.............. .. ....................... .. ..
5.6       Principal Environmental Issues...................................
5.7       Towards a Master Plan; Data Requirements.., ..
VOLUME III: DATA VOLUME
TAMS-ULG baro-akoejo river basin integrated development master plan1
J
"i
l_
1-
J
IVOLUME 2: REVIEW OF PREVIOUS REPORT ON RESOURCES DEVELOPMENT
1. 1.1
Introduction
l-.l
1.2.2     StratigraphyJ ~
1.2.3     Structure...*• • • 
NATURAL RESOURCES
‘ ‘ ' ' ' ' ’ '                . 
1.2.4     Minerals and Major Areas of Mineralization in Baro-Akobo Basin1-3 
1.2.5     Types of Minerals Present
1.2.6     Major Reports- More Detailed Comments ..................................................................... * 1.27     Further Data and Studies RequiredJ
1.3
Soils........................ .........................
1.3.1    Upper Basin Studies ...
1-7 
1.3.2     Lower Basin Studies
L-J2
1.3.3    Soils    of    the    Three    Representative    Plots,    at    Itang,    Pepo    and    Abobo1-1.4 
1.4
Land Use. Land Evaluation and Land Use Planning
1.4.1    Upper Basin Studies.......................................
1.4.2    Land Evaluation...........................................
1.4.3    Lower Basin Studies................................. .
1.4.4    Land Sustability........................... ................ 1.4.5    Evaluation of Data for Master Planning
1.4.6 1.4.7 Forests
Further Data Collection........................ * < - .
Further Analysis ..........................................
1.5
1.5.1    Upper Basin Studies
............
1.5.2
1.53
1.5.4
Lower Basin Studies......................................................................... Woody  Biomass Inventory  and  Strategic Planning  Project
Data Requirement.................................. +............................
1.6
Wild Life..................................
1 6.1 Upper Basin Studies
1.6.2    Lower Basin Studies .
L6.3 Protected Area Recommendations- Previous Reports
1.6.4    Data Requirement........................................................... Tourism.
1
1.7
1.7.1    Upper Basin Studies................ 1.7.2    Lower Basin Studies.
................
1.8
1.7.3     Data Requirement................................................. Fisheries
...   k   16 
...     J.-17 
...    1-20 
. . . .1-23 
...1-24 
.     ,.U26 
.  ..  J,2S
....1-29 
....     1-29 
...   .1.-30 
...     T-30 
....     U31 
...    .1-31 
....1-32 
. . . J.-32 
...     1-34 
...1t35 
.  ..  .1-36 
....     1-38 
...     .1-38 
....     1-38 
... .1-39
181 I
Status of Fisheries in the Upper Baro-Akobo Catchment (>8oo m)J .-40
1.8.2    Status of Fisheries in the Lower Baro-Akobo Catchment (<8oo m3 . . . 1.8.3     Data Requirement.. .
1.9
Apiculture................. ............... 1.9.1 Upper Basin Studies.
4
k
1.10 Mining...........................................
...Ir4l
. . .1-43 
. . M4 
.. .1-44 
. . . J.-46
TAMS-ULG      BARO-AKOBO RIVER ttASW rNTl’CrUATED DEVLt.OPMENT MASTER PLAN
iVOLUME 2 : REVIEW OF PREVIOUS REPORT ON RESOURCES DEVELOPMENT
1.10.1   Gold.............................................................................................................................. 1-46
1.10.2   Plalinum.................................................... ............................................. .............. .. 1-47
1 10.3 Iron Ore.......................................................................................................................... 1-47
1.10.4    Additional Data and Studies Required.......................................................................  Jr47
1.11      Energy Resources Survey.......................................................................................................   L-47
2. 
AGRICULTURE
2.1 
Introduction...............................................-.............. ...................... ..  ............................ ..  2-1
2.2 
Crops Production...... ............... .. ...................................................................... .. .........................2-1
2.2.1     Upper Basin ... ...................... .. ........................... .. ...................... .. ............................... 2-1
2.2.2     Lower Basin....................... .. .................................................................... .. .. ............... 2-8
2.2.3     Requirement- Upper and Lower Basin ........ ........................... .. .................................. 2-12
2.2.4     Data Collection.............................................. .. .................... .. .................................... 2? 13
2.2.5     Phase 11 Activities ..................................................................................................... 2tl3
2.3 
Livestock .. ............................... .................... ......................... .................................................. 2-14
2.3.1     Upper Basin .................................. .. .. .................. .................... .................................2-14
2.3.2     Lower Busin............... ............................. ............................. .. .. .................... ............ 2-22
2.3.3     Data Requirement- Upper and Lower Basin............................. ............... .. ................2-28
2.3.4     Phase II Activities...................... ........................... .. ........................... .. .. ............. .. .. 2-29
3.          water resources
3.1       Climatology and Hydrology. ....... .. ........... .. ............... .. .................. ............................... ..  3-1
3.1  I General................................. ........................................... .. ............. .. ................................3-1
3.1.2     Climatology.................................................................................................................... 3-1
3.1.3     Hydrology.................................. .. ............. .. .................................................... ............. 3-3
3 2 Groundwater Resources.............................. .. ........... .. .................................. .................................... 2-6
3.2.1     Introduction..................................................................................................... .. ........... .3-6
3.2.2     General Geographic Description..................... ...................... .. .................. .. ........... .. .. .3-6
3 2.3 General Hydrologic Features. ........................................... ............................. .. ........... ....  2-7
3.2.4     General Geologic Features. ...... .................. .................................... .. .................... ..  3.-8
3.2.5     Project Area Hydrogeologic Conditions............................................. .. ...................... 3-.L1
3.2.6     Use and Production of Existing Wells............................................... .. ...................... 2-15
3.2.7     Future Ground Waler Development Potential......................... .. ...........................  3r 16
3.2.8     Project Area and Ground Water Policy........................ .................................... .3-111
3.3        Flood Protection Projects.................. ............................... .. ...................... .................... .. .. .. .3-.L8
3.3.1     The Flooding Situation ..... .. ........... ........... .. .. ........... .. .................. ............................. 3-18
3.3.2     Flood Control Potential Benefit.................................................................................  3-1.9
3.3.3     Proposed Protection............................................ ,.....................................  ............ . . 3-20
3.4        Irrigation Projects....................................................................................................................  3r2O
3.4.1     Overview....................................................................................................................   3-20
3.4.2     General........................................................................................................................ 3.-20
TAMS-ULG BARU-AKORO RIVtR I! ASfN INTEGRATED DEVELOPMENT MASTTR PLAN                  11VOLUME 2 : REVIEW OF PREVIOUS REPORT ON RESOURCES DEVELOPMENT
3,4.3      Upper Basin Studies. ................................ ............................. .... .............................  3-24
3.5        Hydropower Projects,................................................... ............... .. ...........................................3-28
3.5.1     General............................................ »......................... ........................... .................... 3-28
3.5.2     Lower Basin Study......................................... ......................................................  .3-28
3.5.3     Upper Basin Study.....................................................................-................ ................3-29
3.5.4     Dam Site Review. ................................................. .. ............................. .......................3-30
3.5.5     Hydraulic Structure............................................................................................ ..  3-2L5
3.5.6     Power Installations............................... ........................................... .. ............. ..  .3.-36
3.5.7     Exisling Projects. .... .................................................................................  3t36
3.5.8     Project Costs.................. .. .................... .. .................................................................... 3-36
3.6        Water Supply Projects. ..... ........................................ .. ............................... .. .................... .. .. .3-37
3.6.1     Overview.... .......................................... .. ........................... .........................................3-37
3.6.2     Institutional and Legal .... ........................................................................................... 3t38
3.6.3     Population.............. .. ............................................... ........................... .. .................... .3-39
3.6.4     Available Resources.................................... ........................... .. ........................... ..  3r43
3.6.5     Present Water Supply and Sanitation, ........................ .. ........................... .. .......3-45
3 6.6 Present Tariffs....................................................... .................. .. .................... ..  3-56
3.6.7     Water Demand Projections. ..... ............................. .. ................................................ ..  3?56
3.6.8     Costs............................................................................................................................ 3r63
3 6.9 Water Supply Projects........... ............................. ............................................................. 3-63
4.           SOCIO-ECONOMICS
4.1 
Introduction.................................................................... .. ........................................ .. ................4-.]
4.2 
Local Government Administrative Structure........... ..............................................................    4.-2
4.2.1     Current Structure..........................................................................................................   4-2
4.2.2     Boundary Changes...................... ............. .. .. ........... .. ...................... .. ............. .. .4.-2
4.2.3     Associations..........................................................................................................        .4-3
4.2.4     Other Institutions............... ............................... .. .................... ...................................... 4-3
4.3 
Ethnics Groups......................................................................................................................     .4.-4
4.4 
Population.............. .. .......................................................................................................... .4-5
4.4.1     Census............... .. ............................................... .. ...................................... .. ........... ..  .4-5
4.4.2      Demography......................................................... ,..................................... ..................4-6
4.4.3      Population Support Capacity Studies. ........................................................................... 4r6
4.5       Health.......................................................................................................................   .4,-6
4.6        Education.................................................................................................................................    4-9
4.7          Rural Infrastructure............. ................. .. .. .................................... ............................................4.-10
4.7.1     Roads...............................................................................................................        .....4-10
4.7.2      Airports........................................................................................................................4-12
4.7.3     WalerT ransport...........................................................................................................4-12
4.7.4     Elec tn fi cation. ..................................................................................................        4-12
4.7.5     Water Supply..................... ............................................................... ........... ..  .4-13
4.7.6     Postal Services
4.8        Commerce..... .............................. .. ...............................  ... ............................... .. ...................  ,4-13
TAMS-ULG b ARU-AKono iuver bastn integrated deveeopmfnt master pi an
1VOLUME 2 ; REMEW OF PREVIOUS REPORT ON RESOURCES DEVELOPMENT
5            ENVIRONMENT
5.   i     Background and Objectives................................... ................................................. ................  -5-1
5.2        Structure and Approachof Environmental Section. .......... ............................................. 54
5.3        Environmental Perspectives of the Project     .Area.................................................................   5?2
5.3.1     Natural Vegetation.......................................................................................................  -5-2
5.3.2     Floodplain Ecology ..................................................................................................       .5-5
5.3.3     Biodiversity and Endemis............................................................................................     5i6
5.4        Current Patents of Land Use.....................................................................................................     5-6
5.4.1    Crop Production..........................................................................................................     5-7
5.4.2     Livestock.... .................................................................................................................   5-7
5.4.3     Forestry.... ...................................................................................................................     5-8
5.4.4     Fisheries .. . .................................................................................................................   5-8
5.4.5     Relocation of People.................................................................................................        .5-8
5.5        Environmental Concerns.........................................................................................................      .5-9
5.5.1     Crop Production and Soils........ ..................................................................................    .5-9
5.5.2     Livestock ...................................................................................................................   5-10
5.5.3     Forestry. .. .......................................... ............................................... .. ............... ..  5-11
5.5.4     Resettlement.................................. ................................. .. ............................. ..  5.-11
5.55      Impoudments currently under construction...............................................................   5-12
5.5.6     Projects Envisaged Under Previous Master Plan Recommendations.......................... 542
5.6        Principal Environmental Issues..............................................................................................     .544
5.6   1   Human Population Pressures ...... ............................................................................    .54.4
5.6 2     Floodplain Dynamics............... .. .............................................................................     5.-14
5 6 3     Wildlife and Forestry Issues. ........ ............................................................................   5-15
5.7        Towards a Master Plan: Data Requirements ...........................................................................   5-15
5.7.1     Policy Refinement...... ..............................................................................................    .5-16
5.7.2     Demography................................................................................................................. 546
5.7.3     Floodplain Ecology...... ................. .. ................................... .. .................... ,.............. $-17
5.7.4     Wildlife........................................................................................................................ 5-18
5.75     Land Use Change........................................................................................................  5.-18
5.7.6     Soils............................................................................................................................  5-.18
5.7.7     Institutional Structure................................................. ............. ,.................,.............. 5 19
r
TAMS-ULG baro-akoikj river basin integrated di.vi-:i.o?mfkt master plan
ivSECTION 1
NATURAL RESOURCES1. Natural Resources
1.1        Introduction
In  this section  natural resources covers geology  and  minerals,  soils,  land  use,  forests,  wildlife, fisheries,  apiculture and  mining.  Sections 2  and  3  cover other aspects ol natural resources namely agriculture and  water resources respectively.  The two  main  studies,  from which  the findings summarised  in  the following  paragraphs come,  are: (I) the ARDCO-GEOSERV study  of the upper basin  which  was completed  in  1995; and  (ii) the Selkhozpromexport study  of the lower basin  completed  in  1990,  hereafter referred  to  as the Russian  study.  To  a large extent,  therefore, the Natural Resources chapter considers the upper and  lower basins separately.  As the content of the various sub-sector reports does not conform to  a single format there are inevitably considerable differences in  the format of the reviews that follow.  In  addition,  because the survey and  analysis methods of the two  studies differ,  the sub-seclor headings for the two  parts of rhe basin  are not always the same.  Where relevant,  this review has also  considered  other studies covering  national and/or regional aspects of natural resources.  Finally,  for each  sub-sector this review identifies what additional data and  studies are required  tor master planning  and  the methodology to be used.
1.2       Geology and Minerals
The Saro-Akobo basin lies to the west of the country, west south west of Addis Ababa and 
consists primarily of Basement crystalline rocks in the eastern uplands with covering Tertiary 
lavas in places, and Quaternary Sediments in the lowlands to the west. Altitude varies from
2 300 m in the east to 400 m in the west.
1.2.1    Physiography
The  Russian  report  on  die  lower  basin  identified  four  major  land  types  from  cast  to  west.  These are;
•             the  Makonnen  Plateau  with  elevations  in  excess  of  2  000  m  to  1  300  m  generally underlain by Pre-Cambnan basement rocks with some Tertiary basalts;
•             the  Musongo  Mountains,  which  are  a  dissected  peneplain  consisting  of  mountain  chains and  intermontane  plains  ranging  in  elevation  from  1  300  m  to  600  m  and  generally underlain by Pte-Cambrian basement rocks;
•            the 1 Ligh llubabor Plain which is descending gently westwards from elevations of 500 m to 440 m and generally underlain by Miocene basalts; and
•            the low- llubabor Plain which is a flat to slightly inclined plain at 440 m to 395 m and is underlain  by  Quaternary  alluvial  and  lacustrine  deposits  and  consisting  of  river  flood plains and swamp.
The upper basin  is an  extension  of the Makonnen  Plateau  region.  The Russian  report includes a Geomorphological map  at 1:200  000  scale which  covers the whole of the lower basin.  Twelve second  order land  types arc identified  and  nine surfaces of planation.  This map  is considered  to be comprehensive. All river valleys in the plain area have Pleistocene and I lolocene surfaces eg
TAMS-ULG DARO-AKOBO RIVER. BASIN INTEGRATED DEVELOPMENT MASTER PLAN         11.           Natural Resources
Gilo. AIwcto, Baro and Akobo which drain the southwest rim of the Western E P
Highlands.
Physiographic forms can  vary  abruptly  across changes in  bedrock  type,  tor example het ccn basalts and  underlying  cry  stalline metamorphic rocks,  between  mafic and  Salic v  o  canic r , and  notably  between  consolidated  rocks and  alluvial sediments Tectonics omnia e e physiography  of the basin  but other factors are important especially  the denu  ation  cause y rivers, volcanic lava (lows and modification to relief caused by deposition.
Rock  type has influenced  topography.  Exposures of pre-Cambrian  schistsand  gneisses of the Paleozoic peneplain tend to form a smoothly undulating countryside. The granitic intrusions into these schists and  gneisses tend  to  be more resistant and  sometimes appear as inselbergs. Outpourings of basalt in  the Tertiary  generally  produce plateau  areas which  have xren subsequently dissected by rivers.
1.2.2    Stratigraphy
The stratigraphic succession in the Bare-Akobo basin as follows:
•           Quaternary - alluvial, lacustrine     deposits, basalts;
•           Pliocene - sandstone, argillites,     basalts;
•           Miocene - basalts, tuffs;
•           Oligocene - Miocene - trachytes, rhyolites, tuffs;
•           Eocene - basalt;
•           Permian - sandstones, argillites, minor conglomerates; and
•           Pre-Cambrian Basement - gneisses and schists, granites.
The Russian  report contains a geological tectonic map  at 1:750  000  and  an  engineering geological map al 1 200 000 scale. On the geological tectonic map the Gilo sandstone is classed as Eocene’Oligocenc, but Davidson (1983) assigns it to the Permian. For the lower basin these maps show the Quaternary and Tertiary formations in some detail
East of longitude 34 30‘
c
the basin is covered by die ARDCO-GEOSERV geological map sheets
at scale 1  ;250  000  There are five map  sheets compiled  from previous geological reports and maps of the area and  a limited  amount of new mapping.  The ARDCO-GEOSERV maps show forty  lithologic units many  of which  are generally  in  stratigraphic order.  The Pre-Cambrian basement rocks are separated in greal detail and fully described in the report.
In  general.  Quaternary  alluvial and  lacustrine deposits occur in  the Gambela plain  (the llubabor Plain) ie the west of the basin.  Progressively  passing  eastwards Pliocene Miocene and Oligocene-Mioccnc formations outcrop in the lower basin area between longhude 34W and
TAMS-ULG baroakobo river basin
;kADEVELOFMEyT
21.          Natural Resources
34°30'.  East of approximately  longitude 34*30'  arc Pre-Cambrian  Basement rocks consisting  of gneisses,  schists and  granites,  overlain  unconformably  by  various Tertiary  basalt flows in  the high plateaux areas.
1.2.3   Structure
The  Prc-Cambrian  cry  stalline  basement  rocks  exhibit  intense  folding  and  foliation  and  they  have been intruded by igneous rocks In the north of the upper basin, particularly, faulting is dense.
Overlying  unconformably  the Prc-Cambrian  rocks are basalt lava flows which  are generally gently  clipping  to  flat.  These lava flows are of Tertiary  age and  are very  clearly  identified  on aerial photographs.  Tertiary  and  Quaternary  sediments are found  in  the Ilubabor plain  west of Gambela. These are virtually horizontal and lie above sandstones and basalts.
Foliation  and  the axes of folding  in  the Pre-Cambrian  basement rocks arc generally  in  a norlh- south  direction.  The Pre-Cambrian  basement has undergone massive epeirogenic movements such  as permitted  the Mesozoic sea to  transgress over the region  from the south  east.  This deposited  sediments with  marked  unconformity  upon  the peneplained  prc-Cambrian  surface. Tertiary flood basalts lie unconformably over the Basement rocks in the project area.
There are two detailed cross sections in the ARDCO-GEOSF.RV geological maps showing the
underlying structure Minor discrepancies were noted between these sections and the
accompanying maps.
1.2,4 Minerals and Major Areas of Mineralization in Baro-Akobo Basin
ARDCO-GEOSERV has investigated  the mineral potential of the Baro-Akobo  basin  and  the report Volume IV C- Minerals is a comprehensive document on  the mineral resources of the basin.  The report cites fifty  three references mainly  geological reconnaissance surveys, prospecting  reports,  and  geochemical reports and  gives an  inventory  of mineral occurrences and deposits of the basin.  Data was obtained  from four regional surveys and  many  semi-detailed  and detailed surveys.
The report identifies fifty  seven  mineral deposits and  occurrences in  the Baro-Akobo  basin mainly  at reconnaissance and  preliminary  investigation  stage with  only  three deposits investigated  in  detail Thirty  one occurrences are of gold  placers,  nine are of primary  gold,  two are platinum,  one is nickel and  four are iron  ore.  marble,  bery  l,  kaolin  and  clay  account for the remaining  nine occurrences Significant occurrences that deserve further consideration  are described in detail together with associated geological conditions.
The main  areas of mineralization  are concentrated  in  the low grade metamorphic volcanic and sedimentary  rocks and  occupy  three major areas known  as the Assosa-Benishangul Mineralized Belt in  the north,  the Metti-Nejo  Mineralized  Belt in  the west and  the .Akobo  Mineralized  Bell in  the south  The first two  belts belong  to  the Western  Greenstone Region  and  the third  io  the Akobo  Greenstone Region  The mineralized  rocks arc known  os greenstones and  arc of Late- Proterozoic age. They consist of low* grade volcanic and sedimcniary rocks of green schist facies
TAMS-VLG II AkO-AKODO RIVER, BASIN' INTEGRATED DEVELOPMENT MASTER PLAN
31.          Natural Resources
along  with  associated  intrusions.  Gold  is  found  as  primary  deposits  within  the  greens  hists  and also as placer deposits.
1.2.5   Types of Minerals Present
Ihe ARDCO-GEOSERV Mineralogical Map  at 1:250  000  scale in  five sheets shows the known mineral occurrences of the basin. Indicated on the maps arc primary gold, placer gold, platinum, nickel,  basemclal,  (Cu,  Zn) chromite,  iron,  garnet,  laic,  kaolin,  marble,  beryl,  sillimanite and clay for bricks.
Gold is by far the most important metallic mineral and the total reserve of gold in various areas of the basin is estimated to be over 100 000 kg. Platinum reserves in the Yubdo ultrabasic massif are estimated to be 12 500 kg. Other significant metallic minerals arc iron ore, and base metals, indicated from geochemical analysis ol stream sediment samples
In  the Assosa-Benishangul Mineralized  Bell there are four important primary  gold  occurrences and  four important placer gold  occurrences with  three base metal occurrences.  The principle primary deposit is at Dul which is 13 km south-southeast of the border town of Kurmuk centered on  latitude 10“ 29’N,  longitude 34“24'E,  and  has been  explored  since 1970,  with  geological mapping  al I; 10  000  scale.  Exploration work is presently being carried out by a US/Canadian joint venture and concessions have been granted.
Also,  near Dul,  25  km south-southeast of Kurmuk  is the Azale-Akendevu  area,  which  is a potential deposit of copper and zinc and has been subject to considerable exploration, Indications are that there is a possible association  of gold  and  silver so  this area is considered  to  be important.
In  the Mctti-Nejo  Belt primary  and  placer gold  has been  found  al Tulu  Kapi.  Bunaya-Ankori, I .ega Baguda,  Komlo.  Yavan-Gedhamso  and  Kobera-Uwa.  Exploration  work  was earned  out some years ago in the Tulu Kapi and Bunay a-Ankori areas. However the most important mineral occurrence in this area is the eluvial and alluvial platinum at Yubdo. centered al approximately longitude 8,>57fN and  Latitude 35"30‘E,  which  was mined  from 1926  to  1988,  The highest production was in 1932 at 230.75 kg but recently production never rose above 5 kg per annum. Extensive exploration has been carried out in this area.
Also in the Metti-Nejo Bell iron orc is found which has been explored since the early 1960s. The deposits occur as lenses within the Metn-Ncjo low grade metamorphic belt. Principally four localities have been identified with total reserves estimated to be approximately 400 000 metric tons. Ollier potential important occurrences in the Metti-Nejo Bell are of nickel at Dalali, 32 km northeast of Yubdo,
In the Akobo Mineralized Belt in the south of the Basin the only mineral of economic importance is placer gold, which has been mined by local inhabitants since early times Based on exploratory work a foreign firm estimated a possible total of one million cubic meiers of auriferous alluvial gold with a possible average grade of 0.7 gm perm .
5
TAMS-GLG f AROAKOBO RIVER BASTS INTEGRATED DEVELOPMENT master pi, an
it1.          Natural Resources
Industrial rocks and  minerals are also  reported  to  occur in  the basin,  Apart from construction materials for local use and  carbonate rocks for soil conditioning,  interesting  industrial minerals occurring  are garnet,  sillimanite,  talc,  beryl and  kaolin,  but all the occurrences are believed  to  be small and further exploration is necessary,
The Gambcla basin  is associated  with  the White Nik  Rift Graben  and  Chevron  has conducted geophysical studies which  confirmed  the presence of a thick  sedimentary'  sequence beneath  the Quaternary'  alluvial sediments,  which  constituted  a deep  trough  favorable for petroleum and  gas deposits.  The area is still under exploration  since oil has been  discovered  in  the adjacent Sudanese rift basins and similar formations could continue into the Gambcla plain
from the analysis of the ARDCQ-GEOSERV report it is evident that the Buro-Akobo  basin contains significant mineral occurrences,  and  most of the prospective areas have been  covered by  regional geological mapping  and  mineral prospecting  by  several international agencies and the Ethiopian  Institute of Geological Surveys in  the past.  The data requirements for more a detailed assessment are given in section 1.10 Minerals.
1.2.6    Major Reports - More Detailed Comments
The ARDCO-GEOSERV study  of the upper basin  and  the Russian  study  of the lower basin  area provide a very  good  inventory  of geological conditions in  the Baro-Akobo  basin.  The following two  sections provide some more detailed  comments on  specific aspects of the two  studies including survey methods, report content and correlation w'ith earlier studies and mapping.
Upper Basin - .4RDCO-CiEOSERl' Report
The ARDCO-GEOSERV report covers the eastern  part of the basin,  or upper basin,  and  the accompanying  geological maps al 1250  000  cover the geology  east of longitude 34°30.  These geological maps,  produced  on  five sheets,  show,  solid  geology  only.  The maps are compilation maps with  some limited  field  work  in  areas not previously  mapped.  The legend  of the maps is quite difficult to  read  due to  the use of rather similar symbols and  a limited  range of colours. Clarity  would  have been  improved  if letters as well as symbols were used  to  designate the geological units.  Also  units are only  grouped  approximately  in  stratigraphic sequence and  no attempt has been made to determine the precise stratigraphic order.
Unmapped  areas were covered  using  aerial photo  interpretation  and  limited  field  work  1950- I960  age aerial photographs at 1:60  000  scale were used  and  photo  geological units were plotted on  overlays.  Information  was transferred  directly  from the overlays to  topographic base maps at 1:50  000  scale.  The photo  interpretation  was completed  by  three geologists and  took  one month,  after which  the three geologists spent two  months in  the field  mapping  field  traverses located  near existing  paths and  roads No  photo  interpretation  was done on  previously  mapped areas.
One of the map  sheets,  Gore sheet,  was compared  with  the published  geological map  of the Ethiopian  Institute of Geology  Survey  (NC 36-16) of this area and  was found  to  agree quite well with  some slight modifications.  An  area of granite on  the published  map  was shown  as gneisses on  the /XRDCO-GEOSERV map.  Also  1.60  000  aerial photographs,  available in  the project
T AMS- U L G DARO-AKcmo river das rN integrated development master pi a n1,          Natural Resources
office,  covering  a strip  E-W of this sheet,  were examined  slcreoscopically  and  generally  major geological boundaries,  for example between  basalt and  metamorphic rock,  were found  to  agree very well with the map units. The Maji sheet was compared with the 1:500 000 geological map accompanying  Bulletin  No  2  of the Ministry  of Mines and  was generally  found  to  agree very well. One unit of Permian sandstone on the Omo river sheet was wrongly designated granite and syenite.  A further unpublished  map  sheet of the Ethiopian  Institute of Geological Survey  (sheet NC 36-7 and 8) Kurmuk-Asosa was coloured up in the office and compared with the ARDCO-
t
GEOSERV Assosa sheet and  found  to  agree quite well.  It is considered  that the geological compilation  maps accompanying  the ARDCO-GEOSERV report are a good  indication  of geological conditions,  in  the upper basin  east of longitude 34*30  The maps arc produced  in printed format but the legend is often difficult to read due to the fact that symbols are confusing,
lhe ARDCO-GEOSERV report on geology describes the geological units in the upper basin very' comprehensively,  although  some difficulties were experienced  in  correlating  geological units from different sources Overall the report is considered  to  be a good  measure of the existing geological information  for this part of the basin.  Previous geological studies arc described  in detail and lhe list of 59 references is comprehensive of which a selection were examined.
The map units are grouped under five major rock unit categories and all are described in detail:
•     gneisses and migmatite;
•     meta-sedimentary, meta-volcanic and metamorphosed shallow intrusive rocks;
•     meta-plutomc and plutonic rocks;
•     I ate-Paleo zoic to unmeta-morphosed Tertiary sedimentary and volcanic rocks; and
•    Quaternary deposits.
Lower Basin - the Russian Report
The Russian report is based on ten years of field work in the lower basin. Annex 2 of Volume IV.  which  includes geomorphology  and  geology  used  aerial photo  interpretation  extensively  as lhe basis for producing lhe maps. Il appears that very little wort had been done in the lower basin before the Russian report apart from lhe Development of the Gambcla Plain study by TAMS in 
1976.  Previous  geological  studies  were  mainly  concentrated  in  the  upper  basin.  In  general  the Russian  report  is  not  easy  to  read  but  represents  a  considerable  amount  of  work.  The  report described   the   geomorphological   structure   in   some   detail   and   identified   nine   surfaces   of planation.  These  are  shown  in  colour  on  the  accompanying  geomorphological  maps  at  scale
1:200  000.  The  four  land  types  described  and  shown  on  the  geomorphologicaj  maps  were reviewed using the 1:250 000 topographic maps and appear to tie in well with reality Generally, in the Baro-Akobo basin from east io west there is a dissected peneplain at 1 500 Io 2 400 m, a sloping dissected Piedmont area al 1 OCKJ to I 500 m and a gently sloping alluvial lacustrine plain at 400 to 500 m. These descriptions appear to correlate reasonably well with those in the Russian report.
TAMS-ULG b aroakobo river basts integrated development master plan
6L          Nat urn I Resources
The section  on  geology  in  the Russian  report has identified  in  stratigraphic order: recent terrestrial deposits,,  Quaternary  volcanic rocks.  Tertiary  pre-rift volcanic rocks.  Mesozoic sedimentary  rocks and  pre-Cambrian  basement rocks.  The report describes the geological units in some detail but in no specific logical order making it difficult to read.
Four sheets at 1:200  000  comprise the engineering  geological map  which  distinguishes seven ty  pes of cover deposits and  five types of underlying  rocks.  Many  boreholes were sunk  in  the plain  for hydrogeological purposes and  four cross sections f IL  HI and  IV were plotted  through these boreholes.  The boreholes reach  down  to  275  m and  pass through  sedimentary  rocks into granites and  basalts.  The sections were reputedly  drawn  at a horizontal scale 1:200  000  and  a vertical scale 1:1  000  which  is a tremendous exaggeration.  In  fact the sections available have been  photo  reduced  making  detail difficult to  read.  The sections generally  start in  the Masongo mountains in  the east,  but boreholes were only  sunk  in  the high  and  low Illubabor plain. Geophysics was also used for these more eastern sections. West o f Gam be! a boreholes penetrate sandstones and  siltstones.  Two  further sections V and  VI are also  available but are not plotted on  the engineering  geological maps.  All the sections in  the plain  area penetrate a variety  of superficial deposits before reaching  the sandstone and  siltstone.  Al the boundary  between  the plain and the mountains sediments are faulted against basalts granites and gneisses.
A   map   at   scale   1.200   000   on   four   sheets   shows   the   areas   of   comprehensive   study,   and engineering    geological    and    hydrogeological    surveys    of    reservoirs    and    sites    for    hydraulic complexes.   Il   shows   four   locations   of   dams   with   many   boreholes   (used   for   hydrogeological purposes)  across  the  plain.  Reservoir  basins  were  investigated  at  1.50  000  scale  and  dam  sites at  1:5  000  scale  with  the  production  of  topographic  maps.  The  reservoir  basins  shown  on  the map arc hang, Bonga. Chira, and Gilo 2/1 and Abobo. south of Gambcla.
L2.7 Further Oats and Studies Required
There  is  a  need  for  a  map  of  the  whole  basin  on  an  Al  sheet  showing  the  geology  from  both studies at an appropriate scale.
13 Soils
This section contains a review' of the soils studies, sampling, mapping and respective reports for both the upper and lower Baro-Akobo basin (ic the ARDCO-GEOSERV and Russian studies). Current constraints for agnculture are noted and soil characteristics of both areas and representative plots are discussed.
1.3.1    Upper Busin Soil Studies - ARDCO’GEOSERV
Soil Survey Method
In brief the soils study were earned out in four stages.
• Review of existing data, including 1:250 000 satellite imagery and the 1:50 OOD topographic cover.
TAMS-Il LG ElARO-AKOBO JUVER, EM SIN INTEORATtD         DEVELOPMENT MASTER PI .AN
71.          Natural Resources
2
•      Selection  of  field  study  areas  and  the  1:60  000  air  photo  cover,  comprising  seven strips covering some 2%; 1 0-10 km , of the upper basin.
•      Field  study  of  the  selected  areas  included  location  and  profile  examination  of  461 auger bores and 212 pits which were sampled for chemical analysis.
•       Post  field  compilation,  re-interpretation  of  the  satellite  imagery  and  plotting  the boundaries  of  landscape  soil  units  based  on  FAOUNESCO  (1988)  legend,  resulting in digitised colour mapping al I ;250 000 (five sheets).
Comments on Methods Used
The ARDCO-GEOSERV lerms of reference called  for both  reconnaissance soil mapping  at 1:250 000 and semi-detail cd soil mapping al 1:50 000 for "selected development areas”. Actual field studies resulted in 673 profile observations of which 212 sites were 2 m deep pits, thus the overall density  of observations was only  one per 77  km2 However,  as the approximate area covered  by  the seven  soil study  strips was 942  km2 the actual density  of observation  was
1:1.40  tnr and  if the other visual observations,  (ie road  cuttings),  are included  in  the total the observation ration is close to I: J.O km2, The field mapping at 1:50 000?, was not available for review.
Sort Mapping-Landscape Units
The 1:250  000  digitized  colour maps legend  shows 30  units.  These units,  which  arc in  five groups, are related to the 30 main and associated soils and are described in Volume ULB Table 3  B .3  of the ARDCO-GEOSERV report.  This table shows the area of each  unit together with lheir respective physical and  chemical characteristics.  Table 1.3  J summarises the details of the 30 units in terms of area, % area, slope, drainage and dominant soil type. The area of each group is given below.
TAMS*ULG h aroakobo river basin integrated development master PLAN1,          Natural Resources
Relict Form
No. of Units
km1
% of Area
Mo - Mountain
6
8 522
16.4
Hi - Hilland
5
ID 858
208
Pl - Piedmont
6
5 792
11.1
Pu - Plateau
8
17 372
33.3
Pl - Plain
5
9 586
18.4
Total                30
52 130
100
The areas in brackets arc taken from the draft reports as (hey are missing from the final version
Slope of Land
Owing to some errors and omissions the total area described is 6% less than the total of the study area.  52  500  km*  The table below ,  indicates the distributions of the above JO units by  slope grouping as against the Relief Unit Areas.
Slope %
km1
%
% of Relief Area
0-8
19 446
42
Pl - 18
9-16
13 138
29
Pu-33
17-30
6 750
15
Pi - 11
31-50
4 226
9
Hi - 21
>50
2212
5
Mo - 16
Note; % may not total to 100 due to rounding
From the above it is clear that slope is a major constraint to development in the upper basin and that the preparation  of a slope map  which  also  indicates the non-arable zones ie bare rock  and screes is a priority.
Soil Cover Characteristics
Table 3.B.3  in  Volume 111.  B gives a summary  description  of the land  and  soil characteristics while Table 3  B.4  and  Table 3  .0,5  describe the soil families and  chemical characteristics of the soil families respectively.  The information  given  is far too  limited  for anything  other than  a first approximation  of the soils in  the upper basin.  In  an  attempt to  make an  estimate of the area of each  major soil unit,  the data in  Table 3.B.3  has been  "unpicked" and  lhe detailed  results are shown below.
TAMS-ULG UARO-AKOGO RIVER IJASFN integrated development master plan
0L Natural Resources
Deep Profiles well drained
Deep Profiles slow/poor drainage
Sub-total
Rocky, sleep, stony Balance
Total
Noils anti Irrigation
33 398 2 873
36 271
12 703 3 169
52 143
64.90
5.49
7039
24.35 5.26
100.00
For irrigation soil salinity and or alkalinity do not appear to be limiting factors, as for example in the Awash valley. The lavas in the upper basin are far less ''alkaline'', with the exception of a few examples with  high  sodium levels The information  provided  in the report* Volume H E Irrigation, to support the selection of the 18 sites would appear to be based upon interpretation of air photos etc. rather than routine soil surveys.
S<n7 Cfiemicfl/ Xnafyrfr
l"hc methods used and the results obtained appear to be satisfactory and compliment the limited number of previous observ ations taken over a large area.
Soil Constrato/j to Agricultural Development
In reviewing the report and with experience of the area the constraints to the cultivation of traditional rainfed crops are:
•     slopes and the need for soil conserv ation works;
•     waterlogging and the need for field drainage;
•     shallow soils and coarse textures;
•     acid soils developed on certain rock types;
•    low levels of N and K and low levels of P and/or fixation of P;
•    lack of humus and organic matter and thus poor tilth;
•     shortage of soil moisture is a nsk if rainfall is poorly distributed;
T A M S- D EG B A RO-AKOBO R fXTR B ASIN INTEGRATED DEVFLCH'M ENT MASTER PLAN
JO1-         Natural Resources
• some fine textured soils have a limited period for cultivation using traditional
methods;
•     present practices are "mining the soil: and
•     population increase is bringing poorer sites/soils into cultivation which speeds tlic
rate of land degradation
Similarly, for irrigated agriculture the constraints are;
•     slope, small plots, frequent hunds-low water efficiency;
•     drainage of bottom lands;
•     relatively small areas of uniform soils;
•     soils with low fertility-thus poor returns;
•     some risk of salinity in few areas/wntcr supplies; and
•     low winter soil temperature.
Finally, for plantation crops and (and fuel/browse spp.) the constraints are:
•     suitable plots for coffee restricted by soil, exposure and altitude (also water for
pulping); and
•      limited  sites  for  tea  and  spices  due  to  climate,  altitude  as  well  as  soil  type  and limited  sites  for  citrus  due  to  soils  and  altitude,  although  ample  lands  for  fuel  and browse crops
Further Studies Required
The review of the ARDCO-GEOSERV report has indicated  that the available soils information is far too  limited  for anything  other than  a first approximation  of the soils in  the upper basin. There is a need  for soil mapping  at 1:50  000  scale of representative areas of the Agro-Ecological zones,  at least 200  km2 per zone,  or 400  km3 in  the larger zones (750  km: is approximately 1:50  000  sheet) to  guide the formulation  and  implementation  of detailed  development proposals
during   the   implementation   of   the   master   plan   Density   of   observation   should   be   at   least   1 
observation  per  knr  of  map  over  "arable"  areas  and  follow  the  FAO  guidelines  for  soil  profile description. Some 10% of the sites should be soil pits and sampled for chemical analysis.
Before preparing the detailed proposals and designs far the proposed 20 to 30 small irrigation schemes each scheme area should be surveyed at detailed level, 1 observation per 0.1 knr and mapped at 1:10 000 or 1:20 000 scale. Soil physical testing should be carried out of representative sites of the main soil land classes towards the end of the field study. Criteria for
r AMS-V L G U A Hf 1-AKQRO nt VER H ASIN rNTTGILATHD DE V ELOPM t Nt M ASTTFl PLAN
It1.          Nafural Resources
soil  suitability  should  follow  USBR  manual  Volume  V  part  II.  These  studies  would  help determine the most appropriate cropping, patterns and irrigation schedules.
As noted  above slope is a major constraint to  agricultural development in  the upper basin  and the preparation of a slope map is a priority using G1S to "unpick" the slope details from die soil landscape units There is also a need to map the soils by G1S (ie dominant soil type(s) and slope). Finally, there is a need to refine soil-slope-landscape-in those areas adjacent to the areas mapped at 1:50  000  (representatives of the Agro-Ecological zones) and  at 1:10  000  or 1:20  000  (for irrigation development) following field studies.
1.3.2     Low er Basin Studies
Volume V of the Russian report contains soils data together with mapping al 1.200 000 scale (4 sheets).
Soil Survey Methods
The study combined original survey work with earlier studies and mapping undertaken in 1986- 89. Difficult access made it impossible to carry' out all the planned field work which resulted in a reduction  in  the number of site observations and  greater emphasis on  interpretation  of air photography, h is not clear from the text how many sites were examined and described. Figure A. I, Volume V Annex 4, shows the location of the soil pits analysed, 126 in number.
The survey covered an area of 24 500 km  within which a part was surveyed by traverse on the
:
ground and a larger portion that wus examined using helicopters and limited ground work.
?V number of samples were collected from bath areas for a wide range of chemical and physical analyses. Also 11 trial sites within the smaller area covered by the ground survey were selected for in-situ physical tests. The number of chemical and physical tests were as follows - soil pits were 15 m to 2.5 m deep
Type of An
iilysis
Sites
Chemical
analysis
Mechanical
163
157
Bulk  density-water  holding % Soil Aggregate
Soil Mapping of ihe Lower Basin
22
18
The soil map at 1:200 000 (4 sheets) printed in colour 1990 is based upon interpretation of the 1985  air photo  cover and  the 1:50,000  mapping  together with  limited  ground  information  and 
overflights by helicopter
] he resulting map sheets are a satisfactory indication of the distribution and areas of the major soil units described in the legend It is also a suitable starting point for the estimation of both
T \MS-l LG BARO-AKOBO kl VER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
111. Natural Resources
irrigated  and  rainfed  amble  potential  in  conjunction  with  the  soil  Suitability  and  Reclamation map at the same scale.
While a lot of the investigative work  was directed  towards the suitability  for surface (flood/fuiTOw) irrigation,  the general comments on  the use of the land  for rainfed  cropping  are significant.
Analysis Results and Physical Tests
Chemical analysis results arc well reported  and  very  close to  the norms used  by  FAO etc. However the mechanical analysis results differ in  that the particle size separations are different to  those used  by  FAO and  others This results in  the "clay" fraction  including  a proportion  of the silt fractions smaller than  0.001  nun  diameter.  Also  the very  coarse sand  fraction,  that fraction between 1.0 and 2.0 mm is excluded from the analysis.
Physical tests differ from those used  by  FAO,  USBR and  11.RI.  but give a comparable indication of the infiltration  rate at the soil surface and  of the approximate uptake of water by  the profile. Soil stability when wetted was also measured.
Soil Type Distribution, Landform and Slope
Neither the map  legend  nor the text on  soils give the area covered  by  each  soil type or sub-type, Areas of major soil units are given  only  (see below).  G1S scanning  will be required  to  obtain  and tabulate (he area of each soil type and sub-type.
A cross section of the area, line ni-TII, is from the east of Gambcla to the Haro river near Jikawo. 
The land form and soil mantle are closely related to the underlying formations as is the slope of
the soil surface. Vertical scale is much exaggerated.
Summary of Soil Mapping and Soil Units
Thirty  two  soil units were recognised  under the Soviet and  International classification,  which  was comparable to  the previous soil studies earned  out in  (1970  and  1976).  Broad  soil grouping  by major tjpes given in the text is tabulated overleaf.
TAMS-ULG BARQ-AKouo river basin integrated development master pi an
it1.          Natural! Resources
Soil Type
Area km1
% of Area
Red laterised (nitisols)
500
2
Ferraltilic (ferrasols)
834
3
ferruginous
Ferraltilics (chromosols)
I 865
8
Brown sia'.lilic (cambisols)
5 445
23
Bleached meadow (planisols)
331
1
Dark meadow (humic planisols)
870
4
Black compact (vertisols)
5 372
23
Alluvial (fluvisols)
5 246
22
Alluvial meadow-bog (gleysols)
2 777
12
Immature soils (lithosols)
875
2
Totals
24 115 100
Soil cover is closely related to the local land forms and Table 1.3.2 gives details of the
soil type, land form and other major profile characteristics.
133 Soils of the Three Representative Plots* at Itang, Pepo and Ababo
T/ie Itcmg plot, 660 ha, is 6 km NbFW of lung on the Jikawo road. It is representative of a large area io  the north of the Bare River on die low plain. The area is flat, about 427 masl, in the grassland-savanna zone. Parts of the area are seasonally flooded and grazed, and a part is used for slash and bum cropping. The soils of this plot, formed on alluvial lacustrine deposits of recent age, are:
%
Vertisols
54.0
Cambisols
29.0
Meadow soils
1.4
Complex soil
15.6
Ground water levels are 20*30 m bclow  ground level, but perched water
r
tables may form in the
wet season or with irrigation.
Top soils are non-salinc and non-alkaline but some subsoils have alkaline features, pH >7.2, as indicated by the values for Na++ and this implies a risk if water tables nse in future. Nutrient levels vary by soil type but generally all are low in N and P and some arc low in K also, organic matter is generally low
All soils are suited  to  rainfed  cultivation  but will require fertilizers and  organic mailer to maintain or improve crop yields. Some areas have lesser or greater flood problems according to topography.  Hie vertisols present a tillage problem and  are difficult to  manage both  under rainfall alone and if irrigated by flood or furrow
T V MS -1’ L G » \R< 1- AKOBO FUVER BASIN INTEGRATED DFVFWPWENT MAST1 R PI. AN
14L          Natural Resources
The Pepo  plot covers 480  ha and  lies 1  km south  of I Lang  on  left bank  of the Baro  river Euroconsult surveyed  part of this plot in  1976  and  their findings arc assumed  to  be included  in the Russian report.
The location  of this plot is representative of the slightly  elevated  eastern  part of the flat low illuvial plain  at some 420  mask  Seasonal flooding  occurs over parts of the plot,  to  a max-depth of 2  m.  This results in  annual deposition  of soil forming  material.  Soil profile drainage is slow
□nd gloving is present in the subsoils, The soil cover of plot consists of three soil units.
ha%
fluvisols         2110              44           Laminated  and  uniform  deposits  of  sills  and  clays
gleysols
29.5 
6
vertisols
238.0 
50
Meadowbog depressions
Depression and hollows-clay accumulation sites,
The report states that this area is characteristic of the Haro  river basin  soils and  their distribution. The vertisols have deep  fine textured  profiles with  characteristic "gilai" surface and  are often seasonally  flooded.  Non-saline; non  alkaline with  high  nutrient and  organic matter level.  Some deeper profiles have low sodium values at depth  The vertisols require,  protection  from flood, tillage and  subsoiling  and  field  drainage for rainfed  cropping.  Irrigation  by  means of flood  or furrow'may be difficult. Fertilizer and possibly gypsum will be required to maintain crop yields.
The fluvisols are generally deep, fine-medium texture, with an increase in clay with depth. 
Neutral at the surface and increasing pH with depth, sodium is low while calcium is high. 
Nutrient levels are good while organic matter is low. The fluvisols are less subject to deep 
flooding and the soil fabric is easier to till and has a high level of nutrients but low in organic
matter. Most soils are neutral al the surface becoming alkaline with depth. pH 7.6. Ground waler
levels are 1-2 m in the wet season, falling lo 3-5 m depth al the end of the dry season. Rainfed 
cultivation may require some flood protection and field drains. Fertilizer will be needed to 
maintain yields in future Irrigation could be a risk due to relatively high ground water levels and 
incipient alkalinity.
The meadow bog  soils (gleysols) are confined  to  small internal drainage basins subject to prolonged  flooding  and  have a fine texture with  abundant organic matter in  (he upper profile. Nutrient levels are high,  sodium is present in  the subsoil.  Soils are acid  at the surface neutral- alkaline at depth.  The latter soil is best left undisturbed  and  used  for grazing,  but could  be irrigated.
The Abobo plot is located 35 km south of Gambeta on the Abobo road and covers 2 137 ha of
the flat high plain between 490-46? mask Soil cover has developed over Pleistocene illuvial and alluvial deposits and consists of Camb iso Is 40%, Chromoso Is 58% and Humic Planosols-
meadow soils 2%. The later arc confined to low lying hollows and channels. In 1989-90 the area
was cultivated, cotton having been grown for some years together with small plots of sorghum. 
No fertilizer was used and cultivation followed traditional practices.
The Chromosol soils are deep,  >lm.  clays with  clay  accumulation  in  the B horizon,  neutral to acid  with  depth.  Topsoil organic matter is high,  low below.  Somewhat low- in  nitrogen  with 
TAMS-ULG DARO-AKUBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
l<L          Natural Resources
moderate levels of P and  K and  non-salinc non-alkalinc with  ground  water levels + 20  m below surface.  The Ground  water is slightly  alkaline-alkaline in  places.  Some 6% ot the area have chromosois with  shallow profiles,  40-100  cm deep.  The soils have a good  structure and  arc well suited  to  cultivation  but require fertilizer to  maintain  or increase yields Some erosion  control
is also desirable.
The Cambisols are underlain  by  lateritic concretionary'  layers and/or weathering  rock  Ground water tables are deep +20 m. Soil structure is good and textures are fine to medium-fine with clay accumulation  in  the B horizon.  Topsoils are neutral,  the profile becoming  slightly  acid  with depth  pH 5.5.  Fertilizer will he required  to  maintain  or improve yields and  some field  drainage may be necessary. Erosion control is desirable.
The Planosols are fine textured.  compacted  in  the subsoil and  have a high  organic matter content on  the topsoil and  high  nutrient levels.  The soil is neutral becoming  alkaline in  the deeper subsoil.  Water holding  is good  but soils are difficult to  work  in  the dry.  Ground  water; depth varies from L5 m during the rains to 3-5 m in die dry season, is probably slightly alkaline.
Further Studies Required
As noted  above the Russian  report and  maps provide a satisfactory  indication  of the distribution and  areas of the major soil units The map  also  provides a suitable starling  point for the estimation  of both  irrigated  and  rainfed  arable potential in  conjunction  with  the soi l Suitability and  Reclamation  map  at the same scale.  For master planning  the following  additional information, mapping and studies arc required:
•     digitize the existing soil map sheets;
•     produce a tabulation of the areas of soils from the GIS map:
•     produce a small scale soil map by GIS and compare it with the FAO map;
•     produce a slope map by GIS and tabulate the slope classes by area and slope - soil
groupings; and
• obtain all soil profile data and assemble for each site the relevant analyses, data etc.
1.4 Land Use* Land Evaluation and Land Use Planning
This section contains a review* of the land use. land evaluation and land use planning studies and mapping  contained  in  the study  reports for both  the upper and  lower Baren  A kobo  basin  and  in several other reports.
FAO has defined  agro-ecological zones on  the basis of a combination  of soils,  landform,  growing periods and  thermal zones.  The FAO study  identified  140  agro^cological zones in  the country and  the Baro  -Akobo  basin  lies within  three regions namely  the Western  Lowlands.  Western Transition and Western Highlands which have 11 agro-eco logical zones
TAMS-ULG baro-akoboRIVER basin integrated development raster plan
l«1.          Natural Resources
the FAO study is valuable as it is a uniform system of zonation applied to the whole of the study area Further,  the report has identified  the constraints and  potentials of each  of the zones which the ARDCO-GEOSERV report docs not do.  A disadvantage is that the map  scale is 1:2  000  000 which is only reconnaissance level hut at the level of master planning this is considered adequate.
1.4.1 lipper Basin Studies
Agro-ecological Zones
The ARDC'O-GEOSERV study  has adopted  the agro-ecological zoning  approach  put forward by  die Land  Use Planning  and  Regulatory  Department.  In  this system,  eleven  agro-ecological zones have been  identified.  The ARDCO-GEOSERV Figure 3.A.2  shows the location  of these agro-ecological zones.  The major drawback  to  the agro-ecological zones used  by  ARDCO- GEOSERV is that no  attempt was made to  identify  the constraints and  potentials for each  unit. The agro-ecological zones,  at least as used  by  ARDCO-GEOSERV.  are entirely  descriptive and for this reason  are of limited  value.  The original Land  Use Planning  and  Regulatory  Department (1989) report should be obtained along with the maps and analysed during Phase 11.
Source of Current Land Use Maps
The land  use/land  cover maps have been  prepared  from the mapping  of the Woody  Biomass Inventory  and  Strategic Planning  Project.  The maps produced  by  the Woody  Biomass Inventory and  Strategic Planning  Project were used  as base maps for field  verification  with  the final land use mapping units being delimited using 1 250 000 scale TM landsat images.
Land Use (Agriculture, Forestry and Livestock)
ARDCO-GEOSREV identified  84  land  use/land  cover units and  mapped  them at ] :250  000  scale. The units are described  according  to  agro-ecological zone units.  The mapping  units identified, and  agreed  with  the Woody  Biomass Inventory  and  Strategic Planning  Project.  The mapping symbols are formed by adding codes eg.
CjCm FBj WO/H, (Sj-Gj)
This stands for moderate cultivation  plus perennial coffee; very  disturbed  broadleaf woodland, open  woodland; mountainous with  slopes of 16-30% and  moderate sheet and  gully  erosion.  The land  use/land  cover legend  does not include the area of each  mapping  unit in  the Upper Basin  or by  administrative unit although  this is very  easy  to  generate from the GIS.  A map  of the cultivated  land  in  the basin  at 1:1  000  000  scale has been  produced  showing  the following categories;
1.          >60% cultivated
2.          40-60% cultivated
3.          10*40% cultivated
4.          <10%
5.          Coffee
6.          State farm
TAMS-ULG BARO-AKOflO RIVER HASIN integrated development master plan1.           Natural Resources
7.           No cultivation
The Forest Resources Survey (Volume 111. E) estimates that there arc 24 000 km3 of forest in the former regions of Illubabor.  Kefa and  Wclcga and  that 60-70% or between  14  400  km: and 
16  800  km*  is  found  in  the  upper  basin.  The  report  also  estimates  that  forest  and  woodland combined  covers  some  43%  of  the  upper  basin.  A  further  6  000  km7   is  mixed  agriculture  and forest land.
Further,  the State Forest Conservation  and  Development Department has designated  11  National Forest Priority Areas in the upper basin covering some 15 000 km: of which over half is heavily disturbed, nearly 20% is open land and nearly 30% slightly disturbed.
.As part of the land useland cover mapping, six forest categories were distinguished and mapped from aerial photographs (1986) and Landsal images (1989). The table below shows the categories and areas.
Forest Type
Area (km )
1
Dense High Forest
9418
Disturbed high forest
5S7!
Dense woodland
10 262
Open woodland
6 844
Highland bamboo
1 590
Lowland bamboo
4 296
Total
38 281
The following sets of maps have also been produced:
Forest Status Forest Cover Agro forestry Forestry
1:1 000 000 
kl DOO 000 
1:1 000 000
I 250 000
The forestry map at 1:250 000 scale gives a good indication of the land usc-cover and the legend is detailed  but the areas of each  unit arc not given  The forest cover and  status maps arc only useful for an  overview ol the upper basin.  The agroforestry  map  legend  does not seem to  be explained anywhere in the report and is of limited value.
The  land  use/cover  map  docs  not  identify  areas  where  livestock  rearing  is  common,  nor  does  it identify areas suitable for pasture range land. The livestock report (Volume III F) makes
T \MS-UL.G BARD-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
IIL          Natural Resources
reference  to  Asosa  woreda  as  having  large  tracts  of  savannah  grassland  suitable  lor  extensive livestock production.
Discussion of Land Use/Land Cover
ARDCO-GEOSERV describe the major mapping  units in  terms of agricultural use,  vegetat ion, altitude,  rainfall and  length  of growing  period.  The areas of major land  use are not given  (ie arable land,  forest,  perennial crops such  as coffee,  tallow or urban).  Hie report gives a static picture of land  use.  There is no  attempt to  determine the changes in  land  use that have occurred 
□ver  time  or  to  say  why  they  have  occurred.  Neither  is  there  an  attempt  to  identify  areas  of  land degradation  In  addition,  the  information  which  is  based  on  1986-89  data  is  almost  certainly  out of  date.  Limited  field  visits  indicate  that  there  has  been  significant  clearance  of  forests  since  the second half of the 1980s (see section 1,5),
ARDCO-GEOSERV has not related  land  use to  present population  densities and  no  attempt is made to  estimate the carry  ing  capacity  of the land  so  that the consequences of population increase can  be predicted.  This can  be done in  a simple way  by  using  the population  support capacities as calculated  by  FAO as part of the Master Land  Use Plan  (UNDP/FAO 1988).  The model has been  updated  and  is available from the Woody  Biomass Inventory  and  Strategic Planning  Project.  The agricultural systems have not been  identified  in  detail.  There is no discussion  of production  sy  stems such  as shifting  cultivation  or mixed  arable livestock  systems. In  effect,  the land  use cover exercise has produced  maps showing  generalised  land  use without any  analy  sis of the underlying  reasons for their adoption  or the temporal or spatial changes that are occurring.
There is a clear need to collect further data on land use and MOA has data on the areas of various land uses by woreda without which no rntional land use planning is possible (see section 1.4.4)
Land Classification
ARDCO-GEOSERV classified  land  according  to  the Soil Conservation  Requirement Classification  (Land  Classification  for Use in  Soil Conservation,  FAO 1988).  According  to  this classification,  the soil mapping  units are assigned  to  one of eight classes in  order of increasing erosion  risk  although  other factors such  as soil depth  and  stonincss are taken  into  account.  The Soil Conservation  Requirement classes arc defined  on  the basis of slope,  soil depth  and  stoniness. Within  these eight classes the major limitation  is identified  as a sub-class or land  class unit Table
13.3   gives   the   results   of   the   Soil   Conservation   Requirement   classification.   The   Study   has. following  the  Soil  Conservation  Requirement  classification  of  FAO.  taken  the  upper  limit  to arable land  as class  IV.  Land  in  class VI  is  suitable  for perennial crops and  grazing  class  VII for forestry   and   land   in   class   VI11   as   suitable   for   wildlife   and   catchment   protection.   This   is summarised below.
TAIMS- U LG narO’a xou o fli ve h liastn i ntegiiated nr v el opkfENT m a ster pi _anL          Natural Resources
Major Land li.se
SCR Ctm
Area (km:)
Annual crops
l-IV
36 000
Perennial crops and grazing
VI
9 500
Forestry
VII
2 650
Wildlife & catchment protection
VIII
3 790
Total
51 940
The use of the soil conservation  classification  to  arrive at a suitability  classification  of land  for uses other than  arable agriculture is incorrect.  To  allocate land  to  forestry.  pasture or wildlife simply  on  the basis that it is unsuitable for crop  production  cannot be defended.  Land  suitable for arable crops is equally  suitable for the other uses and  similarly  land  unsuitable for crops is also less well suited to the other uses. The evaluation exercise should follow the FAO Framework for Land Evaluation and the allocation of land to competing uses based on wider considerations than soils and landform (see below).
1.4.2    Land Evaluation
‘Ehc land evaluation of the upper basin has been undertaken according to the Framework for Land Evaluation (FAO. 1983). A series of 31 land units have been identified on the basis of similarities
of soil, land and agro-climatology.
ARDCO-GEOSERV matched  the environmental requirements of 14  crops with  the land  units and  a suitability  rating  made.  The Report identified  three classes of land  suitability.  These are SI.  very  suitable (>80% of potential maximum yield),  S2-3,  moderately  to  marginally  suitable (20-80% of potential maximum yield) and N, not suitable (<20% of potential maximum yield).
Land can only be evaluated with respect to a defined use or land utilisation type. Although the Report does not define the land  utilisation  type,  it is implicit that existing,  smallholder rainfed agriculture is the land  utilisation  type.  That is,  the potential maximum yield  is that achieved through low input, subsistence agriculture. It is also unclear if the land utilisation type assumes the provision  of soil conservation  measures to  protect the land  The 14  crops in  the evaluation are the major crops grown  in  the upper basin  namely: leff.  grain  legumes (4),  sugar cane, bananas,  ground  nut,  sesame,  enset,  sorghum,  maize,  coffee and  tea,  but there is no  evaluation of the different smallholder crop  rotations or fanning  systems.  The land  characteristics used  in the evaluation are summarised below
TA.VfS-ULG BARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLANI Nalural Resource*
Land Quality
Land Charnel eristics
Temperature
Mean temperature of grow ing period, 
altitude
Growing period
Number of days when rainfall exceeds PET
Moisture availability
Mean annual rainfall
Drainage
Soil drainage
Nutrient status
Soil texture, total N, available P. exchangeable K. Base Saturation. CEC
Rooting condition
pH. soil reaction
Land surface
slope, stoniness
Annex  3.C.2  in  Volume III C of the ARDCO-GEOSERV report gives the crop  environmental requirements for the 14 crops and Annex 3.C.3 a summary of the suitability of each kind unit for each crop giving lhe major limitations
A series of maps at 1:250  000  scale have been  produced  showing  the suitability  classes for the 14  crops/crop  groups.  A series of tables summarising  the suitability  of each  land  unit for the different crops is given  which  also  identify  lhe major limitations to  cultivation  The area of each suitability class for each crop has not been tabulated.
The evaluation hits been carried out in a systematic fashion. The definition of land units, the land qualities used  and  the crop  environmental requirements are sound.  However,  a major limitation of the analysis is that it has evaluated  the land  only  in  terms of smallholder crop  production. There is no  systematic evaluation  of lhe land  for forestry  (fuclwood,  timber or soil and  water conservation),  livestock  (pasture systems,  rangeland),  wild  life (including  recreation  and tourism).  Without an  evaluation  of the land  for alternative uses and  maps showing  the distribution of such land, no land use plan can be made.
Land Capability for Irrigation
Two approaches were used to identify land suitable for irrigation, the first relied on simply identifying the land uniLs suitable for irrigation (land suitability) and the second on a combination 
ot kind suitability and the availability of water. ARDCO-GEOSERV identified 12 land mapping 
units as suitable for irrigation and these are shown in Table 1.3.4
This gives the gross area of suitable land  and  the possible area after allowing  for an  amount of unsuitable land in some of the units. The area of possibly suitable land is 17 424 km2. This figure is believed  to  be a gross overestimation  ol lhe land  suitable for irrigation  The estimate assumes
TAMS-ULG haro-akoiio river qasin integrated dtvelopment master plan
ii1.          Natural Resources
that all land units with slopes Jess than 8% arc broadly suitable for irrigation. Whilst such a slope might he an  upper limit for sprinkler or drip  irrigation,  it is too  high  for basin  or furrow techniques. Although a deduction is made for possible unsuitable soils in some of the units, the criteria used arc not explained in detail The potential irrigable land has been identified according tn the following criteria:
•     land with slope less than 8%;
•     availability of a perennial surface water supply of adequate quality capable of
commanding at least part of the land by gravity irrigation; and
•     suitability of lhe soil.
Following this, an assessment was made of the water available for irrigation to refine lhe above estimate to define an upper limit to the irrigable area. The irrigable area was then evaluated for
the suitability for irrigation using the following criteria adapted from the United States Bureau 
of Reclamation, Table 1,33,
lhe screening  of possible irrigation  sites identified  18  areas with  a gross irrigable area of 109 300 ha. This area is reduced to 58 671 ha when water availability is taken into account The gross irrigable area of the sites identified is shown in fable 1.3.6. The irrigable area of each site has been classified according to the suitability of the land for irrigation. Overall, 2 690 ha (4.6%) is class SI, 37 623 ha (64 1%) is S2 and 18 358 ha (31.3%) is S3. The area of land in each suitability class for lhe individual sites is shown in Table 1.3.7 and the land characteristics in 
.Annex 2.E.3 in Volume HE of the ARDCO-GEOSERV report
The methodology' used to arrive at the area of land suitable for irrigation is unsatisfactory as explained above The evaluation needs to be repeated using more realistic slope criteria. A further problem lies in the identification of the 18 sites with potential for irrigation and in particular the collection of soil data for each site. The soil survey report does not give details of the surveys carried out to characterise lhe soils, it is possible that no field survey has been carried out and that the identification of the soils has been done entirely from remote sensing If this is so, then the areas of land identified for irrigation must be ground surveyed.
Land Tenure
The importance of security of tenure cannot be over emphasised. Farmers are unwilling to undertake long term improvements to land if their ownership or tenancy is not assured ARDCO- GEOSERV have not addressed this issue and this could renect the fact that as ownership of all land is vested in the Government land tenure was not regarded as an issue. However, the Government is aware of the importance of land tenure.
Indications that Government may he considering changing lhe system of land tenure seems to have encouraged farmers to occupy land formally unused to establish their occupancy ahead of any  changes.  This may  explain  why  farmers are occupying  forest land  m the upper basin although declining productivity of arable land may also be important.
T AM S* U1BAR0‘ * KOR0 w■V£R BASTS ,STPCiKATW development MASTTR PLANI.         Natural Resources
Also,  the Study  provides no  data in  the land  use section  on  the size or degree of fragmentation of land  holdings and  how this is changing  over time.  These two  measures might indicate increasing  pressure on  land  and  where land  use planning  interventions might be necessary.  1  his might apply  in  the north  east of the upper basin  where land  use intensity'  seems to  be increasing, but to confirmation this will require further investigation.
Land Use Planning
No land use planning seems to have been carried out as part of the Study. The Final Report gives estimates of the land  suitable for several uses but.  with  the exception  of land  suitable for annual cropping,  this has not involved  a systematic evaluation  of the land,  The Study  has neither identified  the constraints to  land  use nor recommended  improved  land  management techniques. Basic data on  existing  land  use is not provided  and  withoul this as a starting  point land  use planning is not possible.
Insufficient attention  was given  to  soil conservation.  There is a simple soil conservation classification  and  an  erosion  hazard  map  is one of the maps in  the .Atlas.  Unfortunately,  there is no  indication  of how this map  was produced.  Tn  addition,  there is no  description  of the soil conservation  measures necessary  for arable land  or forestry  and  no  discussion  of the alternative soil conservation techniques available.
L43 Lower Basin
The approach  and  methodology  of the Russian  study  differed  from that used  by  ARDCO- GEOSERV.  Therefore,  the sub-headings under which  ihe study  findings arc discussed  are also different,  The data on  land  use came mainly  from the 1983  aerial photographs,  field  observation and  government statistics,  fhe soil survey  provided  the data on  soils and  land  required  for the land evaluation.
Agra-ecological Zones
The Russian  study  did  not describe the agro-ecological zones of the area according  to  any nationally  developed  system.  However,  the Gambela plain  was divided  into  four zones on  the basis of soils,  topography,  hydrology  and  vegetation  Table 1.3.8  summarises the four main features of the four zones.  The agro-ecological units devised  by  FAO and  described  above also cover the Gambela plain.  The units identified  by  FAO as occurring  on  the Plain  are WT4,  WL4T WL5 and WL6.
Land Use
The  area  of  the  Gambela  Plain  is  24  276  knr  of  which  266  km2   was  cultivated  in  J  987  The  land use as estimated by the Gambela Plain study is shown below*.
TAMS-ULG IJARO-AKOBO RIVER BASIN INTEGRATED DEVTl.OPMENT MAST] It Pl AN1.         Matursl Resources
Land Use
1
Area (km )
Cultivated land
266
Savanna
18 048
Forests
5 658
Urban and roads
44
Water bodies
117
Other (rock, gullies etc)
143
Total
24 276
As for the upper basin, there is no attempt to determine if land use is changing or the reasons for any changes if they arc occurring Limited evidence suggests that clearance of forest on the foot slopes and in the savanna woodland is happening but the extern and location of such changes are unknown This deficiency should be rectified
There are several different ethnic groups using  the Gambcla Plain  and  the resulting  land  use pattern  is likely  to  be complex  as permanent cropping  systems coexist with  extensive pasture and  transhumance systems.  The resulting  land  uses and  traditional rights to  land  occupancy should be described and mapped.
1.4.4 Land Suitability
The evaluation of the land in the lower basin has been carried primarily with irrigated agriculture in  mind.  In  the land  capability  assessment used  soil characteristics are matched  against criteria and assigned to a capability group.
The Report docs not state explicitly  which  system of capability'  it uses nor does it specify  the class limits, although it is most likely based on the United States Bureau of Reclamation method The map  legend  and  report indicate that the main  criteria for delimiting  groups are drainage, flooding,  slope and  soil depth  llic class limits used  have been  inferred  from the report and  arc summarised in Annex 4 Volume 5 of the Russian report. These suitability classes are defined as follows:
Class I
Land not requiring fundamental reclamation
Class II
Land not requiring fundamental reclamation but with certain limitations
Class 111
1 and needing run-off control
Class IV
Land needing flood protection
TAMS-VMi IIARO-AKORO RIVER BASIN INTEGRATEDDEVELOFMENT MASHR I'LANL
Aalural Resources
Class V Land needing flood protection and drainage
Cla
ss VI Land
 unadvisable for agricu
ltural improvement
Annex 
4 
Volume
5   summarises   the
characteristics   of   the   suitability 
mapping   units,   the
improvements necessary and the area of each unit.
The Russian  study  assumes that most of the soils in  mapping  units (Class) I to  V are potentially suitable for some form of rainfed  or irrigated  cropping  provided  that the constraints can  be removed  The limitation  of this approach  is that it gives a high  figure for potential irrigable land and it does not cost the provision of the engineering works necessary to alleviate the constraints.
The Russian  study  has not carried  out a crop  suitability  evaluation  as ARDC O-GEOSERA has done for the upper basin  The crops and  crop  groups mentioned  in  Annex  4  Volume 5  arc extremely  broad.  For example,  'all climatically  adapted  crops'  is too  broad  a grouping  of crops; not all climatically  adapted  crops arc suitable for the soil conditions on  the plain  even  in suitability Class 1-1 IL
Also,  rice may  grow well on  heavy  soils,  but the high  swelling  and  shrinking  nature of the vertisols in  class IV and  V might require annual land  levelling  to  maintain  an  even  water depth in  padi fields.  Similarly,  sugar cane can  grow on  heavy  clays but root development and  growth will be reduced  without frequent deep  ploughing.  Some fodder crops will grow on  heavy  clays but it must be stated  which  ones.  The suitability  of crops for irrigation  will require further detailed review during Phase II.
Of particular importance to  irrigation  are the physical properties of the soil especially  infiltration rates and  hydraulic conductivity.  Table 13.9  summarises these parameters.  It is generally recognised  that infiltration  rates in  excess of about 545  rru'day  are excessive for surface irrigation. The Russian  study  has correctly  excluded  mapping  unit lib  from irrigation.  Hydraulic conductivities of <1  m/day  are usually  thought to  be non-irrigabk  as these soils require excessively  close drainage spacings.  This applies to  mapping  units lie,  Va and  Vb.  The Report has excluded  units Va and  Vb  from irrigation  development on  the basis that conductivities are too  slow and  because flooding  would  necessitate empo  Ide ring  and  pump  drainage which  would be prohibitively  expensive.  Mapping  unit Tic is deemed  suitable for irrigation  without provision of drainage.  In  view of the low hydraulic conductivity  values this unit might require sub-soil drainage if irrigated.
A further point to note is that no mention is made in the soil or agriculture sections of the Russian study  of the possibility  of rising  Ground  water tables following  irrigation.  This is an  important consideration  and  the possibility  must be taken  into  account.  Although  Lhe soil report concludes that Ground  waler salinity  is unlikely  to  be a problem in  the lower basin,  high  Ground  water tables per je, especially in die rainy season might hinder crop growth.
Land Use Plan
The  table  below  summarises  the  different  agricultural  uses  recommended  by  the  Lower  Basin Study.
tams-ulg liAROAKOBC RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
25I.         Natural Resources
Zone
Total
Land Recommended for Crop Production (km )
Area
km1
Gross
Net
Irrigated
Rainfed
Western
8 890
40
40
*
40
Northern
3 600
1 160
870
700
170
Central &
7 390
2 760
2 060
950
1 HO
eastern
South-east
4 390
2 880
2 090
300
1 790
Total
24 270
6 840
5 060
1 950
3 110
The Western  zone is not recommended  lor crop  development due to  the problems posed  by flooding  and  poor drainage.  The bulk  of the area is recommended  for natural rangeland,  wildlife and  tourism.  The area of 40  km1 is land  already  cultivated  in  scattered  farms.  Similarly,  in  the remainder of the northern  zone crop  production  is not recommended  with  rangeland  and controlled hunting being the recommended land use.
By taking into account soil suitability, labour availability, access and distance from large settlements, the Russian Study has formulated two development options. The irrigated option envisages the development of 780 km  of which 420 knv is for irrigation and 360 km  for rainfed 
2
2
crop development. The rainfed option would develop 1 074 km  of land. 104 km  for irrigation 
2
7
and 970 knv for rainfed cropping. A series of text maps identify the location of the different areas
of recommended land use.
1.4.5  Evaluation of Data for Master Planning
The purpose of this section  is to  summarise the evaluation  of the data collected  by  the previous studies for the purpose of master planning  and  to  make recommendation  for further data collection  and  evaluation.  The section  begins by  discussing  the deficiencies of the upper basin and the lower basin before discussing further data requirements for both regions.
Upper Basin Data Deficiencies
The  land  use  data  presented  by  ARDCO-CiEOSF.RV  is  insufficient  for  master  planning  purposes for the following reasons:
•  The  Study  presents  no  basic  statistics  regarding  land  use.  There  is  no  indication  of total  land  area,  arable  area,  area  and  production  of  major  crops,  fallow  land,  pasture etc.  Without  this  basic  data,  it  is  not  possible  to  describe  and  quantify  the  present situation.  The  proportion  of  arable  land  currently  cultivated  and  the  potential  arable land  available  for  development  is  not  known.  Without  this  data  the  potential  cannot be assessed Connected to this is the need to identify particular land uses with
T AM S-l TG HARO-AXOBQ RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
26L         Natural Resources
particular  ethnic  groups.  This  applies  especially  to  area  of  potential  overlap  of  land uses such as arable cropping and grazing.
•      There  has  been  no  attempt  to  define  changes  in  land  use  over  time.  Land  use  is dynamic  and  if  the  changes'trends  can  be  identified,  the  reasons  behind  such  changes can  be  analysed.  .An  example  of  an  important  change  about  which  little  is  known  is the  clearance  of  forest.  If  the  changes  can  be  quantified  and  the  reasons  identified, appropriate  interventions  can  be  designed.  For  example,  is  forest  clearance  a  result of increasing population or declining productivity of existing arable land ?
•      Areas  of  land  degradation  have  not  been  identified  and  the  pressures  leading  to  this have  not  been  analysed,  A  zoning  of  the  upper  basin  taking  into  account  population density  and  soil  and  land  type  might  help  in  identify  ing  areas  at  risk.  Once  this  is done, land husbandry' improvements and land use planning can proceed.
The  ev  aluation  of  land  suitable  for  crops  is  satisfactory'.  However,  no  evaluation  has  been  made for alternative land uses. The following comments apply:
•      The  suitability  evaluation  gives  the  current  suitability  of  the  land  it  does  not  give  the potential  suitability  with  improved  husbandry.  Il  is  possible  io  alleviate  some  of  the constraints  limiting  suitability  through  improved  husbandry  such  as  soil  conservation. In  addition,  there  is  no  description  of  the  land  utilisation  type;  it  is  unclear  what system of production has been evaluated.
•      The basis of the land evaluation of the potential irrigation sites is unclear. The precise field  investigations  undertaken  to  evaluate  the  soil  are  not  mentioned.  For  this  reason the area of land suitable for irrigation must be used with caution.
•      No  evaluation  has  been  undertaken  for  alternative  land  uses.  Without  this  competing uses for land cannot be ranked in terms of suitability and no land use plan prepared.
There is no  attempt to  produce a land  use plan  ARDCl>-GEOSREV has not identified  anywhere the major constraints to  development of the land  and,  therefore,  no  attempt is made to  identify the technologies to  alleviate them.  Similarly  there is no  attempt to  identify  areas totally unsuitable Tor agricultural development,  areas where rainfall is sufficient for reliable crop production  or areas which  might benefit from supplementary  irrigation.  There is little analysis to support the development of land and water resources.
Lower Has in Data Deficiencies
As for the upper basin,  there is a need  to  colled  basic land  use statistics.  No  systematic crop suitability has been undertaken. In order to assess more accurately the crops to be included in any with-project situation,  a crop  suitability  for rainfed  and  irrigated  conditions must be carried  out. This will not involve any  further data collection  but a re-evaluation  of existing  dam.  No additional land  use planning  seems to  be necessary  at this stage.  If it does become necessary  as a result of major changes in land suitability, no further data collection is envisaged,
TAMS-ULG b aho-akodo hiver basin entloiu rro oevelopmext master plan
271.         Natural Resources
1.4.6      Further Data Collection
The additional data to  be collected  falls into  two  categories,  digital data from the GIS/database systems and woreda statistics The digital data consists of digitised maps with associated attribute data. The following is required:
•            Forest inventory data from the Woody Biomass Inventory and Strategic Planning Project. This  is  based  on  satellite  imagery  from  about  1988  and  will  provide  the  latest  data  on forest cover
•            South  West  Ethiopia  Forest  Inventory,  Chaffey  1978.  This  is  based  on  interpretation  of aerial photography from 1975.
•           Master Land Use Plan, FAO 1982 (?) Maps of land units, soil and physiography
•            A Summary of the Agricultural Ecology of Ethiopia FAO 1988. Maps of agro-ecological zones and supporting data.
•            Soil Conservation  Research  Project This project has produced maps identifying areas of erosion hazard; these should be obtained and digitised for the project area.
•            The  population  capacity  support  model  as  updated  by  Woody  Biomass  Inventory  and Strategic Planning Project plus maps.
•            The  original  Land  Lise  Planning  and  Regulatory  Department  (1989)  report  along  with the maps for analysis during the Phase II
In  addition  to  this  data,  the  following  maps  of  the  lower  basin  produced  by  the  Russian  study need to be digitised:
•           Soil map; four sheets. 1:2O0 000 scale
•            Soil suitability and reclamation measures; four sheets, 1:200 000 scale
•           Pastures; four sheets 1200 000 scale
It is understood that the maps produced by ARDCO-GEOSERV arc already digitised.
The Woreda statistics required  cover land  use and  should  be for three years,  the most recent available and for five years and ten y ears ago. .Xs well as identifying the extent of such land uses their location must also be mapped. The information required is:
•            land area
»          annual cropping
•            perennial cropping
•            plantation crops
•            fallow
T XMS-ULG b aro-akobo river basin integrated development master han
211- Natural Resources
• 
forest
• 
grazing
• 
waste
• 
waste  and shrub
If available,  data on  size of average holding,  class size distribution  and  fragmentation  should  also be collected,  MW should  collect and  check  the above data and  iraeeepiablc it must be entered into the GJS/Database. Without this basic data, no rational land use planning cart be undertaken.
1.4.7   Further Analysis.
The  further  analysis required  will  be  largely confined io the manipulation of data on the GIS. For the upper basin the analysis will:
•             Identify  land  use  changes  over  lime  through  a  comparison  of  forest  mapping  from  the mid  1970's  and  late  1980’s.  The  degree  of  detail  obtained  will  depend  somew  hat  on  the compatibility of the legends for each dala source.
•             Establish  land  use  intensity  related  to  population  density'  and  soil/land  suitability  by overlaying the thematic maps to identify areas of particular pressure 'degradation
•            Undertake  a  land  suitability  evaluation  for  land  uses  other  than  crop  production.  This  will follow   the   FAO   Framework   for   Land   Suitability,   lhe   environmental   requirements   for forestry,  rangeland  and  wildlife  will  be  established  and  areas  identified  by  matching  these requirements  to  land  units  through  the  GIS.  This  evaluation  will  be  extended  to  include a comparison of lhe economic returns from alternative uses of the same land unit.
•            Produce  a  map  of  soil  erosion  hazard  by  integrating  data  on  land  cover,  soils,  land  form and  climate  on  the  GIS  Areas  presently  at  risk  from  erosion  and  those  potentially  al  risk by project interventions will be identified
•            Identify- and map a comprehensive list of constraints to lhe development of the Upper
Basin imposed by soil and land. 
/'' . /
-I J'"
For lhe lower basin,  the most important outstanding  issue is to  re-evaluate the suitability  of the land  according  to  the FAO Framework  for Land  Evaluation.  This evaluation  will include all major agricultural land  uses (minted  and  irrigated  crop  rotations) and.  in  conjunction  with  other specialists,  forestry  ,  livestock  and  wildlife.  This evaluation  will be extended  to  inefode a comparison of the financial and economic returns from alternative uses of the same land unit,
1,5 Forests
The two major studies which contain information on the forests of lhe Baro-Akobo basin are the Russian study centered on the Giunbela plain of the lower basin and the 1994 ARDCO- GEOSERV data collection studies (Low Intensity) of die upper basin. The ongoing Woody Biomass Inventory and Strategic Planning Project has produced additional information, including GIS data. It is worth emphasising that the Elluopian plateau is one of lhe world's most important
TXMS-ULG UARO-AKOHO RIVER H ASIN ENie<j RATED DEVELOPMENT BLASTER PLANL         Natural Resources
watersheds and dial the forests in Baro-Akobo basin constitute part of that watershed As a result, (he forests in the basin are a vital component in maintaining the river flows. Millions of people throughout a large part of Africa depend upon the water that flow the rivers originating from the highlands of Ethiopia.
1.5.1  Upper Basin Studies
I"he ARDCO-GEOSERV report fails to  distinguish  between  evergreen  rain  forest and  other forms of wooded  stands,  either plantations or wooded  savanna In  section  4.2,  ARDCO- GEOSERV has a species group  which  has the coniferous species as Group  I,  and  being  poorly represented  it is amalgamated  with  Group  2.  This group  is classed  as broadleaf merchantable hardwoods, while Group 3 is potentially usable hardwoods and Group 4 is listed as other species. The Report talks about log volumes and commercial interests.
Such  a species classification  and  approach  for a forestry  resources study  in  a major watershed pinpoints the major problem with  the Study.  Clearly,  ARDCO-GEOSERV considered  broadleaf hardwoods as merchantable timber and not pan of a rapidly diminishing forest which is vital for the protection of the watershed. The broadleaf hardwoods are found in the remnants of llic clump forest on  the plateau  and  in  the slope forest of the escarpment.  This situation  reflects the need for urgent delineation and demarcation of the forest estatefs) as noted by Chaffey in his South- West Ethiopia forest inventory  (1978).  Further more,  the survey  makes no  mention  of the extensive and  very  widely  spread  plantations of fast growing  exotic species which  make up  a sizable part of the woody biomass in the surveyed area
The size and 'spread out' patchwork nature of (he clump broadleaf evergreen forest in the upper basin, the relatively short time spent in the field and the heavy reliance on G1S and past sub sector studies,  suggest that the quality  of the data presented  by  ARDCO-GEOSERV is questionable.  The survey  used  1986  photographs and  1989  l^ndsai imagery  From limned  field visits it is clear that the forests have been and are being rapidly encroached making imagery and photographs from even a few years ago unreliable unless the interpretation is supported by very extensive ground  truthing,  which  was not the case A further problem arises with  the use of satellite imagery  .  The relative similarities of shading  values between  F.l (closed  forest) F.2 (closed * open + fragmented forest) and F.3 (closed + open * fragmented forest + long fallow) requires detailed  ground  truthing  The over reliance on  GIS imagery'  without sufficient ground truthing over selected forests throughout the study area can lead to faulty interpretation. At best, it will increase the chance of errors
1.5.2   Lower Basin Studies
The Russian  study  must be considered  a ’bench  mark  study  of the lower basin.  The botanical studies were detailed  and  thorough  However the study  was not a forestry  study  or survey. Indeed the report itself makes this quite clear. It was a geo-botanical study that included wooded savanna for the purpose of species identification but the comments on lhe wooded savanna and evergreen  catchment forests were descriptive narratives.  While they  surveyed  lhe plain  on  foot, bv vehicle, the steep slope forests of the watershed was visually suneyed from a helicopter
T \MS-ULG BARO-AKOBO RJVER BASIN INTEGRATED DEVELOPMENT MASTER FLAN
10L         Natural Resources
The Russian  study  classified  all trees as pari of the forest,  wooded  savanna,  plantations and evergreen  slope forests while from the point of view of catchment protection  it is necessary  to distinguish between these different groups because of their different values.
Another problem exist in  attempting  to  draw conclusions from the Russian  forestry  observations is that the data is now out of date.  it was gathered  during  1987-8,  During  and  since that time there has been  large scale resettlement and  refugee population  influx  into  the region.  The daily impact of so  many  people on  the forest resource base has been.  and  continues to  be enormous. The Russian  study  ignores the refugee question  entirely.  There is,  therefore,  a total lack  of up-to- date forestry resource data for the lower basin.
1.53 Woody Biomass Inventory and Strategic Planning Project
The Project aims to  develop  a strategic planning  capability  from assorted  resource data bases for long  term biomass management In  many  ways,  the project creates a false sense of security.  By including  woody  biomass from all sources as a possible energy  source,  the perception  is that the country  has more woody  biomass for fuel than  it docs.  Evergreen  watershed  forests,  vital for catchment protection,  are included  with  plantations,  woodlots,  wooded  savannas anti any  other possible energy  sources.  While die nation  needs to  know what biomass energy  resources it has to  promote strategic planning,  the inclusion  of all foresi types,  including  the evergreen  watershed forests,  in  the inventory  is overestimating  the supplies.  Protection  of the evergreen  watershed forests rather than  lheir inclusion  as a potential source of woody  biomass is in  fact required.  The FAO 1988  report slates that the pressure on  forests for fuel wood  exceeds by  60% the forests capability  to  regenerate its self.  A very  significant argument for not considering  them as possible fuel sources. Again, there is an over reliance on GTS data and without ven good ground proofing and legends* there is a real danger of over estimating the resource.
1.5.4 Data Requirements
The ARDCO-GEOSERV Low Intensity Forest Survey has inadequate GIS data upon which to base a master plan study. The amount of ground truthing done in a comparative short time span leaves too much room, for error, A completely new set of matrices covering both the upper and lower basins is needed. The amount at ground truthing should be geared to an "in-depth” forest survey to provide on accurate assessment of all forest types and area; els well as the basis for monitoring future changes. Such a study w ould require several months of field work by four
r
inventory teams supervised by two foresters. Tn addition. Phase II studies would require the
following information which has so far been unavailable to TAMS-ULG;
• 
complete forestry inventory maps for the upper and lower basins;
• 
inventory data and maps of the 1988 to date study from the Woody Biomass Inventory and Strategic Planning Project*
• 
Population Capacity Support Model used by the Woody Biomass Inventory and Strategic
Planning Project;
• 
complete high priority forest areas demarcation and delineation mapping;
TAMS-ULG iiaro-akoto river basin integrated devei iipment master plan
311. Natural Resources
•            fuel wood plantations inventory;
•            FAO Forest Cover Map 1992;
•            woody biomass supply-demand balances data;
•            livestock grazing-systems data;
•            bio-fuel production, collection and consumption data; and
•            information on the forestry department (including research), including organisation, 
staffing, facilities and programmes.
1.6 Wildlife
The  term  wildlife  strictly  speaking  includes  all  fauna  not  under  domestication,  and  as  such greatest  species  diversity  and  numbers  would  be  found  in  the  invertebrates.  For  the  purpose  of this  report  wildlife  will  be  taken  to  refer  to  those  species  which  have  the  potential  to  attract tourists.  How  ever,  this  simplistic  approach  ignores  the  concept  of  biodiversity,  as  well  as  the potential   of   indigenous   flora   (including   endemics)   to   attract   specialist   tourists.   Hence,   this overview  deals  mainly  with  the  larger  mammals  and  to  a  minor  extent  with  the  avifauna Although there have been recent aerial censuses of wildlife in Ethiopia (eg Thouless, 1995), none of  these  have  covered  either  the  upper  or  lower  basin.  Section  1.8  covers  fisheries.  The  main sources  of  information  are  the  Russian  and  ARDCCTGEOSERV  studies,  although  information from other studies has complemented these two dau sources.
1.6.1 Upper Basin Studies
With  regard  to  the status of wildlife in  the higher-lying  areas of the Baro-Akobo  basin,  the situation  is unclear.  The faunal data for forests in  the upper basin  is particularly  poor,  largely because of the manifold  difficulties of recording  and  enumerating  animals in  dense forest.  Only those species which  occupy  the upper canopy  (eg  colobus monkey  CoJcbur guereza) can  be recorded  without adopting  specialised  techniques.  There have been  previous elephant censuses and  these are summarised  in  the table below.  This table indicates that rates of decline have been as high  as 80% (assuming  a degree of accuracy  of the data).  ARDCO-GEOSERV reported  on the presence of elephant in  the Bebeka and  Guraferda areas,  and  this must be assumed  to  be consistent with  Akobo  river - Mizan  Tefen  area of the table.  However,  ARDCO-GEOSERV only recorded  qualitative data (for example,  "upprecioMe in  number**). Note that the Amaya/Bonga area falls outside of the project area,  but its proximity  justifies its inclusion  as an  indicator of general trends
T AMS-U LG baro-akobo river basin integrated development master plan
321.         Natural Reionrccs
REA
1986(1)
1990 (ii)
Akobo river - Mizan Tcferi area
6 000
1 000
Ciambcl a National Park
no data
100
Amaya Bonga area
no data
200
Source: (1) Largcn & Yalden (1987); (ii) Allen-Rowlandson (1990)
Some authorities have used  elephant as an  indicator species,  in  that its presence can  often  be linked  to  the presence of a range of other large mammals,  if this perception  of the elephant as a "flagship  species" is valid,  then  one can  assume equally  drastic declines in  a range of other species which  respond  to  habitat loss,  to  competition  with  humans for land,  and  which  are at the same time susceptible to poaching pressures.
ARDCO-GEOSERV also  compiled  a species list for mammals of the upper basin,  apparently based  largely  on  anecdotal information.  As this section  of the report ( Volume Ill.G) included major errors in  zoological classification  (eg  designation  of the pangolin  Phataginus lemminckii as a primate,  whereas in  fact it belongs not only  to  a separate family,  but to  its own  distinctive Order), the value of the document as a tool for decision-making (as would be needed for a Master Plan) is questionable.
For avifauna,  the ARDCO-GEQSERV study  drew largely  on  existing  data,  and  again  presented a species list.  The designations rare',  common’ and  ’abundant’ (also  employed  for the ARDCO- GEOSERV mammals species list) arc of little value,  particularly  as they  allow for no repeatability were the status of any species to be re-assessed,
Apart from the surveys undertaken  through  Ethiopian  Wildlife Conservation  Organisation,  and the comments contained  in  the ARDCO-GEOSERV report,  there are minor unpublished generalised  statements available through  the then  Natural Resources and  Environmental Protection  authority,  mainly  written  by  foresters with  no  specific training  in  wildlife appraisal techniques.  For example,  the lilubabor Zone Natural Resources Environmental Protection Department has produced  a short report (Ketema & Kebede,  1995) on  the status of wildlife in that area.  Although  listing  those mammals which  are recorded  for the area,  only  qualitative rankings were made ("abundant”,  "common'  and  “rare").  The Ethiopian  Valleys Development Authority  (1990) reconnaissance study  noted  the presence of large mammals including  elephant in  the study  urea.  However,  the brief species list of that report included  species not found  in Africa (eg tiger), and little credibility can be given to these data.
The important point is that no  information  on  fauna of any  scientific value is currently  available for the dense indigenous forests.  Those biological publications of any  use are restricted  to botanical treatments (eg Friis er al, 1982).
TAMS-ULG I1ARO-AKOBO RIVLR HaSTN INTEGRATED DEVELOPMENT MASTER PLANL         Natural Resources
Although  there are proclaimed  forest reserves in  the upper basin,  according  to  ARDCO- GEOSERV,  only  two  areas (Mizan-Tcfcri-Tcpi and  AkohoTebo) have been  designated  as Controlled Hunting Areas.
L6.2 Lower Basin Studies
Lillie (1969) staled "f/ie Garnbeta diffricZ teams with wildlife, including elephant, buffalo, tiang, waterbuck,  roan  antelope,  giraffe,  white-eared  kob  and  Nile lechw't”. This historical status for
the area was doubtless a factor in  attempts to  give the areas some level of protection.  The "protected  areas" are nominally  the Gambcla National Park  (not a National Park  in  the strict sense,  as it remains to  be gazetted),  the Jikawo  Controlled  Hunting  Area,  and  the Tedo Controlled Hunting Area.
Hillman  (1993a) gives a "conservation  area summary1*  for the Gambcla National Park  which gives the major wildlife species being conserved as roan antelope, while-eared kob, Nile lechwe, topi and elephant. Hillman lists no endemic species. However, the kob recorded for this area is the subspecies Kobur tob  JeucofM,  which  is restricted  to  the adjacent region  of south-eastern Sudan  and  part of the Gambela National Park  Under undisturbed  conditions,  the species composition  of this ungulate community  was very  similar to  that of the southern  Sudan,  and indeed  introduces one of the problems in  protecting  and  managing  such  communities: that several species are migratory ( white-eared kob. lechwe. and tiang), and migration in the Gambela region will involve the crossing of international boundaries.
The Russian  study  reported  on  high  densities of wildlife in  the south  and  south-west of their study  area,  based  on  a surprisingly  high  proportion  of the area sampled  from the air (39%), Results were presented for fourteen species, for the wet season of 1988, and for the dry season of 1989. These data were presented together with comparative data for 1977, and included short descriptions of preferred habitat and location for each species enumerated. As these data reflected the abundance of those species which  occur in  habitats with  good  visibility  from the air,  the reliability of the data is to some exlent a function of the habitat sampled, as well as of the size of the animal in question. Of note w-as the dramatic change in numbers of white-eared kob, from estimates of over 100 000 a decade earlier, to no records for 1988/89. To the extent that these data arc reliable, this comparison gives some indication of the rate of change in numbers which can lake place for some species. In any event, these data are now* over six years old, and bear no relationship to present conditions.
With  regard  to  migration  patterns of large mammals,  the Russians inferred  this from air photographs to  model localised  movements around  seasonal water sources.  Although  they  gave a generalised  account of ungulate movement (a predictable dry  season  dispersal to  the wetter grasslands of the west with a rainy season movement to the higher levels of the watershed), no information  was made available on  the patterns of international migration,  including  where for certain species important events like reproduction took place
No data on wildlife of the montane tropical forests of the eastern pan of their project area were collected  by  the Russians Data on  avifauna from both  the Russian  report and  from Hillman (1993a) are too  generalised  to  be of much  ornithological significance Hillman  does how ever provide an extensive species list for the Gambela National Park (152 species from 63 families.
TAMS-H-C b.aro-akopo river basin integrated development master flam
341.         Natural Resources
with   no   endemics),   but   no   ecological   data   are   available   beyond   this   pure   presencc/absence presentation.
For al) previous reviews,  it should  be stressed  that attention  was devoted  only  to  larger mammals, supplemented  by  bird  species lists.  There is a marked  paucity  of information  on  herpetofnuna and invertebrates.  For example,  the Russian  report noted  that the swamps of the Gambcla plain  dry up  in  the dry'  season  •  **and  turn  info  lifeless areas'1. Such  an  interpretation  fails to  address the enormous importance of these depressions for amphibia and  even  lungfish,  which  overwinter in mud cocoons beneath the soil surface.
The present situation  with  regard  to  wildlife is radically  different from any  of the accounts summarised  above.  No  large wild  mammals were seen  during  the consultant s field  trip,  and  it appears that the Russian  predictions have to  a large extent been  fulfilled: i/ prompt and  drastic measures are not taken  to  stop  poaching.  the wild  animals will be killed, in  the main,  within  the next fewr years". Those surviving  populations of larger wild  mammals will be restricted  to  the southwest of the National Park,  or to  the more inaccessible parts of the swamps in  the extreme west.
1.63 Protected Area Recommendations - Previous Reports
Previous  reports  have  recommended  that  certain  areas  in  both  upper  and  lower  basins  be  given “protected" status. These are summarised below.
Upper Basin
The /XRDCO-GEOSERV report made a set of recommendations for the creation  of National Parks (3).  Wildlife Reserves (5),  and  Controlled  Hunting  Areas (2).  However,  the national Wildlife Policy  document has yet to  be finalised  and  several important points related  to protected  areas await clarification.  First,  are the original legal categories of protected  areas to  be retained? Secondly,  does the Ethiopian  Wildlife Conservation  Organisation  intend  to  include additional protected  areas under the existing  'wildlife estate '? And  thirdly,  does the current status of wildlife in  presently  proclaimed  areas (or in  areas recommended  for protected  status) still justify the designation of such areas as categories for either actual or potential protection ?
Low er Bar in
In  recognising  potential impacts on  the wildlife resource,  the Russian  report proposed conservation  zones,  as well as Controlled  Hunting  /Xreas.  There is no  doubt that the status and location  of any  proposed  protected  areas would  have to  be redefined,  so  rapid  have been  the changes in  wildlife resources The Russian  report advanced  the following  wildlife conservation measures:
• arrangement of lands to lake into account seasonal movement of wildlife; in these proposals,  a  rather  simplistic  zonation  was  envisaged,  with  relatively  ’’hard  edges” between  cultivated  lands  and  areas  allocated  for  wildlife  (although  measures  were advanced  for  layout  of  fields  to  include  strips  of  natural  woodland  between  plots  0,5 km in width, no consideration was made of the obvious snaring that would result
TAMS-ULG bar oak 000 river basin integrated development master plan
)51.         Natural Resources
form lliis layout; proposed  measures to  keep  game off crops - construction  ot watering places, channel crossings, and location of artificial salt licks - would in any event have proved unworkable, given experience elsewhere in Africa);
•     proposals for the establishment of prelected areas (both National Parks and 
Controlled Hunting Areas); and
•     the establishment of an experimental farm for game farming, for the "free or semi
free breeding and keeping of wild animals'*.
The first recommendation falls away on two grounds: greatly altered situation in terms of wildlife numbers and  distribution  and  the changed  status of State Fanns,  The second  proposal is geographically  no  longer applicable,  again  owing  to  the changed  status of wildlife populations. Since the time of making  the third  proposal,  there has been  much  work  across Africa to demonstrate that such  approaches are not economically  feasible,  unless for specific species (eg crocodiles and  ostriches) with  defined  commercial outlets.  Besides these specific commercial species,  it is tourism and  safari hunting  which  are those forms of wildlife utilisation  which  are economically feasible.
L6.4 Data Requirements for Master Planning
Hie reviews of prcsious studies has shown that most reports on wildlife proceed no further than 
mere species lists; where population estimates have been made, these are now- hopelessly out of
date. The Russian report acknowledged that no data were collected for the mountainous areas to 
the east of die Gambela plains. A further area of uncertainty concerns wildlife policy and related legislation. At the lime of this Inception Report (November 1995) a draft wildlife policy 
document has been prepared but still awaits ratification at both central and regional government
levels. A workshop is scheduled for January 1996 to finalise wildlife policy details, as well as
details of die proposed Trust Fund. The Master Plan cannot adequately address w ildlife issues
unless proposals arc consistent with both national and regional policies on wildlife. There are key wildlife developments and proposals (and hence, data collection recommendations) which can 
only proceed once both Wildlife Policy and Wildlife Legislation documents have been approved 
and ratified.
Policies
The finalisation of the wildlife policy should coincide with the start of Phase 11 and it is hoped 
that the workshop in January will provide the guidance required for master planning and hence
will be able to assist in finalising the data collection requirements. A failure to finalise the
wildlife policy will impede the preparation of the master plan In addition, the finalisation of the
Nanon Conservation Strategy is under review by the Council of Ministers and any master plan 
needs to be consistent with this both at the national and regional levels.
Swrwy of W ildlife Status
There is a need for basic information. largely ot a scientific nature, regarding the status of natural resources and all planning on wildlife utilisation and its integration into any Master Plan must
y S-l I LG BARO-AKOBO RIVER R ASIN INTEGRATED DEVELOPMENT MASTER PLANL         Natural Resources
be  seen  as  academic,  until  an  internationally  acceptable  survey  has  been  undertaker.,  there  are essentially three approaches that could be taken, with regard to the collection of outstanding data:
•    collection of the data before proceeding to Phase 11;
•     collection of the data whilst the Phase 11 programme is underway; and
•      in  cases  where  the  amount  of  data  required  is  unduly  extensive,  to  relegate  the  data collection  phase  to  the  context  of  a  national  programme,  essentially  separate  from  die Phase II programme
For both  upper and  lower basins,  there is a clear need  to  link  the status of wildlife (expressed ideally  in  lernis of population  dynamics,  to  the availability  of species-specific habitat). Approaches to  wildlife census techniques should  be discussed  with  Ethiopian  Wildlife Conservation  Organisation  and  with  similar projects like Omo-Gibe so  that a consistent, standardised and repeatable system is adopted for Ethiopia.
lower Sorin
There is no  doubt that the Lower Basin  has enormous potential for wildlife; even  though depletion  of wildlife resources has taken  place,  the habitat is largely  unaffected  outside of areas of human  settlement,  and  there is the potential for natural restocking  for some of the migratory' species. Current species lists need to be supplemented by ;
•      population estimates for larger mammals;
•     habitat appraisals, specific for different "suites" of species, including relative extent
of these habitats;
•     biodiversity inventories and collections for avifauna, herpetofauna, small mammals
and invertebrates;
•    an understanding of migration patterns, including those associated with the livestock 
herds of the Nuers; and
•      information  on  the  perceptions  of  the  local  communities  as  a  basis  for  integrated approaches    to    wildlife    planning    and    to    develop    a    framework    for    Integrated Conservation and Development Programmes.
The biological output w ould  include: definition  of population  status,  habitat requirements and extent (for key  animal species); understanding  of international migration  patterns; development of a standardised  system of vegetation  classification  (to  be consistent with  the Ethiopian  Wildlife Conservation  Organisation  and  other relevant Ethiopian  organisations).  Once the basic data have been  collected,  a workshop  would  be held  to  present proposals for park  management and boundaries to  all stakeholders,  including  peasant associations,  representatives of tribal groups, local authorities. National Conservation Strategy representatives, etc
TAMS-ULG iiAitoMKono river ijastn integrated development master flan
r1.         Natural Resources
To  meet  these  needs  effectively,  it  is  estimated  that  a  team  will  be  required,  comprising 
a  wildlife  biologist,  a  limnologisl'plant  biologist,  and  a  sociologist.  About  six  months  of the field would be needed, with an adequate budget for air surveys.
Upper Basin
Wildlife habitats in the upper basin are largely gallcricd forests which do not lend themselves to conventional aerial survey techniques and will require a more specialised long-term approach
1.7      Tourism
1.7 J Upper Basin
The ARDCO-GEOSERV study  drew attention  to  the aesthetic potential of landscape and  forest areas to  attract visitors,  the interesting  ethnic groups,  as well as commenting  on  the potential for
"adventure sports" like white water rafting.  Certainly,  the potential of the indigenous forests to meet the needs of specialist botanical interest groups is enormous.  Thus,  the potential for tourism, although sound, does not centre around wildlife, which was at one time the focus for the Lower Basin.  However,  there is a need  for supportive infrastructure,  including  transport, improved  road  systems,  promotion  and  tourist accommodation  These standard  tourist needs arc summarised  by  ARDCO-GEOSERV,  including  proposals for both  ground  and  air tourist
’circuits”.
In  addition,  the ARDCO-GEOSERV report made a set of recommendations for the creation  of National Parks (3),  Wildlife Reserves (5),  and  Controlled  Hunting  Areas (2).  These proposals are of course directly related to tourism, but are dealt with under Section 1.6 of the Inception Report. In summary, the major problems with the proposals advanced by ARDCO-GEOSERV are that:
• tourism development is dependent upon the securin’ of the resource base, and this has yet
to be demonstrated; and
the proposals fall short of recommending  the investment procedures to  establish  the initial infrastructure,  (assuming  that these will not develop  spontaneously  from the private sector),  aside from a statement that "such  costs should  be sanctioned  by  the central and/or regional governments”.
1.7.2    Low er basin
At the time of Little's report,  Gambcla was identified  for its wildlife attractions,  and  for the traditional ways of life of the local people,  the Anuaks in  particular In  fact Gambcla was seen as forming an important harbour navigable for nearly 1 400 km to Khartoum, "the door to the big game lowlands of the lllubabor Province".  To  develop  this tourism potential,  construction  of a 36-bcd  Bare river Lodge was recommended,  to  be built in  Gambcla town  at the confluence of the Bare and  Jebjcbe rivers.  The report even  included  detailed  design  specifications.  Other general recommendations were advanced,  including  setting  up  a feny  across to  the southern  bank 
TAMS-ULG B AR<>AKOBO RIVER BASIS INTEGRATED DEVELOPMENT MASTER PLAN
JSI. Natural Resources
of  the  Baro  river  (a  bridge  has  subsequently  been  built  in  1981),  and  "to  implement  the  wildlife department's recommendations'1.
A  five  year  tourism  development  strategy  was  proposed,  with  recommendations  on  immediate actions  to  stop  the  destruction  of  wildlife,  and  commitment  to  protected  areas  and  appropriate legislation.  Both  the  report  by  I  ANUS and that by Curilla concentrated on the area around Ji ma and Nekemte, and were thus not applicable to the Lower Basin
1*7.3 Towards a Master Plan - Data requirements for Tourism Sector
The following  are required  for Phase IL  to  allow incorporalion  of tourism into  a Master Plan  tor Baro-Akobo  basin.  First,  the outcome of several policy  documents will be required; however, this lies outside of the control of the master plan  study.  Secondly,  there are those data which  can be actively  collected  by  the project.  With  regard  to  the first need,  clarification  is awaited  on  the intention of government with regard io:
•             park  accommodation  investment:  for  example,  will  lax  concessions  be  offered,  while  at the  same  time  requiring  investors  to  pay  a  turnover-based  lease  fee  for  Ethiopian  Wildlife Conserv ation Organisation, and/or to local communities?
•             people  occupying  wildlife  protected  areas  what  will  be  government  policy  when  forced to make a choice between loss of biodiversity and resettlement?
•             who   should   manage   protected   areas,   Ethiopian   Wildlife   Conservation   Organisation   or regional governments?
•             approaches  to  management  of  wildlife  outside  of  protected  areas,  the  concern  here  is  the modus   optrandi   of   non-consumptive   safari   operators   (for   example,   those   wishing   to establish camps along the Baro river for the purpose of conducting canoe safaris).
The issue of occupation  of "protected" areas is one of extreme complexity,  given  the impact that resettlement has on  peoples'  lives.  These issues are discussed  fully  in  the draft Wildlife Policy document,  the final outcome of which  should  in  theory  give the necessary  guidelines on resettlement, which will be critical for any wildlife/tounsm planning in the project area.
With regard to the requirement for collecting new "hard' data, there are no major changes in the
"tourist interest base" of the upper basin that would justify expanding or updating the ARDCO- GEOSERV recommendations, except revision in the light of changes in policy direction. For the
lower basin, tourism will be a direct function of the status of wildlife, and data requirements for
that sector have been covered in Section 1.6.
The conventional strategy  for tourist development is to  identify  discrete sectors of operation, which  will normally  include: promotional campaigns,  competitive pricing  strategies,  training  of personnel,  removal of constraints,  subsidies for construction  and  operation  of hotels and  tourist infrastructure etc.  In  the case of the project area,  such  an  approach  would  be premature,  until such  time as commitment to  wildlife and  forestry'  protection  is evident.  Some measure of the
TAMS-ULG uaro-akoborivpr basin integraiw development master plan
391.         Natural Resource?
level of commitment will be made clear when wildlife and tourism policies are released, hut there still remains the question of commitment al local level
The Ethiopian Tourist Commission is making an assessment of tourist needs and market potential so  there is no justification  for duplication  of effort in this sector. What is important is that any of the master plan  proposals do  not detract from the tourist attractions of the region; most of these concerns are however addressed  under the Environment Section  (section  5),  where due consideration is given to environmental safeguards.
1.8      Fisheries
1.8.1 Status of Fisheries in the lipper Baro-Akobo Catchment (> 800 m)
The Report devoted  less than  one page to  fisheries and  provided  no  detailed  information.  In addition,  little data is available from oilier sources.  No  formal biological studies have been carried  out in  the region  and  no  assessment of the status of the fisheries has been  made.  The exception is an ad hoc survey around the Gore awraja by the Russian Delegation of the Science and Technology Commission
Specter
fhe Russians found that 40 fish species identified in the lower Baro-Akobo plains were present. 
The fish fauna of this region is probably made up of distinct community structures. On the
Fit Mekonnen plateau the river has a typical meandering pattern and floods extensively. As the
river descends from the plateau to the lowland plains it cuts through steep gorges and is fast Bowing. In this region rheophilic (fast water) species such as Barbus and Labeo will be found. 
On the floodplain species with a preference for slower flows, similar to those found in the upper floodplain area, will probably be found
Fishing
In comparison lo the lower catchment there is little fishing in the upper caichment. Some fishing 
is done on the Baro, Sur, Wabcr* YabL Dibo and Uka rivers, but this is purely on a subsistence
basis using traditional gears. The dominant species caught arc Oreochromis niloticus,
Oreochromis zilffi and Barbus spp. No data exist on the number or intensity of fishing. The
principal reasons for the lack of fishing are die absence of any suitable-sized, slow-flowing 
walers or lakes, inaccessibility of the major rivers and tributaries for most of their course, and lack of tradition amongst the local ethnic groups. The majority of groups (Oromes. Kcfa and Amharal are farmers and hunters and rely on livestock for their source of protein. Agnuaks (the traditional fisherman), only found in the Welega region in the Dale awraja, and they only fish to supplement their agricultural activities
Fisheries Development
There   has   been   some   evidence   of   attempts   to   increase   fish   production.   The   Fisheries Department of MOA stocked Lake Bishan Waka near Tepi with 11 000 tilapia fingerlings and
T A MS- VLG BAROAKOBO RIVER ElASfN INTEGRATED DEVELOPMENT MASTER N AN
40L         Natural Resources
Horta reservoir,  about 5  km west of Dembi Dole.  constructed  by  the Lutheran  World  Federation for irrigation  purposes,  with  58  000  fingerlings.  Unfortunately  there has been  no  follow up  ol these activities
1.8.2  Status of Fisheries in the Lower Baro-Akobo Catchment (< 800 m)
Studies on  fish  and  fisheries of the lower basin  are also  limited.  A comprehensive study  of the fish species of the region w as carried out by the Russian Academy of Science in the late 1980s as part of the Russian  study  This Study  examined  the species composition,  trophic status and parasitology  of the fish  populations but provided  no  information  on  the fisher)’per ve.  A limited amount of data is available from other reports and statistical sources.
Fish Species
Seventy tw o fish species were found in the lower basin of which the Nile perch (Lcr/ej niloticus), Nile tilapia (OreocAroffifc m/oricus) catfish (C/arias sp), Bogrus. ftar&us and Labeo species were
the most important both in ecological and commercial terms.
Fishing
Fishing  in  the region  is mainly  on  a subsistence basis both  in  the main  river channels and  many of the floodplain  lakes.  Virtually  every  family  who  lives near water fishes to  supplement their diet These people use traditional gears or hook  and  lines.  Members of the Agnuak  group,  who are traditionally  fishermen,  also  catch  fish,  mainly  in  the dry  season,  and  sell it on  the road  side near the river.  These fishermen  use gill nets in  the river and  spears in  the ponded  areas.  There are,  however,  three fishing  cooperatives al Pinindo  (at Tata) Pinkew and  Hang  which  were established  by  the Lutheran  World  Federation  Mekane Yesus project.  These cooperatives are supplied  with  gears by  the Lutheran  World  Federation  and  fish  is bought from the cooperative at Birr 1.50/kg al Pinedo. Birr 1.25/kg at Pinkew, and Birr 1,00/kg at [tang The cooperative then sells on the fish in Gambela for normal market prices.
Fishing  is highly  seasonal in  the lower Baro  Akobo  basin.  Flooding  between  June and  October prevents most fishermen operating and thus the main fish seasons is restricted to the drier periods between  October and  April.  However,  fish  are still caught and  sold  in  the wet season  but availability'. particularly in the major urban centres such as Gambela, is restricted
jVu/wfierjr of Fishermen
No  estimate of the number of fishermen  operating  in  the region  is available.  The fisheries department docs not collect data on  effort and  no  indication  of the proportion  of the population fishing cither to supplement their diet or to sell on the local market, has been made.
The main  problem is estimating  lhe number of subsistence fishermen  in  the community  who  use hook and line to supplement their diet. Their contribution to the exploitation is likely to be high, particularly  in  a region  where food  resources are limited  A further problem concerns the presence of refugees in the area who may also be fishing.
TAMS-ULG BAROAKOBO RIVER BASIN INTEGRATEDDEVH.OMENT MASTER flAN
411,         Natural Resources
Fish Production and Marketing
The majority  of fish  caught is sold  (consumed) locally,  either whole,  gutted  or filleted.  A small quantity  of dried  fish  is sold  in  the region,  mainly  during  the wet season,  or transported  to Wdega and Gore.
Problems in  accessing  the markets in  Gambela and  the fact that the Fish  Producers and Marketing  Corporation,  a parastatal set up  by  the Government to  regulate prices,  does not operate in  this region  also  means no  market information  is available on  which  to  estimate production  indirectly.  However,  an  FAO study  made an  estimate for the Gambela area.  Based on  a conservative estimate of per capita fish  consumption  (10  kg  year) total fish  production  in Gambela would be around I 240 mt year. Local fisheries officers feel this estimate may be high.
Further evaluation  of production  can  be gained  from the records of the Lutheran  World Foundation  and  the Department of Fisheries relating  to  the fishing  cooperatives (sec Table
1.8.1).  Care  must  be  taken  when  interpreting  these  data  as  their  accuracy  is  somewhat  suspect. However,  they  do  show  the  potential  for  exploitation  of  the  fisheries.  The  large  declines  in catches  reflect  the  loss  of  fishing  gear  on  the  change  of  Government.  Table  1.8  2  shows  the contribution  of  each  fish  species  to  the  total  catch  in  1994  from  the  Gilo  and  Baro  These  data illustrate  differences  in  the  species  composition  of  the  catches  between  the  rivers,  but  show  the importance of Nile perch to the fisheries.
The estimated catch is lhe lower Baro-Akobo basin is well below the potential of 5 000 mt per
annum estimated in 1995 by Aubray He gave no indication of the method of assessment but if
a figure of 5 mt per kilometre of river, equivalent to about 200 kg per ha as would be expected from tropical floodplain river channels and associated lakes (FAO. 1985) then the rivers of the
region should yield this amount of fish.
Demand' for Fish
Demand  for fish  in  the Baro-Akobo  region  is high.  In  a market survey  in  the Gambela area carried  out by  the Lake Fisheries Development Project,  availability  of fish  was the primary reason for not consuming (64% of households) or serving (61% of hotels) fish. Local households tend to eat fish throughout lhe year when available and few conform to lhe habit of eating fish during  the fasting  periods (12%) or on  Wednesday  and  Fridays only  (10%).  While most households (93%) consider the price of fish too high they also say it is lower than meat (67%). The majority (88%) of households that buy fish make a deliberate effort and go to buy either al the road  side or from lhe market in  Gambela.  Nile perch  is the preferred  fish  because of taste (72% of households. 94% of hotels) followed by catfish.
Demand for fish from outside the lower basin is high, mainly from the international community and  wealthier groups in  Addis Ababa.  The inhabitants of Addis Ababa are tending  to  eat more fish but they arc restricted by supply. In the upper Baro-Akobo basin, the demand for fish is low because (he local people have no tradition of fish eating. Under the new economic environment, FAO in 1995 estimated that the demand for fish in Ethiopia would increase to 10 000 mi per year by the year 2000 but most of this would probably be satisfied from lhe Rift Valley lakes.
T kMS-ULG rvro-axobo RIVER BASIN INTEGRATED devei ofment master planI.          Natural Resources
£conomfc and Financial Analysis
No  economic appraisal of the capture fisheries of the basin  has been  undertaken  and  is impossible because no  data are available to  support the analysis.  Similarly,  there is no  information  oi the returns earned by fishermen from their fishing activities.
1.8.3  Data Requirements and Further Studies
The information  available on  the fisheries resources of the Baro  Akobo  catchment areas is basically  biological,  comprising  species lists,  superficial species distribution,  methods of capture, fish  parasite fauna and  some genetic data.  These data arc inadequate to  make a full appraisal of the impact of future developments within  the masier plan  or how fisheries will be integrated within these proposals
To  provide the necessary  data a regional sector study  should  be carried  out which  provides information  on  the fisheries resources per sc,  levels and  methods of exploitation,  socio-economic contribution  of fisheries both  at the local and  national levels,  potential for marketing  w ithin  and outside the region,  institutional and  legal requirements The study  would  have the following elements.
•              An   assessment   of   the   artisanal   and   subsistence   fisheries   in   the   Baro   Akobo   basin covering  the  following  subjects:  when  they  fish,  how  often  they  fish  and  for  what  length of  time,  what  gears  they  use  and  their  cost  (including  ancillary  equipment  such  as  boats or   canoes),   what   species   they   catch,   how   much   of   each   species   they   catch   (usually recorded by weigh but account must be made of the contribution of juveniles to the catch, what  proportion  of  the  catch  they  sell,  consume  themselves  or  process  (eg  dry  or  smoke) and   where   they   sell   the   catch   (4   man   months   over   a   year   involving   two   fisheries observers/sc ienti sis).
•              A   study   of   the   socio-economic   status   of   the   fishing   communities   including   the importance  of  fishing  to  the  earnings  of  the  artisanal  fishermen,  the  importance  of  fish to  the  diet  of  the  fishermen,  community  health,  nutritional  status,  access  to  education, housing etc. ( Fisheries Economist for 2 months)
•             A  study  of  the  economic  and  marketing  issues,  including  the  volume  and  value  of  fish production for each part of the basin and the supply and demand for fish both locally and in  major  towns  This  will  include  estimates  of  per  caput  fish  consumption,  average  per caput  protein   consumption   and   a  review   of  distribution   channels   (Fisheries   Marketing Consultant for 3 months)
•  An  evaluation  the  status  of  the  fisheries  upstream  and  downstream  of  the  Abobo  dam  on the  Alwcro  river.  This  study  would  cover  the  same  aspects  as  the  regional  study  to provide  data  on  the  fisheries  before  and  after  the  closure  of  the  dam  (1,5  man  months over □ year involving two fisheries obsen ers/scientisls)
T AMS-ULG Baro akobo RIVER basin INTEGRATED DEVELOPMENT MASTER PLANL           Natural Resources
An  assessment of the legal and  institutional requirements to  support the sector must be made once data on  the status of the fisheries and  socio-economic aspects are available (Fisheries Economist for 2 months).
1.9      Apiculture
There is no information on apiculture in the lower basin as the Russian report did not study this sector.  The ARDCO-GEOSREV report includes a section  on  apiculture in  the Final Report Volume 11I.F Livestock, but this does not amount to a comprehensive review and analysis of the sector* despite the statement that “serial# attention w grven ro tiie deve/opmenf of apiculture".
1.9.1  Upper Basin Study
Potential for Apiculture Development
The Study identified the most important parameters for the development of apiculture as;
•     presence of suitable natural vegetation;
•     presence of native honey bees;
•     presence of fresh water;
•     weather conditions; and
•     altitude.
Over most of the Upper Basin these five conditions exist This contention is supported by a study in  1991  by  Ayelew Kassaye which  considered  temperature,  rainfall,  altitude and  vegetation  and ranked the beekeeping potential of the various parts of the country' between I and 4, where 1 was the highest potential and  4  the lowest.  Most of lhe Baro-Atobo  basin  was accorded  categories I and  2.  However,  as lhe survey  states lhe information  is not sufficient to  predict whether the current level of beekeeping could be increased.
PrejenX Situation
The ARDCO-GEOSERV Report stales that apiculture represents a significant activity  that augments the incomes of a good section of the region’s population. However. there is no data to support Illis statement The information  provided  comes from several woredas only  and  consists of lhe following mostly for a single year usually 1990:
•     production of honey and bees wax,
•     number of hives.
•     number of hives per farmer, and
•     yield per harvest.
There is no  information  on  lhe income earned  by  farmers from apiculture or of the total production  of honey  and  bees wax  from the upper basin  or its value The data in  most cases is
TAMS-ULG raro-akobo river basin integrated development MASTER FLAX
441*         Natural Resources
for a single year,  1990.  and  so  there is no  indication  on  production  trends,  although  an  increase in  the number of Tcj Bets is reported in some places. A Ministry of Agriculture surv ey in 1994 revealed  that within  the project area the number of hives owned  by  individual beekeepers ranged between 5 and 500. The average value of production was not provided.
Traditional beekeeping  in  the Baro-Akobo  area is practised  at two  levels.  The first level is to have between  1  and  10  hives hanging  in  the eaves of a house.  The second  level is forest beekeeping  where one beekeeper may  own  up  to  500  hives.  Traditional hives are made from bark,  hollowed  logs,  woven  sticks,  or baskets.  Traditional beekeeping  is a sustainable,  skilled activity using technology that has withstood the test of time (2000 * years).
ARDCO-GEOSERV provide some information  on  the species from which  honey  is produced, their time of flowering  and  color of the resulting  honey  (Volume III.F Table 3.F.46),  The report confines its attention  to  bee keeping  and  docs not comment on  whether honey  hunting  occurs in the upper basin and the extent to which it contributes to total production.
The main concern reported in the Study is the damage to bee colonies by spraying coffee In fact
the only recommendation included in Section 10 Future Directions and Recommendations is
sprary/wg nf coffees jeopardizes the lives of bees and this problem needs to be
addressed *
However,  in  the Agronomy  Report Volume II1.D ARDCO-GEOSERV states that insecticides, fungicides and  herbicides are hardly  used  by  farmers in  the area.  The situation  is therefore unclear: possibly  there is some factor causing  the death  of honeybee colonies in  coffee producing areas.  Pesticide use will not only  reduce apiculture by  killing  bee populations,  but may  also reduce the value of exported  honey  and  beeswax.  (Honey  importers analyse for pesticide residues in both commodities.)
Earlier in  the apiculture section  night time spraying  of coffee is proposed  and  the removal of the bee hives to  keep  them out of reach  of the spray.  However,  there is no  indication  that either of these two  suggestions has been  discussed  with  coffee farmers and  whether either is in  fact practical.  The Report contains no  assessment of the support provided  to  bee keepers by  MOA and data on  these services is necessary  for the formulation  of a development strategy  for apiculture as part of the master plan.
Deve/opmenr Proposals
The Study  concludes (Volume Hl.F section  9  3.6): 5r can  be staled  that significant amount of honey is readily available in  the various wore das of the study area,  and  that a  pre-requisite exists for substantially increasing  the level of honey production  in  the region  through  extensive Introduction  of modern Mvej'  However,  a modem hive is assumed  to  be a frame hive,  but no analysis is provided  to  show that the associated  increase in  costs would  be to  the financial advantage of the farmer There is a single statement to  support the proposal namely  that 22  kg of honey  was obtained  from one modem hive compared  to  only  3  kg  from the local hive (page 106).  Elsewhere on  the same page,  however,  the Report states that each  traditional hive yielded an average of 9 kg per harvest.
TAMS-ULG DAHO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PI-AN
♦51.          Natural Resources
The Report makes no  proposals or cost estimates for an  investment programme to  introduce improved  modem hives.  In  addition,  there is no  indication  of the institutional support required 
to promote such a programme eg extension, credit etc.
However, the proposal for the introduction of modem hives ignores four major factors.
•      A  study  of  different  bee  keeping  technologies  in  Zambia  shows  that  traditional  bee keeping is more economic than the use of frame hives: the situation is likely to be the same in the Baro-Akobo basin.
•      Currently  beekeepers  own  up  to  200  traditional  hives  and  accept  that  at  any  time perhaps  only  25%  will  be  occupied  by  honeybee  colonics  The  financial  investment required for frame hives makes it impossible for a subsistence farmers to own similar numbers of frame hives.
•      Beeswax  (rather  than  honey)  is  an  important  export.  Existing  reports  and  proposals favour   the   increased   honey   production   possible   with   frame   hives.   However   .the honey  efficiency  of  frame  hives  is  derived  at  the  expense  of  wax  production:  little beeswax  is  harvested  from  frame  hives  as  the  beeswax  combs  held  within  wooden frames are recycled - returned to the hive for the bees to reuse.
•     Depending on the race of bees being utilised, it might be completely impossible to manage them in frame hives (especially Apis mellifera adansonii).
To  quote from a report by  Ayelew Kassaye in  1992: Tnrr 20  000  (frame) hives have been distributed  since 1976.  Only very few are in  use*  The ARDCO-GEOSERV does not address the issue of why farmers do not adopt frame hive technology
1.10 Mining in the Baro-Akobo Basin
Since ancient limes the three mineralized belts have been known for their gold (section 1.2.5) and 
local people have exploited platinum and iron ore as well as gold Mining is an artisanal activity 
at present
LIO J Gold
Existing  mining  developments in  the basin  arc restricted  to  the semi detailed-detailed  primary gold  exploration  which  is underway  by  a Canadian-American  joint venture company  at Dul In the Metti-Nejo belts alluvial and eluvial deposits in areas 1 ike Yubdo, Kapi, Gedemso and Komto have heen  known  for their gold  and  platinum.  In  the Akobo  mineralized  belt,  placer gold  is the only  known  economically  extractable mineral and  as in  the Assosa-Bcnshangul region  local inhabitants,  usually  1  or 2  man  enterprises,  mine these deposits.  There arc no  estimates of the quantities extracted
1.10.2 Platinum
The  discover)'  of  platinum  in  the  vicinity  of  Yubdo  dates  back  to  the  early  1920s.  The  main platinum deposit is residual derived from the bedrock. The total platinum mined from 1926-1988
1' AMS- VLG BARO-AKOBO RIVFR HASfN INTEGRATED DEVELOPMENT MASTER PLAN
4ft1. Natural Resources
was 2 129.67  kg  excluding  that produced  from 1973-1983.  Pie maximum recorded  production of the metal was in  1932  which  amounted  230.75  kg.  [n  recent years production  has not exceeded 5  kg.  Major exploration  programmes were undertaken  at 5  uhdo  since 1961,  and  total platinum reserves were estimated as 12 500 kg grading 0.336 gm/ton.
1.10.3  Iron Orc
From very  early  limes,  the local inhabitants in  Ethiopia (including  the central Wellcgn  region) have smelted  iron  ore for the manufacture of agricultural implements,  utensils,  weapons,  etc.  A report submitted  to  the Ministry  of Mines,  indicated  a total reserve of 800  000  metric tons averaging 65% Fe in the basin.
1.10.4  Additional Data and Studies Required
The Baro-Akobo  basin  is a large area and  since systematic studies only  started  during  the last couple of decades there are many  gaps in  the knowledge of specific areas within  the basin.  This lack  of data is a constraint to  any  planning  of the development of the mineral resources in  the
basin. The critical areas for follow- up studies are the:
• 
lower Birbir area-Tulu Kapi, Bunaya and Ankore;
• 
Akobo Basin; and
• 
Gissen area-Kurmuk Belt.
Such  studies will lake a long  time and  will have to  be undertaken  as part of the implementation of the master plan.  In  addition,  there should  be a proper assessment of the geology  in  terms of plate tectonics to  establish  metallogcnic zoning  in  order to  help  predict areas of interest for future exploration. Extensive use should be made oT meigt-d data sets using G1S.
Future development should  lake account of the potential role of the artisanal miners and  a socio economic study  of this group  should  be undertaken  so  that a strategy  for the development of this sector can  be formulated  which  is appropriate and  takes account of the perceptions of the miners Some data for this group  could  be obtained  during  the Participatory  Rural Appraisal activities planned as part of the proposed data collection phase prior to the start of Phase II.
Although  the outline of die existing  geological maps was carried  out there should  be a detailed database covering  areas were geochemical,  geophysical and  detailed  mineral exploration  have been  carried  out,  including  a summary  of how many  stream,  soil,  trench  and  chip  samples were collected
1.11      Energy Resources Survey
An  Energy  Resources Survey  was carried  out by  ARDCO-GEOSERV in  1994.  fallowing  the methodology  adopted  in  the ECEN/CESEN nmicmnl energy  study  of 1984  A two-tiered stratified  sampling  technique was used  lo  select urban  and  rural settlements,  lhen  a random selection was made of households within settlements.
TAMS-LLG DAXO-AKClHO RIVER I1ASTN I NTT,GRATED DEVELOP MINT MASTER PLAN
-17L         Natural Resources
The survey  divided  the basin  into  three agro-ecological zones following  the ECEN/CESEN study,  but docs not identify  the proportions of the area in  each  (these were obtained  by  TAMS- ULG from the CSA and  will be mapped  later in  the GIS).  The zones and  proportions of population and area arc:
A       %       Central parts of the Ethiopian Highlands
B        %      South-western parts of the Ethiopian Highlands
C        %      Western Lowlands
The   urban   and   rural   populations   of   each   wareda   were   abstracted   from  the  CSA   Statistical Abstract for 1990 (sic), but the same estimates are to be found in the 1992 Bulletin.
The study  estimate of population  unfortunately  contains a large error in  the population  quoted in the CSA projection estimate for the town of Dcmbi Dolo (shown as 52,413 instead of 25,736), and  in  the total population  for Meko  (9,209  instead  of 20,052).  The second  error is not very’ important,  but the first one significantly  affects the overall estimate of urban  population,  raising it 16% from 167,000  to  193.000.  This will affect all calculations of the overall demands for energy,  although  the desegregated  one (rural and  urban) art not affected.  Tabic 3  was prepared following the same methodology, but extended into the lower basin, with these errors corrected.
The sample process was heavily weighted towards urban populations.
Urban
Total households Surveyed households Proportion
418,618 
Total households
39190
665 
Surveyed households
764
0,15%
Proportion 
1.95%
Whilst it is understandable that greater weighting  should  be given  to  urban  populations,  since they  are generally  larger consumers of energy,  the ration  in  this case is disproportionate,  and  the sampling level of the rural population is so low that the results may be open to question.
Three questionnaires were administered:
i           Household Energy Consumption
II           Sectoral energy consumption and supply
III         Economic and social infrastructure study
The   study   provides   some   basin   measures   of   existing   infrastructure.   Although   the   results   of Questionnaire 111 are not given, a summary is given as
• Communications: 8 post offices, 23 telephone service stations and 9 airstrips
T AMS-ULG b ar o-akobo river basin integrated development master plan
4»L         Natural Resources
•    Education: 6771 teachers of 179,922 students in 590 schools, of which ] 12 junior
secondary. 360 primary and 41 kindergartens
•    Water supply: to 22 (owns
•   Transportation: Poorly maintained.
Some idea of income distribution  is available from the results of the questionnaire,  allhough  die small size and  the approach  used  (direct question  by  questionnaire) make the findings very suspect
Per Capifu
Settlement Type Income
Persons   per household
Household income Birr/m
Minimum Median Average Maximum
Birr/yr
Rural
187
Urban < 5,000
532
Urban 5-15,000
655
Urban >15.000
846
1
9
9
3
Overall
232
5.4
5.0
5.4
5.5
5.3
59 84 1,925
75 222 3,117
171 295 8,875
188             385 103
12,169
The findings indicate a highly  non-monetarised  economy  and  a high  degree of skew in  income distribution  In  any  plan  to  develop  the cash  economy,  credit schemes will play  an  important part. These schemes will have to  deal with  lending  to  groups with  no  physical assets who  arc able to provide only unsecured collective guarantees.
The study  provides a useful insight into  the energy  balance of the region  as a whole,  but the results require further interpretation  and  some Additional figures to  allow it to  make a useful contribution  to  the Master Plan.  Table 6  summarises some of the findings.  The forecast for the year 2013 shows that even then, 99% of demand for energy will be from biomass, at a cost of 7% of the average price of electricity  of 0.37  Birr/kWh  (the cheapest supply  identified  in  the Russian Report is 0.97  Birr/kWh).  Even  then  the demand  for firewood  will only  be 55% of the sustainable supply  from existing  forests.  Provided  that the forests are protected  from destruction  by  timber extractors or irrigation developers, then the forests will be an adequate source of energy supply.
The analysis would  probably  change if disaggregated  to  a lower level of resolution,  e.g.  wereda. Fuel wood  is collected  by  women  and children w»ho spend, in rural areas,, around an hour a day on  a 2.3  km round  trip  on  the task.  However,  if local supplies of fuelwood  were to  become exhausted,  then  it would  still be more economical to  buy  it as wood  or charcoal that buy electricity. If costs rose above affordable levels, then the present low- efficiency (<10%) wood
TAMS-ULG DARO AKODO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
4*31.          Natural Resources
burners could be replaced by more efficient stoves of (he kind developed in Kenya to keep 
effective costs well below* that of electricity.
A conjectural scenario has been worked out in Table 8(a) to assess the demand that might occur if HEP were to compete effectively with biomass fuels, based on a substitution of 50% of these with HEP through an effective programme of rural electrification and aiming at consumers such as bakeries. The effective HEP cost would have to undercut biomass fuel costs by an amount sufficient to  cover the costs of installing  and  maintaining  electric cookers at a much  higher efficiency of use. Even then, there may be considerable consumer resistance due to unexpected benefits of wood burning, such as control of flying insects in the house and termites in the thatch Under this scenario,  the demand  for installed  hydropower operating  at 25% load  factor (as al present) would rise from less than 10 MW to over 4000 MW.
TAMS-ULG baro akobo RIVER basin INTEGRATED DEVELOPMENT master PLAN
SOJablcLlL Summary of the 30 Mapping Unite Idcntifi-cd in the Upper Basin by ARDCO-GEOSERV
km’
% Area
Slope %
Drainage
Dominunl soil
Mountain
Mol
Mo2
Mo3
Mo4
Mo$
Mo6
i Hilhnd
Hil
Hi2
Hi2
Hi3
Hi4
Hi5
HS
1396
2029
580
2587
1356
537
867
1286
857
3420
3558
40% Hi5
(867)
2.7
3.9
I 1
5.0
2.6
1.6
1.7
2.5
1.6
6.6
68
0.8
(17)
>50
37
50
16
3
1
35
30
10
8
4
1 (5)
rapid
rapid
Welt drained
n |i
tl H
rapid
rapid
Well drained
Rock and Scree
l ithic Cambist)!
Lithic Leptosol
Haplic Acrisol
Rhodic Nitosol
Dyslric Fluvisol
Lithic Leptosol
II               H
i*            *i
llaplic Luvisol
Rodic Nitisol
n
Poorly drained Well drained
»           hi
'
I Pied mon I
Pil
Pi2
Pi 3
Pi3
Pi4
Pi5
Eutric Vertisol
Haplic Nitisol
207
658
1605
30% Pi3
2241
937
0.4
13
3.1
0.9
43
1.8
40
25
16
16
2
♦
rapid
Well drained
11 ii
Hi |i
n II
Well drained
Lithic Leptosol Chromic Cambisol
Haplic Nitisol
Chromic Luvisol
Haplic Nitisol
Eutric Gleysol
Plateau
Pul
Pu2
Pu3
PlM
Pu4
Pu5
Pu6
Pu7
Pu8
236
1123
729
3997
30%Pu4
6727
4055
191
311
0.4
2.3
1.4
7.7
2.4
12.9
7.8
0.4
0.7
>50
48
30
20
18
10
4
2
1
Rapid
Well drained
»             |i
11 ii
I*            R
Ik 11
R II
Rock and Scree
Lithic Leptosol
Haplic Airisol
Hapic Nitisol
Hapic Lixisol
Hapic Nitisol
ii           n
Poorly Drained
if    ■ •
Eutric Gley so I
Plain
Pll
PI2
PI3
1 PI4
1 Pl5
738
(1365)
627
915
286
46431
1.4
0.6)
1.2
1.76
(Li
94 06
25
9
3
2
1
Well drained
*»           ii
Poorly drained Well drained
n             1r
Dystric Cambisol Dystnc Cambisol
Eutric Vertisol
Dystric Cambisol Eutric Fluvisols1 able I 1 2 Seal Type Land Form and Major Profile Characteristics
Soil Type
% Area
Land Form
Major Profile Characteristics
Lithosols 
2
Hills, ridge*
Shallow, acid, coarse and gravelly Low OM
content rapid drainage little vegdalion cover
Glevsgls Alluvial meadow
12
Internal
Deep, generally fine and medium textures,
basins of flood
(luninialcd) HighOM content, slightly acid to
plain and ox-bows
neutral Base nch very poorly drained Perennial
hvdrophvtic vegetation subject to seasonal flooding
Llum?h
22
Lower plains
Deep, generally fine but also coarse, laminated
Alluvial
soils
River valleys
high OM content Base nch slightly acid to
Relict  meanders,
neutral, occasionally slightly sodic Very poor
ox-bows and deltas
tn moderate drainage Hydrophytic vegealaion
Annual flooding
Vertisols 
23
Older flood
Deep fine textured cracking days, “Gilgu"
Black
compact
soils
plains
surface features Calcareous Low OM content
and low niitncnt levels some gley subsoils,
subject Io seasonal water logging Savanna and
hydrophytic grasses
Humic PlnnnsQk 4 Dark
meadow soils
Bleached meadow soils
Minor river
valley s of the fool slope zone, between the hills and plamsornc
High plain* in scattered patches slightly stoping to sloping relief
Variable, generally deep fine textured slightly 
acid Io tkuitbI. Compact B+C horizons, high
OM content Base nch seasonally waterlogged
good lo moderate drainage sodium present in profiles nt B'C horizons Hydrophvuc grasses 
Rapid loss of nutrients once cultivated
Coarse topsoils arc loamy and or loamy clays, generally deep, acid and low m OM and
nutrients. Rapid to good drainage, some finer lexrured sites have moderate sodium content in sub
soil Open savanna with annual grasses.
Cambisms
Brown
Siallatic 
soils
23
High
Usually brown, acid, low OM, lupaoth over
plain widespread on
less acid more clayey sub-soils with Fe*Mn
both flat and
cones. Low nutrient levels Under forest higher OM
undulating surfaces
Most sites reflect underlying nxk type
Chmmosols Fcnugenous
soils
8
High Flam
Reddish brown, generally deep, slightly acid
commonly- over a
with laterite in lower profile, low to moderate
rolling, disected
OM and moderate nutrient, but low in |> Open
surface
fcrest and annual grass Frequently cultivated
Fcnasols Ferraltilic
soils
3
High Plain,
Well structured top soil Deep. OM and
broken dissected
nutrient nch aver laterite. Neutral Io avid nt
surface and
depth, good water holding capacity Mixed
on ridges 
forest, little cultivation.
up to 600 ITiasl
Ntliwls
2
Plateau,
General!) deep, 'rd acid clsyey loams wuiJi a
Red lalcrizcd
rolling and ndge,
OM rich top sod Non calcareous low in P+K
soils
same steep slopes 
moderate N. Base rich over Basalt, well
and valleys. Erosion
drained Widely cultivated Little forest left
common
now cultivation savannaTabic 1.3.3: Land Classification Assessment, UppeT Basin.
SCR
class (1)
Slope
limit (%)
Land class
Limiting
factor
Area
(km )
3
I
II
III
IV
VI
vn
VII]
<3
3-8
8-15
15-30
30-50
>50
>50 + rock outcrops >85% of
surface
I
Ik
Ilk
Illw
IV we
Vis
Vie
Vile
VUIds
VUIde
none
erosion
erosion
drainage waterlogging 
. erosion
stoniness
erosion
erosion
rock 
outcrop, shallow soil
shallow   soil, erosion
1,240
3.768
13,394
31
17,739
3,316
6,142
2,647
2,500
1,292
s V, permanent swamp/waterlo. 
the Upper Basin.
, has not been
jd inTabic 1.3.4: Land Suitable for IrrigEitinn in the Upper Basin
Map Symbol
Mapping Unit
Area (km*)
Gross 1     Possible
MO6
Hi4
Hi5
Pi4
Pi 5
Pi6
P116
Pu7
Pu8
PL3
Pl .4
PLS
inter   mountain 
valleys
undulating   and gentle side slopes undulating   side slopcs/low terraces gentle  inclined  foot slopes
flat   undulating 
terraces
depressions   on 
plateaux
undulating plains Flat flood plains depressions   with poor drainage undulating   plain   of middle terrace
flat  plain  of  low
terraces
flat floodplain
573
3558
867
2242
937
145
4056
192
311
6280
915
287
516               I 2846
781
2018
937
145
3418
192
3!
5338
915
287
20,363
17,424Table 1.3.5: Land Suitability Classification Specifications for Irrigation.
Land
Characteristics
Class 1
Class 2
Class 3
Class 6 
non-irrigable
Soil Texture:
sandy loam to 
permeable cloy
loamy sand to 
permeable clay
loamy sand to 
vertisol
sand & gravel
min. depth to 
rock
ISO
100
75
<75
min. depth to sand;gravel
100
75
50
<50
Slope (%)
0-2
2-4
4-8
>8
Levelling requirement
low to 
moderate
moderate
moderate to 
heavy
very heavy
Clearing
low to 
moderate
moderate
moderate to 
heavy
very heavy
1 Drainage requirements
none
moderate
Source: Survey
Analysis          e
:r Haro-.
moderate
1
High
vol lie. Im gallon
5Table 1.3.6: Location  of Potential Irrigation 
Sites and Gross Irrigable Area.
Irrigation site
Gross Irrigable
Area (ha)
Koji
Sako Guda
Beko
Kiln
Lafo Kotu
Baro
Birbir
Fani
AJwero
Guy
Godarc
Achani
Awaya
Bebeka
Guracha
Gumero
Akobo 1
Akubo 11
Total
288
3010
544
4994
19
2000
8000
282
240
110
651
312
137
24
144
4467
4528
28,521
58,671
Source:  Survey  and  Analysis  of  the  1  pper Baro-Akobo Basin, vol lie. Irrigation, 1995.Table 1.3.7: Land  Suitability  Classification  for Potential
Irrigation sites.
Irrigation site
Irrigation Suitability Classification
SI
S2
S3
Koji
Sako Guda
Beko
Kilu
Lafo Kotu
Bare
Birbir
Fani
Alwero
Guy
Godare
Achani
A way a
Bebeka
Guracha
1 Gumaro
A kobo I
Akobo II
Total
L
288
0
0
0
0
2000
0
0
0
0
0
312
0
0
90
0
0
0
2690
-
1810
544
2994
19
-
8000
282
140
70
400
-
137
424
54
2900
2728
17,121
37.623
-
1200
-
2000
-
-
-
*
100
40
251
-
-
-
-
1567
1800
11,400
18,358
Source: Survey and    Analysis of tl he L pper Ban a-Akobo Basin vol lie. Irrigation,Table 1.3,8: Major zones of the Gambcla Plain.
| Zone
Physical characteristics
Major land use
Area (km )
1
Western
Northern
Central and eastern
flooded from
2-5 months
foolslopes of mountains and Barn floodplain foot slopes of mountains
Baro-Gilo interfluve
east and southeast areas steeply dissected
western area gently undulating
permanent pasture, 
wildlife habitat
extensive
arable
forest,    scattered cultivation
savanna woodland, arable cultivation
8,890
3,600
1.870
2,760
South-eastern
1.840
2,550
Source: Baro-Ako
bo Basin Master P
an Study ot Water and i and Resour
of  the  Gambda  Plain,  Draft  Final  Report,  Volume  V  Annex  4.  soils  and  their possible uses.Table 1.3.9: Infiltration and Hydraulic Conductivity by
Suitability Mapping Unit________________________
Mapping
Unit
Infiltration
M/day
cm/hr
Hydraulic
Conductivity 
(M day)
la
lb
Ic
ria
lib
lie
Illa
[lib
IVa
IVb
IVc
Va
Vb
1.58
1.4
0.72
1.15
7.2
1.15
1.3
1.3
1.7
0.07
0.07
0.04
0.04
6.6
5.8
3.0
4.8
30.0
4.8
5.4
5.4
7.1
0.3
0.3
0.2
1.7
0.57
0.3
0.43
0.29
7.5
0.04
0.14
0.14
0.86
n/d(1)
n/d
004
0.04
Source: com    >iled from data in      Baro-Akobo Ba
in Master Plan 
Study of Water and Land Resources of the Gam be la Plain, 
Draft Final Report, Volume V Annex 4, soils and their possible uses 1990.
(1). Not determined.Table 1,8.1. Catch rates of the fisheries cooperatives between 1988 and 1994
1 1            Pinudo/Tata
Pinkcw/Gilo
llang/Baro
Total
1988
1990
1991
1993
21994
Number of members Distance from
|| Gambela (km)
7901
1140
500
8205
N/A
30
105
7320
1105
1944
615
6065
2)
18
10.227
17,170
10,371
30
2920
28
36
25,449 19,415 12.815
8820
8985
79
Table I 8.2: Species composition of the catches (weights) in 
the Gilo and Baro rivers 1994, based on data from the fishing 
cooperatives
Species name
Local
Agnuak
Gilo
Raro
Distichodus ndofieus Latex nUoticus
Mr/roephj jbshWH
('/arias
Gymnatrha
Het erot is nd otic us Cilharinus
Bagrus
Barb us
Tilapia
Syndonlis
Pofypterux bichir
Pure
Gur
Weirj
Ague 1 la
Wect
Ullaw&k
Ujaka
Udwara
Ukura
J ary’
Urwedo
Ukok
Udwella
228
944
37
4
318
139
IOS
224
l
84
164
1209
33
190
132
207
27
209
65
70
13
50
402SECTION 2
AGRICULTURE2.          AGRICULTURE
2.1       Introduction
In  this section  agriculture is split into  crop  production  and  livestock.  This treatment coincides with  that used  during  the iwo  major previous studies of the Baro-Akobo  basin,  although  both crop  production  and  livestock  are elements of the various farming  systems found  in  the upper and lower basins.  In  a country  where livestock  is an  integral part of many  farming  systems,  the absence of an  approach  in  the previous reports which  considers farming  systems as a whole is unfortunate
2.2       Crop Production
Combining  the data from the two  studies,  albeit for a period  5-6  years apart,  the total cultivated crop  area in  the Baro-Akobo  basin  is al least 554  (XX) ha (Table 2.1.1),  with  Illubabor and Wollega together accounting  for three quarters oT  the total,  The striking  observation  is that over 95% of all cultivated  land  fails in  the upper basin.  The assumption  is that most of this is in  the highland zone and reflects the high density of arable farmers on the land above about 2 000 m
2.2.1 Upper Basin
The ARIX.O-GEOSERV Study  is technically  comprehensive with  regard  to  a description  of the geology  and  soils (see section  I above),  but is light on  agricultural statistics,  analysis and recommendations,  Volume III.D Agronomy  includes just over one page of general recommendations,  in  a 66  page report.  The inclusion  or recommendations assumes that ARDCO- GEOSERV’s TOR included  such  a requirement.  The report does not say  how the recommendations would  be implemented  or give an  estimate of the costs involved.  This review concentrates on Volume III. D of the ARDUO-GEOSERV report.
Method of Approach
ARDCO-GEOSERV used  two  surveys to  provide the bulk  of their data: one a survey  of 353  farmers in  19  woredas and  the second  a survey  of MOA woreda offices.  In  addition. ARDCO-GEOSERV used  information  from discussions with  government staff,  NGOs, private sector firms and  from the related  socio-economic survey.  The farmer survey reported  data by  woreda und  provided  averages by  region  and  for the catchment as a whole.  While the answers to  some questions varied  considerably  betv/een  woredas and zones,  the report does not generally  elaborate on  these variations which  is unfortunate as the upper basin  is not a homogeneous area.  This review only  comments on  the catchment averages and  does not attempt to  re-analysc the ARDCO-GEOSERV data.  ARDCO- GEOSERV presented  official data on  crop  areas,  yields and  production  by  zone and commented on tire variations.
Crops and Cropping Patterns
The total cultivated  crop  area in  the upper basin  is 536  000  ha (Iable 2.El) with Illubabor and  Wollega together accounting  for three quarters of this total.  Cereals are by far the most important crop,  74% of the total cropped  urea followed  by  coffee |7% Pulses 6%, oil seeds 2% and spices and vegetables were the other crops grown
TA MS-ULG raro-anobo river basin Integra ted development master pi an
2-11. AGRICULTURE
In the farmer survey nearly all farmers reported mixed farming with 3 rotation of crops. Table 2.1.2 shows the cropping patterns in the main zones of the upper basin. In the two largest zones. Illubabor and Wollega, and in Kaffa, maize is the most popular cereal hut in Assosa, sorghum, as in the lower basin, displaces maize, perhaps on account oT its greater reliability  under drought conditions.  Teff,  the source of the staple food  of the highland population, is particularly unportant in Illubabor with more than a quarter of the crop  area,  but much  less so  in  other zones.  Other small grains are present but of insignificant importance in the upper basin.
The above figures exclude tea grown on the Gumcro estate, 300 ha. It also seems likely that the area of coffee excludes the 12 500 ha on the Bebeka and Tcpi estates, since the area of estate grown tea has been excluded.
Crop Yields and Production
Table 2.1.3 gives the average yields reported by the woreda offices ofMOA for 1992/93. Yield levels mostly correspond with data in the latest report of the annual sample survey of farms in Wollega published by the Agricultural Statistics Department, where there is a marked difference the values of the ARDCO-GEOSERV report are preferred. Fot the upper basin as a whole, except for the maize yield which is rather high, 1.280 kg/ha, most yields  conform  to  the  levels  expected  from  mainly  subsistence  production.  Yields  of other  major  crops  were  sorghum  1.070  kg/ha,  teff  630  kg/ha  and  coffee  540  kg/ha Average yields in Lllubabor were higher than tn the other regions, although Assosa had the highest sorghum and wheat yields and Wollega the highest coffee yield. The yield reported from the Bcbeka coffee estate is 400 kg/ha, less than that achieved by the small holders, while the Tepi Coffee Plantation Project has a yield which ranges from 600 to 800 kg/ha. The average yield of made tea from Gumero is 2 000 kg/ha,
Table 2.1,4 gives figures for total production by multiplying the area and yield 
information respectively from Tables 2.1.1 and 2.1.3. It should be noted that the
ARDCO-GEOSERV figures for production in Table 3.D.3 use the average basin yields rather than the zonal yields. The revised figures in Table 2.1.4 show the importance of
cereals and especially maize The whole of the upper basin produced over 234 000 mt of maize of which Illubabor produced 45% and Wollega 31%. Coffee production, 
48 000 mt is almost exclusively found in these two zones As noted above this total is
likely to exclude production from the estates which w-as 1 563 mt in 1993 down from
2 770 mt in 1992. The corresponding figure in 1991 was 1 471 mt.
Number o f Farmers. Farm Size. Family Size and Availability'of Labour
Land holdings are reported to average 1.2 ha, although the farm survey did not collect this information  from the farmers interviewed.  Die report does not indicate the number of farming  households The average farm family  had  7.8  members of which  4,5  were more than 13 years old and regarded as economically active and 2.8 worked full time and the remainder farming  part time.  Only  20% of the farms reported  using  hired  labour. Master planning  requires information  on  the number of farmers and  farm size for the different agro-ecolcgical zones.
TAMS-ULG baro-akobo Rf\er basin Integrattii development master plan2.           AGRICULTURE
Crop Husbandry and AgricuifuraJ fnpufs
Land  preparation  is normally  undertaken  using  draught animals,  usually  oxen  and  up  to six  ploughings are necessary  for leff while two  or three arc sufficient for other crops Such  operations expose the soil to  erosion  requiring  appropriate soil conservation measures on  sloping  land.  On  average farmers own,  a pair of oxen,  75% of fanners use their own  oxen  as a source of farm power,  while 16% rented  oxen  and  6% used  hand tools.  No  farmer used  a tractor.  The socio-economic survey  provides some information on oxpower used in terms of hours per ha, but does not specify the crops. Therefore, the information can not be used when calculating crop gross margins (see below).
ARDCO-GEOSERV noted  thai improv/d  varieties of the major crops exist but data from the woreda offices indicated  that only  1% of the cropped  area was cultivated  using  seed of improved  varieties.  Most farmers use their own  seeds which  are almost all local provenances.  A quarter of farmers exchange seeds with  neighbours and  nearly  the same number buy  them back  from the grain  merchants prior to  planting.  This fact seems to indicate a storage and.  or a cash  problem,  although  ARDCO-GEOSERV made no  such comment. Average seed rates reported were close io those recommended by MOA. Three in four farmers staled there was either no source of seeds of improved varieties or that the supply  was inadequate.  One in  three farmers had  insufficient cash  to  buy  them anyway and  18% were unaware of their existence (or one supposes) the possible benefit.  While ARDCO-GEOSERV did  not comment upon  these figures,  the answers reflect badly  on the input supply, credit and extension services.
According to the surv ey, the use of fertilizer is insignificant in the upper basin at present. In  the 1992/93  season,  only  294  mt of fertiliser,  mainly  diammonium phosphate (DAP, 20%N:22%P),  was used: this is an  average of 0  22  kg  of nutrients per ha of crops and compares with  more than  100  kg/ha in  developed  countries,  nearly  500  times more.  The farmer survey  suggested  lhe main  reasons for not using  fertiliser was a lack  of cash  to buy  even  if it were available (46%) and  secondly  that none is available (42%).  Late supply  was a minor reason,  as was lhe lack  of knowledge about how to  use it (15% of fanners in both cases).Totals exceed 100 as more than one answer was sometimes given.
Almost no pesticides arc currently used in the basin even though only 5% of farmers said 
they had no pest-problems. Nearly half of fanners stated that they arc not locally 
available and almosi as many said they could not afford to buy them anyway. The
proportion admitting to lack of knowledge was again 15%. Control of weeds is most
commonly by hand and fanners on average wved twice. with the greatest atteniion being 
paid to maize and sorghum.
Harvesting  of crops is exclusively  by  hand  Threshing  is predominantly  manual with  35% of fanners using  animals to  assist and  only  8% using  only  animals.  Threshing  machines are not commonly  available.  Post-harvest losses by  rodents,  insects and  rotting  in  the domestic grain  stores,  which  are most commonly  used,  are believed  to  be very considerable.  No  information  was obtained  concerning  lhe labour requirements of the various crops either in total, by gender or for (he various operations.
TAMS-ULG BARO AKOBO RIVER BASIS INTEGRATED DE STI-OPMENT MASTER PLAN
2-32.          AGRICULTURE
The socio-economic survey  provided  information  on  labour hours per ha lor ploughing, weeding  and  harvesting  but as these figures are not for specific crops they  can  not he used  to  establish  crop  labour profiles or to  calculate the crop  gross margins per labour day,  The latter is a key  measure for comparing  the profitability  of individual crops and cropping  systems under smallholder production  which  would  form part of a master planning  exercise In  addition,  the wide ranges exhibited  by  the figures eg  75  to  40  061 man hours per ha for weeding must cast doubt on the reliability of some of the figures.
/tale of Women
The field  survey  showed  that women  participate in  almost all farming  activities.  Weeding is the most important task  being  undertaken  by  98% of women,  follow ed  by  harvesting. 84%.  and  threshing  76%.  For planting  and  cultivation  the percentages are 47% and  42% respectively.  While men  normally  do  the ploughing  on  10% of farms women  arc involved in this operation
Zengt/r of Growing Season, Cropping Patterns and Smallholder Farming Systems
rhe ARDCO-GEOSERV report indicates a 'length  of growing  season” which  divides the basin  into  three zones: the floodplain  with  4-7  months,  the mid-slope with  7-9  months and  the highland  with  9-12  months,  TAMS-ULG have started  to  analyse the climatological data from 25  meteorological stations,  within  and  close to  the project arcEi boundary  from the FAO Agro-meteorological Unit data base,  and  an  additional five from the ARDCO-GEOSERV report to  refine (he assessment of the length  of growing  season using alternative methodologies.
In  the ARDCO-GEOSERV survey,  most farmers,  nearly  90%,  practised  mixed-farming and  followed  a crop  rotation.  Only  15% of farmers interviewed  said  they  practised shifting  cultivation,  and  only  52% were able to  fallow their land  for 1  to  2  years.  In consequence,  cropping  intensity  in  the highland  zones is about 70% and  only  in  Assosa is it low at 22%.  There were no  details of the rotations followed,  which  a comparison  of alternative smallholder farming  systems during  the master planning  phase will require, 
-rv 
Ihe report does not comment on  the implications of this situation.  Although  potentially a sustainable system with  low yields per ha,  it is in  practice one in  which  yields almost invariably  decline.  National annual growth  in  agricultural production  in  the last decade has averaged  0.9% and  while TAMS-ULG does not have the data is likely  that the rate of increase in  the cultivated  area during  the same period  has exceeded  0  9% indicating a fall in  productivity  of land  overall.  There arc undoubtedly  several factors operating,  not least of which will be the:
•    degradation of cropland by erosion;
•       use   ot  ever  more   fragile   and   marginal   land   as   population   pressure   increases,   an annual growth rate of nearly 3%; and
•    shortening ot the fallow cycle leading inevitably to continuous cultivation.
.       .
TAMS-ULG BARO-AKOBo RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLANAGRICULTURE
Cash, Gross Margins and Farm Incomes
The major source or cash  reported by farmers was lhe sale of grain, 75%, followed by sales of animals and  animal products,  51%.  and  sales of coffee.  36%.  In  general,  the survey  found  that crop  prices were 50% higher at planting  time than  at harvest.  In  the upper basin  off-farm employment is rare and  only  11% of farmers referred  to  off-faim employment as a source of cash.  Credit as a source of cash, although not income, was mentioned by 6% and lhe lack of credit facilities was noted.
ARDCO-GEOSERV did not calculate the gross margins from lhe various crops. A gross margin  analysis from which  it would  have been  possible to  calculate the net farm incomes from the various smallholder cropping  systems would  have provided  data for the land evaluation exercise which will form part of lhe master planning in Phase II. This analysis provides a financial and  economic comparison  of alternative uses of areas of land with similar soils, slopes, climate etc,
Coffee and Tea Estates
Yields  of  the  coffee  estates  are  far  below  their  potential  and  ARDCO-GEOSERV identified lhe problems at Bebcka as;
•     shortage and uneven distribution of rainfall with the dry season repeatedly coinciding with the flowering period;
•    unsuitable soils;
•     an acute shortage of labour especially during peak periods;
•     diseases (unspecified) which are difficult to control effectively because of the
large size of the plantation; and
• wild anima] damage.
The Tepi Project has an acuie labour shortage especially during harvest time and as a result 20-30% of the productive area remains unharvested.  ARDCO-GEOSERV make no proposals for tackling these management issues hut says that under the new policy each estate is to have a separate Board of Directors to direct and supervise operations. No assessment of the performance of the Gumero tea estate was made
Agricultural Services
ARDCO-GEOSERV assessed  the services provided  by  government and  other organisations, but the report provides a limited amount of data. The report comments that the extension service’s links with the farmer have deteriorated. There is one extension agent for 811 fanners and 46% of lhe farmers surveyed reported that they received no visits from their development agent during the 1992/93 crop season. Of the remainder
T AMS-ULG u aro-akobo wvfr basin integrated dev elopment master elan
2-52.         AGRICULTURE
22% received  more than  ten  visits which  was considered  adequate by  ARDCO- GEOSERV, although in many other countries a fewer number of visits would have been considered sufficient The woreda offices of MOA indicated that a shortage of trained development agents, lack of funds and vehicles prevented them from achieving a belter extension coverage.
The Agricultural and Industrial Development Bank has only one main branch at Metu and two sub-branches at Gimbi and Dcmbidolo in the upper basin. The Commercial Bank of Ethiopia has nine branch  offices which  provide credit to  a few selecicd  farmer cooperatives for the purchase of inputs and  advance payments for coffee collection, processing  and  marketing,  fhe survey  found  that 4% of fanners interviewed  obtained credit from the Agricultural and  Industrial Development Bank  and  7% from the Commercial Bank  of Ethiopia.  ARDCO-GEOSERV did  not report the terms and conditions of the loans or repayment rates.  The most important sources of credit are relatives and friends according to 53% of respondents No mention was made as to what extension messages the development agents were giving to the farmers and how relevant they were.
Two NGOs are mentioned, Menschen fur Menschcn and Rural Integrated Basic Services but,  while a little data is provided  on  their activities,  there is no  assessment of their performance.
GwfrwTr/r /□ Increased Agricultural Production
Constraints to  agricultural development were classified  as follows: 78% of the farmers reported  problems of weeds,  pests,  diseases,  wild  animals etc.,  while only  28% experienced problems related to shortage of rain, floods, poor drainage, erratic rainfall, soil salinity  and  erosion.  According  to  .ARDCO-GEOSERV most of these problems could be managed by the farmers if they were provided with suitable guidance. Economic problems were also prominent. About 80% of farmers reported that a lack of money was a constraint while 58% experienced  transport and  marketing  problems in  addition  to  a shortage of cash. The conclusion from these data are dial in the upper basin factors such as rainfall, drainage and erosion arc not perceived as major problems where as weeds, pests, diseases, wild animals and a shortage of cash arc serious problems.
Potential/or Irrigated and Rainfed Agriculture
ARDCO-GEOSERV reported that there were 18 possible irrigation sites (see section 1 above).  However,  the study  noted  that each  site would  require hirther study  before a decision lo invest could be made. The report also noted that there were opportunities to increase household  food  security  by  diverting  streams to  irrigate home gardens and developing micro-dams andLor ponds foT irrigating vegetables crops at the village level. Other development opportunities identified  were the production  of spices eg  ginger, tumeric. black pepper, cinnamon, cardamon and fruit eg citrus and banana but the report contains no detailed proposals,
TAMS’ULG baro-akobo river basin INTEGRATED DEVELOPMENT MASTER plan2. AGRICULTURE
Cor/?, Gross Afargins and Farm Incomes
The major source of cash  reported  by  farmers was the sale of grain,  75%,  followed  by sales of animals and  animal products.  51%.  and  sales of coffee,  36%.  In  general,  the surv  ey  found  that crop  prices were 50% higher at planting  time than  at harvest In  the upper basin  off-farm employment is rare and  only  11% of farmers referred  to  off-farm employment as a source of cash.  Credit as a source of cash,  although  not income,  was mentioned by 6% and the lack of credit facilities was noted.
ARDCO-GEOSERV did  not calculate the gross margins from the various crops.  A gross margin  analysis from which  it would  have been  possible to  calculate the net farm incomes from the various smallholder cropping  systems would  have provided  data for the land evaluation exercise which will form part of the master planning in Phase IT This analysis provides a financial and  economic comparison  of alternative uses of areas of land with similar soils, slopes, climate etc.
Coffee and Tea Estates
Yields   of   the   coffee   estales   are   far   below   their   potential   and   ARDCO-GEOSERV identified the problems at Bebcka as:
•     shortage and uneven distribution of rainfall with the dry season repeatedly 
coinciding with the flowering period;
•     unsuitable soils;
•     an acute shortage of labour especially during peak periods;
•     diseases (unspecified) which are difficult to control effectively because of the
large size of the plantation; and
• wild animal damage.
The Teoi Project has an  acute labour shortage especially  during  harvest time and  as a result ”>0-30% of the productive area remains unharvested.  ARDCO-GEOSERV make no  proposals for tackling  these management issues but says that under the new policy  each  estate is to  have a separate Board  of Directors to  direct and  supervise operations. No assessment of the performance of the Gumcro tea estate was made.
Agricultural Services
4RDCO-GEOSERV assessed the services provided by government and other organisations but the report provides a limited amount of data. The report comments that
'he extension service’s links with the fanner have deteriorated. There is one extension 1  for 811 fanners and 46% of lhe farmers surveyed reported that they received no 
from their development agent during the 1992/93 crop season. Of the remainder
TAMS-ULG haro-akojio river basis- integrated de vi lopment master planI
2. AGRICULTURE
I ---------------------------------------------------
J
f NJ
22% received more than ten visits which was considered adequate by ARDCO- GEOSERV. although in many other countries a fewer number of visits would have been considered sufficient. I"he woreda offices of MOA indicated that a shortage of trained development agents, lack of funds and vehicles prevented them from achieving a better extension coverage.
J]
J]
3
J
J
J
J
1
3
J
jl J
J
?!
1
The Agricultural and  Industrial Development Bank  has only  one main  branch  at Metu and  two  sub-branches at Gimbi and  Dcmbidolo  in  the upper basin.  The Commercial Bank of Ethiopia has nine branch  offices which  provide credit to  a few selecled  farmer cooperatives for the purchase of inputs and  advance payments for coffee collection, processing  and  marketing.  The survey  found  that 4% of fanners interviewed  obtained credit from (he Agricultural and  Industrial Development Bank  and  7% from the Commercial Bank  of Ethiopia.  ARDCO-GEOSERV did  not report the terms and conditions of the loans or repayment rates.  The most important sources of credit are relatives and  friends according  to  53% of respondents.  No  mention  was made as to  what extension  messages the development agents were giving  to  the farmers and  how relevant they were.
Two  NGOs are mentioned,  Menscbcn  fur Menschen  and  Rural Inlegratcd  Basic Services hut,  while a little data is provided  on  their activities,  there is no  assessment of their performance.
Cornrroinw to Increased Agricultural Producnon
Constraints to  agricultural development were classified  as follows: 78% of the farmers reported  problems of weeds,  pests,  diseases,  wild  animals etc.,  while only  28% experienced  problems related  to  shortage of rain.  floods,  poor drainage,  erratic rainfall, soil salinity  and  erosion.  According  to  ARDCO-GEOSERV most of these problems could  be managed by the farmers if they were provided with suitable guidance. Economic problems were also  prominent About 80% of farmers reported  that a lack  of money  was a constraint while 58% experienced  transport and  marketing  problems in  addition  to  a shortage of cash.  The conclusion  from these data arc that in  the upper basin  factors such as rainfall,  drainage and  erosion  arc not perceived  as major problems where as weeds, pests, diseases, wild animals and a shortage of cash are serious problems.
Potential for Irrigated and Rainfed Agriculture
ARDCO-GEOSERV reported  that there were 18  possible irrigation  sites (see section  1 above).  However,  the study  noted  that each  site would  require further study  before a decision  to  invest could  be made.  The report also  noted  that there were opportunities in increase household  food  security  by  diverting  streams to  irrigate home gardens and developing  micro-dams and/or ponds for irrigating  vegetables crops at the village level. Olhcr development opportunities identified  were the production  of spices eg  ginger, tumeric,  black  pepper,  cinnamon,  cardamon  and  fruit eg  citrus and  banana but the report contains no detailed proposals.
TA MS-U LG barclakoho river basin Integra ted development masterplan
22.           AGRICULTURE
However,  the farmer survey  did  not make clear the views of the farmers concerning these developments.  Participation  of fanners at all stages of development from identification  through  to  implementation.,  monitoring  and  evaluation  is essential if the proposed  activities are to  succeed  and  to  have a chance of being  sustainable.  Planning development without the participation  of the people is also  contrary  to  the policy  of the current Government:
" The participation  of the people,  specially (he rural population  is decisive for fast and sustained  development All (he economic policies of the Government should  enhance and encourage this participation "
Recommendations
According  to  ARDCO-GEOSERV  low  levels  of  production  are  due  to  a  number  of factors including
•     inadequate and/or low grade extension work;
•     a lack ol agricultural inputs and credit;
•     poor fanning practices;
•     poor or a incomplete absence of physical and social infrastructure: and
•     an absence of appropriate technical packages.
To  exploit  the  agricultural  potential  of  the  upper  basin  the  report  identified  that  the following were required:
•      a  master  plan  for  the  catchment  which  would  include  the  identification  of  viable development projects which should be able to attract local and foreign investors by
•      providing  the  necessary  infrastructure  eg  roads,  irrigation  facilities,  extension,  credit etc.;
•      the  provision  of  agricultural  inputs,  extension,  credit  and  improved  marketing  to encourage farmers to increase production as a priority;
•      immediate  steps  Io  stop  deforestation  and  to  rehabilitate  conservation  and  protection works in the endangered areas:
•     an intensification of agricultural research;
•       introduction   of   strict   policies   covering   conservation   and   exploitation   of   natural resources including soils, forests, water and wildlife; and
•        development   of   small   and   medium   scale   agro-industrics   to   process   surplus agricultural   produce   in   a   way   that   would   encourage   farmers   to   increase   their agricultural production.
r 4MS*ULG HARO AKOBO WVER BASIN [Nil GRATED DEVE1OPMFNT MASTER PLAN2.          AGRICULTURE
The above is a set of non  controversial recommendations.  However,  the report does not contain  proposals for their implementation  or how those which  might conflict ie conservation  of natural resources (eg  forests) and  agricultural development might be reconciled.  Nor docs the report include the views of (he farmers concerning  the development possibilities.  Experience has shown  that development proposals which  do not accord with the priorities of (he fanners do not succeed and arc unsustainable.
2.2.2 Lower Basin
The Russian  study  provides a comprehensive and  detailed  summary'  of crop  production  in  the lower basin  as it was in  1987.  Although  technically  thorough  the Study  is now oul-of-datc with respect to  development strategy'  and  statistics.  Il considers in  considerable detail two  agricultural development alternatives based  on  the establishment of stale farms a “minimum" which  has an irrigated  farming  and  a rainfed  farming  option  and  a "maximum" which  would  involve the development of sugar cane cultivation and processing.
Farming Households, Methods of Praducrion and Culrivafed Jrea
Al the end  of 1986  (here were 142  peasant associations,  of which  36  comprised  fanners resettled  by  the government from outside the lower basin,  and  a recently  established  state farm.  The /Xnuaks and  the Nuers formed  the bulk  of the local population,  48% and  38% respectively.  The Anuaks arc sedentary  agriculturalists who  farm along  the river banks m the central part of the Gambela plain.  They  keep  small numbers of poultry  and  also fish  and  hunt.  The Nuers are semi-nomadic herdsmen  who  also  cultivate some small areas of subsistence crops.  The Nuers inhabit the western  part of the plain.  Whereas members of the local peasant's associations cultivated  their plots individually  members of the resettlement associations practised  collective fanning.  Overall 71% of the population  were reportedly  engaged  in  crop  production,  although  no  data was available as to the employment of the rest.
By  the end  of 1987  only  26  600  ha or 1% of the total land  area was cultivated  In  1986 the area of land  cultivated  by  the peasants associations totalled  24  900  ha involving 23  400  families,  an  average of 1.06  ha per family.  All tillage operations w ere undertaken 
r
manually,  except for land  preparation  which  was undertaken  using  tractors in  the
resettlement areas Farmers lacked  even  the most primitive farm implements and  there
were  “no  farm  service  facilities ,  This
1
last  statement  being  taken  to  imply  that  farmers
had no access to fertilizers or other inputs. Animal draught power was not used.
The state farm concentrated  on  cotton  production,  all rainfed,  from 1987  after experimenting  with  a whole range of crops during  the preceding  years In  1988  the area under cotton was 3 335 ha compared to 1 800 ha in 1987 and only 100 ha in 1986. Except for picking  all operations were mechanised.  This area is not included  in  the 1987  crop area statistics.
Crop Areas’ and Fields
TAMS-ULG baro^akobc river basin integrated devei            OPMKNT MASTER. PLANAGRICULTURE
lite data for crop areas covered the period 1983 io 1987.The figures show that there was a rapid increase in the total area of crops from almost 6 800 ha in 1983 to 24 600 in 1986 due to the development of new lands by settlers coming into the area. However, in 1987 the total area fell to  18  000  ha because of a lack  of rainfall in  April and  early  May. Between 1983 and 1987 there was a change in the cropping pattern with lhe proportion of lhe cropped area under maize falling from 72% to 23%, while sorghum increased from 26% to 66%. The increase in lhe area of sorghum and reduction in that of maize probably reflects the latter*s drought tolerance.  Overall the proportion  of cereal crops fell from 98% to  91% as small areas of legumes and  oil seeds were introduced  by  the resettled farmers. Tables 2.1.1 and 2 J .2 provide lhe data for crop areas in 1987.
The Report states that rain fed farming in the Baro-Akobo gives only poor and unstable yields and  is entirely  dependent on  rainfall and  floodwater.  Il goes on  to  say  that productivity  very  low despite considerable physical efforts But primitive manual labour,  howeverpersistent it might be,  can  never ensure high  srableyields under /he lack of farm implements. transport fact lines, means ofpest and disease control, fertilizer efc.”
Yields were low with the Report quoting for 1987 an average of 1 500 kg/ha for maize
for the local farmers associations and 1 170 kg ha for the rescttlcrs who were occupy ing less fertile lands. The corresponding figures for sorghum were 1 040 kg'ha and 390 kg ha respectively. Table 2.1.3 provides yield data for 1987.
Total Value of Production and Farm Incomes
for (he period 1984-87 lhe Study calculated the total value of crop production from the lower basin. The value of agricultural production was 4.4 times higher in 1987 than in 1984 following the introduction of the resenlcrs and the establishment of the state farm. Area expansion rather than an increase in productivity was lhe cause of the increase.
Faking 1986 as a representative year lhe report calculates that the average income per family was Birr 383 for local farmers associations and Bin 182 for resettlers. Comparable per ha figures were Birr 326 and Birr 214 respectively. These figures excluded the value for vegetables and fruit grown on their household plots.
Changes' Since 798”
Although  no  detailed  and  more recent data are available to  the consultants,  some significant changes since 1987 are known to have occurred For example, there are now four stale farms, the existing population has increased, large numbers of refugees have been, and are still, living in the area and a dam on the Alwero river is under construction which it is planned will irrigate 10 000 ha. In addition, there may have been further in migration from other pans of the country The Russian report provides information at a time when major changes were occurring in terms of crop cultivation and it is impossible to know from that data what is the present situation.
Prospects for Agricultural Development
TAMS-ULG BARO-AKOBO river basin integrated development master plan
1-02. AGRICULTURE
The Report proposes two  development allematives for the Gambela plains based  on  the soils and  available water.  These alternatives were based  on  the development of large mechanised  state farms in  line with  the philosophy  in  vogue in  socialist countries at the time and  since discredited.  There is no  evidence to  suggest that during  the course of the study  the consultants involved  the local people when  planning  their development proposals.
As a result of these two  factors the proposals for crop  production,  despite their great detail in  terms of input requirements,  manpower and  machinery  needs,  costs,  yield  and revenue estimates,  can  be disregarded  as a basis for master planni ng.  However,  some of the basic elements on  which  their proposals were based  eg  the soil studies,  crop rotations,  some of the crop  input output data,  machinery  requirements etc.  are relevant for (he present master planning  process as private sector Investment in  agricultural development in  the lower basin  is one possibility  to  be considered  in  Phase IL  The Report proposed two alternatives “minimum" and "maximum* .
1
The “minimum" alternative was mainly  aimed  at attaining  self-sufficiency  in  food products,  especially  crops commodities,  on  the Gambela plain  for an  increasing population  between  1990  and  2010.  The 'maximum'" alternative was designed  to,  in addition  to  meeting  the food  requirements of the local population,  provide production for export to areas outside the Gambela plain.
The Report proposed two options for the “minimum alternative' a rain fed-oriented option 
and an irrigated option. The irrigated option would involve the development of the
fol lowing ,in addition to the 26 600 ha under cultivation at the lime of the study;
i
•     10 000 ha of new minted lands in flkana-Kijang area with four state farms by 1995;
•     10 400 ha of irrigated land in accordance with the Al wen> project on which work
started in 1987 by 2000; and
•      13  500  ha  of  irrigated  land  in  Ubala-Pashata  area  with  five  state  farms  between  2001 and 2010; the Gilo might provide the water for this scheme.
Under this option,  the area under crop  rotations would  be 60  500  ha.  with  23  900  ha tinder irrigation  .Allowing  for double cropping  the area actually  sown  would  be S3  800  ha.  These developments would  enable the region  to  become self-sufficient in staple food  crops and  by  2010  producing  the following  quantities of marketed production: cotton 41 800 mt. groundnuts 17 700 mt and sesame 1 300 mt.
The  tainted  option  would  involve  the  development  of  the  following  in  addition  to  the 26 600 ha under cultivation at the time of the study;
•     10 000 ha of new rainfed lands in Ukana-Kijang area with four stale farms by 1995;
•       10  400  ha  of  irrigated  land  in  accordance  with  the  Alwenj  project  on  which  work started in 1987 by 2000; and
TAMS-ULG BARO AKOW! FUVER BAStN [NII-GRaITDLM-VELOPMI-NT MASTER PLAS     Z-lGAGRICULTURE
• 61 300 ha of new rainfed lands to lx developed with 17 stale farms between 2001 and 
2020 on the best land in the central part of the Bare-Akobo valley.
Under this option,  there would  be a total 108  300  ha under cultivation,  of which 10  400  ha would  be irrigated  and  96  900  ha rainfed.  Including  double cropping  of the irrigated land the total crop area would be 118 700 ha. Under this option crop production would reach similar levels lo that of the irrigated option.
The ‘maximum alternative’ would  involve the construction  of a sugar complex  in  the years from 2001  to  supplement the “minimum alternative-irrigation  option*.  The additional developments would involve the following.
• an additional 18 000 ha of irrigated land in the Ubala-Pashala area with eight state
farms and a sugar factor}', producing 180 00 to 190 000 mt of cane a year.
/nvertmem Casta and Rates of Return
As the development of state farms is no  longer an  option  for the future development of the area no detailed review of the cost estimates or the calculation of the associated rates of return  has been  attempted.  However,  it appears that the methods used  to  calculate the economic return  might not be in  total accordance with  accepted  procedures.  For example, the capital costs appear to  be in  financial pnccs.  although  it is difficult lo  follow the analysis. Further, the calculation of farm gale values was not shown.
Future Policy re Large Scale Farming
The  Russian  Study  based  its  •'maximum”  alternative  on  the  cultivation  of  a  large  area  of sugar  cane  and  an  associated  sugar  factory  while  its  "minimum  alternatives"  also  involved large  scale  farms.  While  state  farms  are  no  longer  an  option,  the  unofficial  "Agricultural Sector Strategy’ which is believ ed to originate from the Prime Ministers Office envisages the  promotion  of  commercial  scale  farming,  with  long-term  lease  of  land  not  occupied by   smallholders   and   in   a   way   which   would   “nor   jeopardise   the   livelihood   of pastoralists"...  "if  is  in  these  regions  that  most  of  the  major  rivers  could  be  put  into  use for  increased  agro-industrial  production  through  large  scale  irrigation  schemes''.  Most
such  land  is "in  maccextf&te lowland  areas that are malaria  and  tsetse infested", Early difficulties centre on  lack  of infrastructure but it is "expected  that entrepreneurial capacity will participate in  such  ventures.  At the same time divestiture of the existing state farms will continue". Intensive commercial agriculture is envisaged  to  gather momentum towards the end  of the decade,  with  foreign  investment an  important f eature, but "...the most critical factor is farm management". In  the main  it "will be based  on large-scale irrigation  schemes,.  producing  large and  reliable harvests w ith  higher levels ofproductivity per man-hour’
It is quite clear from the foregoing  statements that Government anticipates commercial sector exploitation  of water resources.  Access to  land  and  water resources will,  it seems, involve a commitment to contributing to the cost of developing the infrastructure.
TAMS-ULG baro-akobo river basin integrated development masti h plan
2d!2. AGRICULTURE
[ caving aside the issue of whether or not the private sector would accept the risk of such developments,  there  is  a  potential  for  irrigated  development  on  the  lo  wer  slopes  of  the Baro-Akobo  basin.  However,  howr   such  a  development  would  conflict  with  the  existing grazing  and  wild  life  is  not  known.  Further,  the  area  concerned  is  very  inaccessible. Master  planning  during  Phase  II  will  examine  and  evaluate  alternatives  land  use  systems for the lower basin.
2-2.3 Data Requirements * Upper and Lower Basins
The  analysis  above  has  identified  that  there  are  significant  gaps  in  the  information  which  a master planning exercise will requite for the various agro-climatic zones:
•            views  of  the  fanners,  male  and  female,  as  to  the  constraints  they  face,  their  aspirations
and    their    perception    of    the    development
opportunities
available
information on the constraints they 
face in the
upper ba
sin i
s
availabl
e
to    them:    some from the ARDCO-
GEOSERV study;
• 
number  of  farmers,  the  area  cul
tivated  and 
number
and
type
of
animals  owned  per
household by the main ethnic groups;
• 
information  on  the  crop  rotations
followed  a
nd  for  e
ach 
c
rop 
the
cultivation  practices
, 
inputs  used,  including  draught  power  and  labour  (male  and  female)  and  yields  under traditional and improved systems;
•            information  on  the  type  and  severity  of  crop  pest  and  disease  problems,  including  post harvest.
•            labour  availability,  wage  rates  by  season:  this  information  is  required  for  both  men  and women;
•           suppliers of inputs, prices and seasonal availability and suppliers scale of operation;
•            terms  and  conditions  of  credit  provided  by  the  Agricultural  and  Industrial  Development Bank  and  the  Commercial  Bank  of  Ethiopia,  quantities  being  provided  and  repayment rates;
•            farm  gate  prices  on  a  seasonal  basis  for  the  major  commodities  produced  by  small  scale farmers,  identification  of  the  main  buyers  (eg  private  traders,  government  agency  etc  ), location of depots and scale of operations;
•           details of the extension service, structures, staffing, programmes, constraints etc ;
•           details of crop production research being undertaken in the upper and lowrer basins.
T AMS-U LG UARO-AKOEIO RIVER BASIN INTEGRATED DEVELOPMENT mas ter plan
3-122.           AGRICULTURE
2-2.4 Data Collection
The above list consists of two  types of data.  The first is information  which  should  be available from offices eg  traders,  banks and  government and  parastatal organisations The second  is information  which  would  require discussions with  farmers.  The former types of information would  be obtained  through  visits to  the relevant offices and  the completion  of simple questionaries,  while the second  would  be most easily  and  quickly  obtained  using  PRA techniques.  This approach  would  not involve a formal questionnaire survey  but meetings with key  informants among  the farming  community  and  a checklist of topics for discussion.  Both these exercises should he undertaken during the first quarter of 1996,
2.2.5 Phase fl Activities
The following  paragraphs indicate in  outline some of the activities which  TAMS-ULG w ould undertake during  Phase II as part of the master planning  exercise.  As noted  in section 1 a land evaluation  exercise would  form a major part of the master planning  exercise for the natural resources sector,  including  agriculture.  Specific development proposals for the smallholder and commercial farming sectors arc considered in the paragraphs below.
Smallholder Crop Prodacfion
Development strategy  is likely  to  focus on  promoting  low-cost development of small holder production.  In  view of the rainfall pattern  in  the valley  and  the over-nding limitation of nutrients in crop production. TAMS-ULG would:
•     complete the analysts of the climatic data including the provision of full climate data (temperature, RH. wind, sunshine, rainfall) for a station located in the basin at about 750  m:  very  approximately  it  would  lie  on  the  longitude  35&   E  and  failing  this, undertake some very' careful interpolation using the GIS;
•      review  carefully  the  available  mean  climate  date  in  the  light  of  the  identification  of some doubtful values both in the FAO- Meteorological Unit data used for 25 stations, and in the 5 extra sets extracted from the ARDCO-GEOSERV report;
•      prepare  representative  crop  and  farm  models,  illustrating  typical  crop  production practices  under  traditional  and  improved  systems,  and  providing  information  on  costs and  returns,  labour  use  etc,  for  use  in  land  evaluation  and  in  assessing  alternative crop production options;
•       assess   these   constraints   to   improving   smallholder   crop   production   and   make proposals to ov ercome these, including the promotion of a programme centred upon the supply of organic and inorganic fertilizers;
•      identify  the  constraints  in  distribution  of  fertiliser  and  the  availability  of  credit  and make proposals to overcome them;
•     investigate the opportunities for the development of irrigation schemes.
T kMS-ULCJ B VROAKOBO MVTR BASIN INTEGRATED DEVELOPMENT MASTTlt ELas
2 13I
i
1 AGRICULTURE
•     prepare project profiles for all priority projects and prefearibility studies for priority 
irrigation projects:
•     prepare crop production GIS overlays.
Large Scale /rrigafed Crop Prodbcrion
The most promising  form of development in  the lower val ley  is commercial irrigated farming.  In  view of the very  high  capital requirement to  develop  the whole infrastructure,, an  exceptionally  stable economic environment would  need  to  be created  by  the Government an  order to  attract venture capital.  This raises questions about the developmeni options of the local administration  and  the attitudes of local people.  The second  issue which  would  need  to  be addressed  is the high  cost and  practical difficulty’
of marketing produce from this area.
(n  the  context  of  Government  policy.  TAMS-ULG  would  re-examine  the  potential  for
sugarcane  during  Phase  II.  This  crop  undoubtedly  would  grow  well  and  yield  heavily  in 
j
j
j
j
the    area,    but    with    mean    minimum    lemperature    not    falling    below    18 C    in 
fl
December.fJanuary  there  is  serious  doubt  about  the  successful  maturation  of  the  cane  for
j
harvest  A  heavy  yield  of  cane  with  say  6%  recoverable  sucrose  is  much  less  economic
to  process  than  half  the  yield  of  cane  with  12%  recoverable  sucrose.  Cotton  grown 
wholly  under  irrigation  to  avoid  the  period  of  )ow   radiation  during  the  rainy  season 
r
should yield well and should be profitable if the yield was greater than about 3 mt per ha
. 
Bananas  would  grow  and  produce  well  in  the  area  but  marketing  such  a  perishable  crop 
would   be   the   main   obstacle.   Citrus   and   mangoes   likewise   represent   a   considerable
irrigable  poiential  hut  also  a  marketing  problem.  A  river  route  down  the  Nile  would 
J
J 
J
change the situation.
2.3 Livestock Production
As for crop  production  the main  sources of information  concerning  the livestock  sector in  the Baro-Akobo  basin  are the Russian  report covering  the lower basin  and  the ARDCO-GEOSERV report which  covers the upper basin.  Some comparative data was obtained  from the Livestock Research Report of the Institute of Agricultural Research for 1991.
2.3*1 Upper Basin
Many  parts of Volume Ifi F Livestock  of the ARDCO-GEOSERV report arc difficult to  read, especially  sonic of the tables.  Table 3.F.21  purports to  relate lo  Trypanosomiasis distribution  in Kaff a region  but in  fact is the number or bee hives in  sampled  woreda LWortunaicly,  the analysts and  presentation  of the data does not do  justice to  the large amount of potennaily valuable information  collected.  It is also  unfortunate that the report seems to  rely  on  data (eg concerning  tse-tst and  trypanosomiasis) some of which  is 20  years old  A lack  of consistency  m the data provided  by  the wortdas makes it difficult to  analyse and  use such  information  in  any planning exercise.
TAMS-ULG BARO-aKOflO river basin integrated development master planAGRICULTURE
Method of Approach
ARDCO-GEOSERV based  their approach  upon  surveys.  The first involved  67  randomly selected  livestock  farmers.  This data was supplemented  by  information  collected  from
14  woreda  offices.  In  addition,  83  farmers  were  interviewed  concerning  the  disease situation  in  their  particular  areas.  Blood  tests  were  taken  from  94  sick  animals  in  nine woredas,  107  ticks  were  collected  from  animals  in  eight  woredas  and  17  flies  were collected  from  two  sites.  Clearly.  ARDCO-GEOSERV  collected  their  data  from  a  very small sample of fanners and animals etc.
L/vcsrodr Breed*, and Types
The Report notes that the cattle of the upper basin  are of a non-descripl (local type). There is some evidence of recently  introduced  genetic material eg  Friesian.  The local cattle are typically  small (200-250kg),  generally  brown  (this colour is reputedly the least attractive to  tsetse),  less commonly  with  white and  black  colouration,  they  are normally horned  but some naturally  polled  types were observed.  The majority  arc described  as 'undefined  breeds'  and  ARDCO-GEOSERV mention  only  one specific breed,  the Sheko which has a potential to yield up to 5 litres of milk per day. Although other distinctive types occur they  arc not mentioned  The report does not mention  breeds of sheep,  goals and poultry.
Animal Numbers
The ARDCO-GEOSERV report sets out the livestock  numbers for a part of the upper basin  in  two  tables: Table 3.F.I(a)(Wclega and  llubaborland  Tabic 3Flb  (W.  Wdcga). There is some doubt over the accuracy of these figures owing to the w ide discrepancy between those for 1987 and 1989 in Table 3.F. 1(b) Figures for those parts of (he upper basin that lie in the Southern Peoples Autonomous Region arc not included in the report. From conversations with  government stall’ the overall impression  is of a gradually increasing livestock population in spite of obvious constraints of health and nutrition. In llie case of cattle and sheep, this impression is tentatively confirmed by the very limited information that has been provided on die sale of hides and skins through Bcdelc.
Herd Composition
Table 3.F.2  shows the composition  of the cattle herd  from the farmers surveyed  but as two  columns (F&G) appear to  include cattle of the same age group  it is difficult to deduce anything  worthwhile from the tabic.  In  1992  in  eight woredas in  Gimbi Awraja
the livestock holding were as follows.
cattle
sheep
goal
horse
mule
donkey
poultry
average
6.5
0.8
0.2
0.02
0.04
0.3
1.4
range
4.5-9,9
0,2 * 14
0.03 - 0.5
0.02-0.1
0 02-0.1
0.1-0.8
0.&-3.0
The outstanding  importance of catlle in  livestock  herds is highlighted  by  these figures. The relative unimportance of poultry  is another notable feature,  and  is thought to  apply to the basin generally.
T AMS-TJLG HARO-AKOBO RIVFR BASIN INTEGRATED DEVFl OPMENT MASTER PLAN
2 45AGRICULTURE
The ARDCO-GEOSERV report also  highlighted  the range of the sizes of the livestock units as shown  by  their survey  with  cattle holdings ranging  from 5  for poor households to JO for rich ones. The criteria for the definition of rich, middle and poor are not staled, but it ts assumed that it refers to households with large, medium and small herds
Livestock Performance
The range in  milk  yields reported  by  lhe ARDCO-GEOSERV study,  14-2.1  litres per day, related to the age of the cows with the lower figure for 2-3 year old heifers and the higher Figure for cows more than  four years old.  The overall average milk  yield  for the upper basin  quoted  by  ARDCO-GEOSERV is 2.5  litres per day  [lie Report also  noted that lhe potential of the indigenous cow is much  higher than  these average figures suggest,  particularly  if selection,  improved  nutrition  and  health  care arc applied.  Mo information  on  milk  quality  was provided.  For the upper basin  as a whole ARDCO- GEOSERV report that the age of fimt calving and average birth weights are 2.9 years and 
15-25 kg respectively,
The ARDCO-GEOSERV also  report quoted  parameters for sheep  and  goats,  bill do  not indicate whether the data are derived  from the survey  or from government statistics. Some of the figures for goats appear somewhat suspect First the average number of kids per kidding  is two  but the percentage of nanny  gnats kidding  having  1,2  or 3  kids only totals 55% (page 22).  Secondly  ARDCO-GEOSERV gives two  different average live weights for adult males ic billy  goats 48  kg  and  bucks 65  kg.  Since in  English  a buck  is identical to  a billy  goat it is difficult io  interpret these figures.  Die same occurs with nanny goats 28-42 kg and does 55 kg.
Livestock Consumption and Marketing
ARDCO-GEOSERV provide some information  on  livestock  consumption  reporting  that "98% of the goats in  the Asosa  region  are consumed  locally,  while 80% of cattle, 75% of sheep  and  5% of goats are consumed  at Gimhi where rhe number of consumers is relatively greater than  the rest of lhe woredas in  the region  ', it is assumed  that the percentages in lhe preceding sentence refer to lhe animals produced in Gimhi.
Many woredas and other centres of population have market days (eg once or twice a
week). For selected woredas lhe report provides some information on lhe numbers of
animals sold but it is difficult to quantify this trade: lhe ARDCO-GEOSERV report
contains information such as: 'there are 21 markets in Welegd and rhe cattle markets
of (limbi, Nekcmpte, Yubdo. Mugi. and Segno-Gebeya handle 67% of the cattle marketed
in the region'. 'The main market day - there are two per week - at Dega is reported to
attract 200 cattle. 400 sheep. 200 goats and 250 chickens, '
It is also  implied  in  lhe report that significant numbers of livestock,  especially  from areas close to the border are sold in Sudan. The information below is derived from market information contained in Table 3.F.6 (undated) and obtained from the survey.
TAMS-ULG baro-akouo Rjvtit basin Integratbd development master plan
2-112.        AGRICULTURE
Oxen
Cow
Calf
no.
kg lw
Birr/kg
no
kg lw
Birr Teg
no.
kg lw
Birr/kg
Uta
4
258
2.28
2
200
22
4
95
2.7
Metu
6
298
233
4
229
1.9
Tepi
2
no
1.20
1
240
13
1
123
1J
Milk is marketed in the larger cenLres al Birr 1.50 to 2.0 per litre: it is sold loose and untreated. !l is often  adulterated  with  water and  is usually  boiled  before consumption  as TH and  Brucella is a risk. Hie high price (about the wage for three hours work) reflects the strong demand for the product.
Cattle prices arc higher al holiday  limes and  lower during  the rainy  season  when  farmers sell to  minimise the risk  of carrying  the animals through  the rainy  season  when  the risk of loss through  disease is at its maximum.  The data provided  does not give a coherent picture as the same information is not provided for each woreda.
ARDCO-GEOSERV sets out statistics for hides and  skins despatched  from Bedelc in  the period  1991-1993  (Table 2.2.1).  These figures do  not represent the production  from the upper basin  as they  include some unquantified  adjacent areas.  However.  they  probably indicate some of the trends in  respect of this important trade.  The report also  observes that 30% of hides are used  for domestic purposes; 15% of sheep  skins for saddles; and 75% of goat skins for honey  bags.  The following  observations can  be made from the figures in Table 2.2.1:
•     the delivery of cattle hides increased by 44% in the period 1991 to 1993;
•     the quality of hides improved;
•      the  importance  of  air  drying  of  sheep  skins  increased  greatly  from  an  insignificant proportion to over 50% during the period; and
•     the trade in goat skins declined by 30%.
It is not possible to  draw categoric conclusions,  but it appears that the volume of trade and  the quality  of the products increased  over the period,  perhaps due to  an  increase in livestock  numbers,  and  an  increasing  awareness of the value and  the importance of the quality  of hides and  skins.  Increased  extension  activities of the hides and  skins technicians at woreda,  zonal and  regional level may  have had  an  impact the apparent increase in the marketing of hides and skins.
The manure from livestock  is an  important by-product and  is used  both  for enriching  the soil,  particularly  for perennial crops and  also  as a fuel.  For the latter purpose it is dried and ARDCO-GEOSERV reports dial it is sold sometimes at prices ranging from
T AMS*ULG IIARO AKOB0 wx er basin integrated development master plan
2-i?J 1 1 J J n
] u
j u u a 2 2 ’ll 0 2 2 2 ’J
2.          AGRICULTURE
Birr  0.16  to  0.38  per  kg.  The  use  of  manure  as  a  fuel  represents  an  irretrievable  loss  ol soil condition and fertility.
Financial Returns from Livestock Production
ARDCO-GEOSERV did  not assess the costs and  returns from livestock  production.  No information  on  labour required  for livestock  production  is provided.  An  analysis of the costs and  returns from livestock  production  is necessary  if alternative land  use systems arc to be compared during a master planning exercise.
Feed Resources
The graying  is very  diverse in  respect of composition,  reflecting  the ecology  of the vegetation  types and  also  the intensity  of grazing  to  which  it has been  subjected.  This grazing intensity in turn is determined by;
•     density of livestock numbers;
•    proximity' to centres of population; and
•     incidence of trypanosomiasis, malaria and other diseases of man and animal.
ARDCO-GEOSERV report extensive areas of savanna lands especially  in  Asosa and  in Guraferda,  which,  although  they  have been  degraded  through  excessive burning,  offer potential for livestock  development,  in  conjunction  with  measures to  control the tsetse fly.
Grazing  comprises land  that has never been  cultivated,  abandoned  crop  lands and  during the dry  season,  animals often  graze within  (he forests and  plantations.  Tn  general,  grazing is not adequate for the livestock  populations in  most localities.  However,  this was not the case at Nlizan  Teferi,  perhaps because the production  of coffee instead  of annual crops reduces the need  for draught oxen.  ARDCO-GEOSERV did  not quantify  the areas or yields of the grazing  resources of the upper basin.  In  very  general terms,  the dry  season represents the period  of shortage of natural grazing,  while in  the wet season, approximately from June to September, it is relatively abundant.
ARDCO-GEOSERV report the successful introduction at certain unspecified sites in the
upper basin of species of jVfetffcago, Desmodium. 
Fj’da and Cynodon The
residues left from annual crops is the main supplementary feed for livestock. They 
are also used for fuel, house construction and bedding (ARDCO: Table 3.F. IB)* 
The report does not mention whether the brewers grains from the brewery at Bedele
are used as a source of livestock feed
Health and Disease
The ARDCO-GEOSERV report contains about 50  pages on  the incidence of animal diseases,  their vectors and  the veterinary  services in  the upper basin.  For the most part the data is comprehensive,  although  ARDCO-GEOSERV did  not analyse fully  the results of the tanner questionnaire. Information on the incidence of diseases, other than
T AMS-ULiG b aroarobij reveel hasin integrated development master, pi-an
2-lSAGRICULTURE
trypanosomiasis and lick home disease, and their treatment al woreda level is less good 
The following section draws out the main points and makes further analyses of the data
in the report
The main  source of information  on  the distribution  of tsetse and  trypanosomiasis continues to  be the Langridge report of 1976.  At that time,  nationally  about 98  000  km- were infested.  ARDCO-GEOSERV quoting  more recent studies report that the area had since increased  by  over 20% lo 120 000 km*. The reasons for this trend include changes in  land  use.  the increasing  movement of people and  their cattle from the highlands into the valleys in  response to  population  pressure,  and  the changing  nature of the tsetse fly. It is reported  by  National Tsetse and  Trypanosomiasis Investigation  and  Control Centre that the fly now breeds at altitudes up to 2 000 m. Formerly it was restricted to altitudes below 1 700 m.
ARDCO-GEOSERV describe die distribution of tsetse and trypanosomiasis in the upper
basin in detail. This information may be out of date having been taken from earlier
reports, as ARDCO-GEOSERV refer to the possible expansion of the fly into the Didessa Valley (p.50), although on p.53 a contradictory reference is made to the infestation of the
valley by G. nj suimorsftony). Staff at National Tsetse and Trypanosomiasis
Investigation and Control Centre reported to TAMS-ULG that this valley has been 
infested for many years and is the site of a control and eradication programme.
ARDCOGEOSERV   covers   ticks   and   tick   borne   diseases   in   detail   but   only   quotes previous  studies  and  so  the  information  could  be  out  of  data  as  the  main  work  referred to  was  undertaken  in  1986/88.  One  important  result  of  this  work  relates  to  the  absence (at the time of the survey) of evidence of Theiieria purvu and the lick which carries it R appendiculatus.  This  agent  is  responsible  for  East  Coast  Fever  (ECF),  which  is  a  maior and  fatal  disease  of  much  of  East  .Africa.  With  the  uncontrolled  movement  of  cattle across  the  borders  with  Kenya  and  Sudan,  its  introduction  into  Ethiopia  with  very  serious consequences,  is  a  major  hazard.  ITic  tick  species.  ftoophifus  annulafus  was  introduced into Gambela during the 1980*5. and the view is expressed that it is only a matter of time before  ECF  is  introduced,  unless  strong  precautions  are  taken.  The  ARDCO-GEOSERV report  also  makes  the  important  observation  that  as  tsetse  control  is  implemented  in  the border  areas,  this  will  enable  livestock  in  general,  and  cattle  in  particular  to  move  into Ethiopia on a wider front. In other words, it will expose a longer length of border to the ingress  of  affected  livestock  so  increasing  the  risks  of  the  introduction  of  new  vectors and new diseases.
According  lo  .ARDCO-GEOSERV numerous dip  tanks have been  constructed  in  the upper basin  but most of these are non-functional Veterinary  authorities now believe that regular dipping  is unnecessary,  and  if it is applied  irregularly,  it can  adversely  affect the health  ofcattic.  If ECF is inadvertently  introduced  to  the Basin,  regular dipping  would then become essential.
The ARDCO-GEOSERV report contains much  information  on  animal disease from a number of woredas.  The data are not analysed  and  it is therefore difficult to  understand easily the pattern of importance and frequency of the different diseases. In the -'review ’
1
T AMS-l FLG baro-akobo river basin intfgmtcd df vei opmekt masti « pi an
2-!9I I 1 n 
ji
:ii j
u 
ii
2. AGRICULTURE
on  page 9K,  pasteurellosis,  anthrax  and  blackleg  are recorded  as the most frequent in  the upper basin,  although  the accuracy  of identification  is questioned  as a result ol inadequate equipment al the veterinary clinics.
In  addition  to  diseases,  heavy  losses of livestock  are reported  from other causes, accident,  plant poisoning  (especially  during  the dry  season  when  forage is scarce), predation  by  animals (one suspects lhat with  the demise of most wild  carnivores,  the main  culprits may  be human) and  theft.  Additional fatality'  and  morbidity  also  reportedly occur as a result of poor nutrition and shortage of trace elements.
Zfvestocfc Supporf Services
The  veterinary  service  is  of  primary  importance  to  the  sector.  Since  about  1992,  the
ji
department   has   been   within   the   MO   A.   Resources   like   transport   that   were   formerly 
controlled  by  the  vet  section  arc  now  shared  with  other  units  al  the  MOA  offices.  The
main point of service provision are the veterinary' clinics at each wo re da office, together
ji
with  □  small  laboratory  and  equipment  for  diagnostic  purposes.  A  crush  for  restraining 
animals  is  normally  located  nearby.  The  report  cites  a  shortage  of  equipment,  drugs, 
medicines  and  vaccines  as  the  main  difficulties  encountered  At  the  zonal  level,  similar
facilities arc available, and there are diagnostic laboratories at several regional centres.
Within   the   Basin,   the   National   Tsetse   and   Trypanosomiasis   Investigation   and   Control
Centre  is  located  at  Bedele.  The  centre  also  controls  the  campaign  against  rinderpest, covering an area that is a little larger than the basin.
The report includes no description or comment on the livestock ex tension services. There are no centres of livestock research within the basin. The National Livestock Research Centre at Holleta and the Bako Research Centre undertake research lhat is relevant to 
livestock and forage development in the basin, but the Report contains no details.
u 
d)
u
The   main   centres   have   a   basic   facility'   for   the   slaughter   of   livestock,   which   is
administered  by  the  local  council  which  charges  a  fee  ranging  from  Bin  5-16.  Veterinary 
staff  were  reported  to  be  in  attendance  for  both  ante  and  post  mortem  inspection.  There
are no facilities reported for the processing of milk, meat or hides and skins in the basin
The  following  NGO’s  are  active  or  arc  considering  starling  projects  involving  livestock 
in (he project area but no details are provided:
J
•     Menschcn fur Menschcn in Metu; and
•     DED in Asosa.
/fecornmrndanoraj /Or Figure De Vf/opmenf
fjl
ARDCO-GEOSERV  make  three  brief  recommendations  for  the  development
of
l  ivestock 
production:
jl z1
1
TAMS-ULG BAROAKOBO1UVTR BASIN INTEGRATED DEVELOPMENT MASTER PLAN
2 -2DAGRICULTURE
•      rational  use  of  the  dense  forest  through  the  opening  up  of  access  roads  to  overcome
the shortage of pasture lands,
•     intensification of livestock production in areas where there are ample pastures; and
•      integration  of  crops  and  livestock  in  the  lllubahor  and  Kcffa  farming  communities
The  crop  production  aspects  can  be  linked  to  intensive  beef  production  using  crop residues as well as the introduction of improved fodder crops.
The first recommendation  conflicts with  the need  to  stop  the destruction  of the remaining forests which  are located  on  steep  slopes and  is environmentally  unsound.  Making  use
of these areas for livestock  production  conflicts with  the need  to  protect the watersheds. In  addition,  the recommendation  seems to  be at variance with  die second  one.  I f there an: areas where ample pastures exist why is there a need to open up the remaining areas of dense forest for livestock  grazing? The third  recommendation  is appropriate and  has no  doubt been  made countless limes before,  but the ARDCO-GEOSERV report contains no  strategy  to  achieve the objective of integrating  crop  and  livestock  production.  There arc no detailed proposals let alone a costed programme.
For the veterinary services the major recommendations were as follows;
•      MOA  should  revise  its  policy  and  give  the  Veterinary  Department  full  autonomy including  control  ovct   budget,  manpower  and  facilities  including  vehicles  (one  car per  awraja  and  3-4  motorcycles  per  woreda)  with  the  provision  of  sufficient  funds  to cov er running costs, per diems etc. and regular refresher courses for staff:
•      the  veterinary  climes  should  be  properly  equipped  and  have  sufficient  equipment, drugs,  vaccines,  reagents  etc.:  (he  provision  of  adequate  drugs  would  discourage  the widespread illegal drugs trade leading to resistance due to dilution and improper use;
•     lhe introduction of coordinated legislation regarding meal inspection, quarantine and 
sale of veterinary drugs (no further details were provided);
•      veterinarians  should  be  able  to  undertake  research  into  the  livestock  disease problems being faced by farmers for which they require incentives, transport fuel etc.;
•       lhe  National  Veterinary  Institute  should  receive  sufficient  funds  io  produce  enough vaccines  to  meet  the  needs  or  all  regions  and  their  subsequent  distribution  should distributed with lhe minimum of bureaucracy;
•      the establishment of a strong quarantine and cattle route system to stop diseases from being introduced into the country.
•      dip  tanks  should  not  he  built  and  ticks  should  be  controlled  where  necessarv  using hand sprays; and
TAMS-ULG daroakobo      RIVER BASIN INTEGRATED DSVtl OPMEXT MASH R PLAN
2-21AGRICULTURE
•   sufficient   funds   should   be   allocated   to   the   National   Tsetse   and   Trypanosomiasis Investigation  and  Control  Centre  to  use  the  target  technology  to  eradicate  tsetse  <tnd control trypimasamiasis in the basin.
The first five recommendations amount to  a request for more money,  ARDCO- GEQSERV does not appear to  have considered  any  new approaches to  the provision  of veterinary  services.  Government budgets are and  will always be under severe pressure and  ways need  to  be found  to  provide an  effective service within  the likely  future financial constraints Inevitably  this will mean  a budget substantially  less than  that considered  neccssarv  by  the Department.  No  mention  is made in  the report of the possibility  of using  village vaccinators who  would  charge tees,  other cost recovery measures or the involvement of the private sector in the supply of drugs
2.3.2 Lower Basin
The Russian  report was the result of several years of painstaking  field  work,  data collect ion, collation,  analysis and  editing,  although  the livestock  sector,  part of Annex  5,  is not as comprehensive as some of the other sections perhaps reflecting  the greater difficulty  of collecting data on  livestock,  especially  in  respect of those that arc migratory.  The Russian  report is based on field work that was begun nearly ten years ago and which finished in 1987-The study although technically  thorough  is now out-of-date with  respect to  development strategy  and  statistics.  The former concentrated  largely  in  terms of the development of stale farms.  The following  areas received insufficient attention:
•           the subsistence sector and its support and development;
•           seasonality of labour requirements for the various development proposals; and
•           the possibilities of improvement of efficiency through yield increase
A/crhcJ of Approach
The survey  of the natural resources of the lower basin  was complemented  by  statistical data from MO A covering  livestock  numbers,  production  and  support services.  In  the lower basin  data collection  was,  and  still is,  difficult because of the nomadic life style of the livestock owners, the Nuers. and their herds.
Livestock Breeds and Production Systems
All lhe cattle are of the zebu type, small and well adapted to the harsh climatic conditions in the lower basin with resistance to a number of cattle diseases. The local Ab i gar breed which  belong  io  the Nuers is also  found  more extensively  in  Sudan.  According  to  the Russian  study,  quoting  the Veterinary  Department,  reported  that the breed  is perfectly suited to the conditions in the lower basin. It also considered that the Ab i gar was a meat rather than  a milk  breed.  Sheep  and goats were generally  of local breeds and  very  small. In 1995 some larger brown sheep, more closely related to the desert ty pe of sheep were observed in the lower basin.
TAMS-ULG BAROAKOBC RIVER basin integrated development m aster plan
2 222.         AGRICULTURE
At the time of the Russian  study  livestock  in  the lower basin  belonged  to  the Ncur.. nomads who  lived  in  the northwestern  part of the Gambela plain,  within  Akobo,  Ilang and  Jikawu  wotedas During  the dry  season  the Nuers were to  be found  on  the hanks of the permanent rivers where they could obtain water, grow- sorghum and where there was grazing  for their livestock.  In  the wet season  when  these areas were flooded  the Nuers moved  northwards back  to  their major pastures which  had  regenerated  with  the onset of the rains.  Howr and  if these migration  patterns have changed  during  the last 7-9  years is unknown as is the impact on die Nuers traditional system of the resettlement programme, dam construction,  influx  of refugees and  the Fullata who  also  have livestock  herds. Information on these aspects of the livestock sy stem is necessary' for any master planning exercise.
Animal Numbers and Herd Composition
The Russian  study  notes that there arc *>w relevant official statistics. 'The data supplied by the audiorities in Gambela and MO A contradicted each other and are at variance with data provided  in  project study  reports.  According  to  the authorities in  Gambela,  the
numbers of livestock were as follows in 1986 (*000):
Worcda
Cattle
Sheep
Goats
Poulin
Akobo
135.0
32.4
43.7
13.0
Jikao
105.0
25.2
33.9
13 0
Itang
60.0
144
19.4
19.5
Gambela
-
-
•
26.0
Abobo
-
♦
-
32.5
Gog-Jor
*
•
-
26.0
TOTAL
300.0
72.0
97.0
130.0
equines nut recorded
But in June 1988, the Gambela authorities reported 321 300 cattle and 180 000 sheep and goats  representing  increases  of  respectively  a  7.1%  and  6.5%  over  a  two  year  period. While  this  is  feasible  especially  in  a  situation  where  there  is  cross  border  livestock migration,  other  sources  quoted  by  the  Russian’s  gave  much  lower  estimates.  According to the TAMS report the largest cattle population ever kept in Gambcla was about 150 000 head.  In  the  Alwcro  project  report  official  data  supplied  by  MOA  and  the  Gambela authorities stated that there were only 29 000 head of cattle and 33 000 sheep and goats Discussions with Neur. tribesmen in the western part of the lower basin m 1988 indicated that  individual  households  owned  no  more  than  20  head  as  there  was  insufficient  pasture to keep more. This finding led the Russian study to conclude that towards (he end of the 1980s there were no more than 150 000 livestock in the plain.
Livestock  statistics  provided  to  the  consultants  in  September  1995  gave  exactly  the  same numbers for cartie sheep and goats in Akobo and Jikao woredas as those provided by the
T A M S U LG IH RO- AKOPO R f VER P AS IN INTEGRATED DEVELOPMENT MAi l EH PLAN
2-23II
I1
2,         AGRICULTURE
J 
-Il
Gambol  a  authorities  in  1986.  Ihe  conclusion  to  be  drawn  from  the  figures  provided 
above is that I lie figures are so inconsistent as io make them invalid without some further
corroboration   and   understanding   of   migrations.   For   master   planning   a   more   reliable
estimate of the livestock numbers in the lower basin is necessary.
-H
The   Russian   study   provides   no   breakdown   of   individual   herds   between   animals   of
different  ages  or  by  sex,  but  notes  that  up  to  30%  of  the  cattle  population  comprises  old 
bulls  and  castrated  males  and  other  low  productive  animals  Each  family  was  reported  us
J
keeping 15-30 cattle and 7-8 goals and sheep, allhough some families had up to 80 cattle. JLivfts/odl Per/ormaflce
J
3
J 
J 
The  cows  weigh  on  average  250-300  kg  and  4-5  year  old  bulls  300-350  kg.  On  average
calves  weigh  18-20  kg  at  birth  Milk  yields  are  2.6-3.11  litres  per  day.  giving  550-650 
litres  per  lactation  of  210-215  days.  The  average  weigh!  gain  for  calves  is  180-240  gm
per  day.  Unlike  ihe  ARDCO-GEOSERV  report  for  the  upper  basin  the  Russian  report
includes  no  other  parameters  eg  age  of  first  calving,  calving  interval,  number  of  calves
per  cow  etc..  Sheep  usually  weigh  25-30  kg  on  the  hoof,  yielding  13-17  kg  of  carcass
weight.  Weight  at  birth  is  2-3  kg.  As  for  cattle  the  Report  gives  no  other  production 
parameters.
Zfrejrtodt Conswmprffln an*/ .Warfcerirtg
J 
3
J 
The absence of relevant statistics prevented  the Russians from estimating  the actual consumption  of meat in  the lower basin.  According  to  the Gambela authorities I 800 cattle,  1  300  sheep  and  goats were marketed  in  Gambela during  1986.  'Fhe corresponding 0gores for ihe first four months of 1987  were I 100  and  400  respectively.  A lot of animals were reportedly  sold  at I tang  but the absence of suitable records prevented  the Russians from being  able to  make any  definite statements as to  the quantities of meat or
other livestock commodities produced in the lower basin.
J 
J 
J 
Instead  the  Report  includes  estimates  based  on  the  livestock  numbers  giving  totals  of
8 200 mt of beef and 700 mt mutton, both carcass weights. However, the Report does not
include  the  assumptions  used  to  calculate  these  figures.  Further,  it  concludes  that  these
quantities  seem  to  be  more  than  sufficient  to  meet  the  needs  of  the  local  population  and 
suggests  that  a  proportion  “becomes  ihe  booty  of  armed  bandits*  and  ends  up  in  Sudan, 
while  livestock  bought  in  Gambela  is  also  not  consumed  in  the  area  but  driven  to  the
highlands.  There  is  clearly  the  need  for  additional  data  to  understand  livestock  marketing 
in the lower basin, as the pattern in 1986/88 is incomplete and changes are likely to have
J vll
J
I
occurred during the last 7-9 years.
Financier/ Returns from Livestock Production
The report provides no  information  of the returns from the existing  systems of livestock production  in  the lower basin.  For master planning  purposes the consultants will need  to have information on the gross margins from different livestock systems if the returns
TA MS-U LG HARO-AKOUO RIVER BASIN INTEGRATED DEVELOPMENT MASTER Fl-AN
2 242.
AGRICULTURE
from  livestock  arc
to  he  compare
d  wi
th  those
fr
om
other  compe
ting  uses  of  the  l
and 
(ie
other Land Utilisatio
n Types LUTs),
Feed Resources
The  vegetation  of
the  Plain  was
the
subject
of
a
very  detailed 
study  undertaken 
by 
the
Russian  team over a period  of two  years February  1987  to  February  1989.  The depth  of the study  was such  that the authors of the report described  the survey  as unique (at that lime) in  Africa.  The results were presented  in  a schematic geobotanieal map  at a scale of I : 200  OOO.This study  classified  the vegetation  into  three main  ‘pasture zones’ as follows:
•     seasonally flooded pasture, dominated by Hyporr/iema, Ec/nnocWocr, cifld Penniyerum spp.;
•     pastures in die plain that are not normally inundated, including some open savannah and forest lands: Andropogon, Hyparrhenia, Echinochloa  Heteropogon and Panicum
r
spp. dominate this zone; and
•     pastures on the piedmont and adjoining uplands, including savannah and forest lands: Hyparrhenia, Panicum, Pennisetum and Loudetia spp dominate this zone;
The area,  productivity  and  yield  for each  zone as assessed  in  the study  are set out in  a
simplified form below together with the consultant’s suggested season for grazing.
Past urt Type
Hectares
(■000)
Estimated
Season of Grazing
Yield (‘000
mt) Green
.Seasonally flooded
2 008
6 986
October to July
Non * flooded lowland
739
7 420
May to October
Piedmont pasture
803
1 245
May to November
Total
2 550
15 651
The seasonally  flooded  zone lies below 425  m and  is characterised  by  a period  of inundation  of four to  five months each  year from about August to  December,  The length and  period  of flooding  varies throughout the zone according  to  the local topography  as well as the rainfall conditions of that particular year.  As a result there arc several grass associations within  the zone.  Where (he flooding  period  is long,  Echinochloa spp. flourish  and  grow- up  with  the rising  water level.  These species are particularly  nutritious and  arc favoured  by  grazing  stock  Echinochloa spp.  also  have the capacity,  when  the water level recedes and the stems fall down on the moist soil, of rooting at the nodes and thereby  producing  another flush  of forage.  In  the areas with  a shorter period  of inundation  Hyparrhenia  hirta predominates,  yielding  an  inferior feed  for livestock. Species of Paspa/t/m.  Pernikretom and  Andropngon also  occur but less frequently.  It is reported  that where grazing  intensity  is high,  superior grass species establish  such  as Cynotion and Dactylocicnium.
TAMS-ULG BARO-AKOBO river basin integrated development MASTER PLAN
2-25AGRICULTURE
The non-flooded  zone lies between  425  m and  550  m.  Echinochloa sp  persists in  the moister areas of this zone,  with  occasional trees of Pd  iostigma  thorn  ingif,  Terminalia mollis, and  Cornftrefiim collinum. Secondly,  the slightly  higher localities are characterised  by  Heteropogon.  Andropogon.  Bothriochioa,  and  Digitaria. In  open forests and  woodlands,  stands of Pcnnisetum.  Hyparrhenia,  Heteropogon  and Dichanthium are the grasses characteristic of these productive soils.
The piedmont pasture lies above the 550  m contour: it is heterogeneous,  but of generally much  lower productivity.  The characteristic grasses are Hyparrhenia  rufa.  Thenteda triandra. and Loudetia sp
Animal Health and Disease
The Russian  study  included  no  special research  into  livestock  diseases in  the lower basin. Data was taken  from the 1975  TAMS Gambela Project Report and  was provided  by Ethiopian  specialists.  According  to  the TAMS report the livestock  in  the lower basin suffered  from practically  every  disease known.  Particularly  widespread  were rinderpest, pasteurellosis,  brucellosis,  anthrax,  aphtha,  viral bronchopneumonia,  trypanosomiasis and tick borne diseases.
Despite the absence of official statistics there was a consensus that anima] diseases cause large losses.  Rinderpest and  trypanosomiasis were the most serious with  the former a particular problem in  Akobo  woreda on  the border with  Sudan.  Death  rates of 20-25% were reported  in  some years.  It is not clear from the Russian  report when  these losses occurred  and  with  the control programme the situation  may  now be much  improved. Viral pneumonia and  pasteurellosis are most dangerous for young  cattle and  death  rates for this age group were as high as 15%.
Trypanosomiasis was reported  to  be the greatest peril,  especially  in  the central part of the plain  which  was why  livestock  were not kepi these despite an  abundance of forage in 
1986/87  The Report considered  that it was time to  take action  to  tackle the problem of trypanosomiasis as practical ways of combatting  the problem were available.  Drugs had not been  tested  and  their efficiency  was unknown.  The report noted  that other control measures such  as clearing  and  burning  of bushes and  scrub,  spraying  with  disinfectants, creation  of special buffer zones had  not been  attempted  t As noted  above new techniques are now available to  reduce the population  of tse tse flics and  hence the incidence of trypanosomiasis.
It is clear from the above that there is a need  to  update the information  in  the Russian report on  animal health  and  diseases and  possible disease control measures if the master plan is to be able to propose a rational strategy' for the future development of the livestock sector in the lower basin.
ifmfodc Support Services
According  to  the  Russian  Report  the  veterinary  sen  ice  was  “m  rAe  moimg'at  the  time of their field work during the late 1980s. There were three Veterinary surgeons, two
TAMS-ULG BARO-AKoHO RIVER BASIN INTFGRATFD development master Pl AN2.         AGRICULTURE
assistants and  four attendants whose tasks were lo  undertake vaccination  programmes when  necessary  and  to  treat animals when  called  upon.  Clinics were to  be found  in Akobo and Gambcla woredas with two each in Jikawo and ltang. The Report notes that there was a shortage of medical instruments,  equipment,  drugs and  transport and  no diagnostic facilities and that the service was unable to influence the disease situation in the lower basin  there is no  information  in  the report concerning  the presence of any livestock extension service
For master planning  the consultants will need  to  update information  on  the livestock support service in  the lower basin  including  an  inventory  of the veterinary  clinics,  an assessment of their condition  and  facilities,  details of the staffing  of the Veterinary Department and information on the slaughter houses and similar facilities.
.Recommendations for Future Development
The Report notes that: lhefacf that the entire livestock population  is privately owned, mainly by nomads, which makes planning and control of livestock population and yields an  extremely dijfificull task ’ and  later 'homadic nodtAreediJig  poses imww problems for zootechnical and  veterinary serv ices Breed  improvement is practically impossible
to organise. *
In the light of the above comments the Report makes only general recommendations and proposes the establishment of a dairy  farm on  every'  crop  state farm proposed  for the central crop  producing  area.  These farms would  supply  meat and  milk  for the consumption within the lower basin. Initially three farms were proposed because of the lack  oI experience in  trypanosomiasis control.  Proposed  locations were the Abobo  and Alwcre state farms and possible in Gogjor wcreda. After 4-5 years and positive results further dairy- farms would be established on other state farms. Each farm would have I 000 cattle including 500 cows of the local Abigar breed which w ith proper feeding were expected to produce 1 600 kg of milk per year. Crop and processing byproducts would provide part of the feed.
To improve the social and economic conditions of the nomadic herdsmen in the western part of the lower basin the Report makes the following proposals:
•    construction of an all-weather road to connect Gambela town with the most densely populated livestock production areas;
•    establishment of stockbreeders’ associations comprising 150-350 Neur. families, with 4 000 to 6 000 livestock in each association;
•    construction of central farmsteads for the stockbreeders' associations, close to the all- weather road;
•     provision of drinking water supplies in the pastures to improve their utilisation; and
TAMS-ULG [I ARO-AK.OH0 RJVER BASIN INTEGRATED DEVELOPMENT MASI I R PLAN
2-271
1
J J 1
] ] J
.1
2.
AGRICULTURE
• an increase the number of privately owned poultry and the construction of an egg 
factor)' for 60 000 layers producing 12 million eggs annually.
There  is  no  evidence  from  the  report  that  while  making  these  proposals  (he  views  of  the nomadic herdsmen were sought and taken into consideration.
As the stocking  rale in  the grazing  areas was high  the Report noted  that it was not expedient to plan on any increase in livestock number in the next 15-20 years, except for the numbers of catlie on  the proposed  dairy  farms,  fhe recommendations arc dated  and reflect the emphasis of the time on  centralised  planning,  collectivisation  of herders,  as happened in Mongolia, and a command economy. As noted earlier this approach has been discredited and replaced with an emphasis on herdsmen and participation,
233 Data Requirements - lipper and Lower Basin
From the preceding sections it is clear that the data provided both in the ARDCO-GEOSERV and Russian  reports provides a far from complete and,  in  the case of the latter,  an  up-to-date picture of the livestock  sector.  The following  information  is required  for the preparation  of the master
plan;
]
3
1
• 
completion of the livestock inventory of the basin, (TAMS-ULG have already collected data on livestock numbers from most of the woredas) including an assessment of the
numbers of livestock belonging to refugees and the Fullata; there is a particular need to check the numbers provided in September for the lower basin as in a number of instances
they are the same as for 1986;
• 
more detail of representative herd structures, growih rates, labour and feed requirements
3
J
ja
J
( male and female), and off take rales for the main species;
•           poultry breeds, management, diseases and prospects;
•           views of the livestock owners, male and female, as to the constraints they face* their
aspirations and their perception of the development opportunities available to them;
•           quantification of agricultural residues and by-products for livestock feed by wroreda
(TAMS-ULG have already collected data on crop production from some woredas);
•           further assessment of livestock trading movements and migration patterns;
•           prices of livestock producis at retail and wholesale levels, including Addis Ababa and 
retail costs of inputs eg drugs, vaccines etc.;
•             details  of  the  existing  facilities  for  livestock  slaughtering,  marketing,  processing  and distribution,   although   ARDCO-GEOSERV   provide   some   information   for   the   upper basin;
TAMS-ULG haro-akobo river rasin Integra i to development master plan
I
2-28AGRICULTURE
numbers of staff by woreda and zone, who are concerned with livestock production, veterinary care, and hides and skins technology;
details uf the veterinary' clinics and other livestock facilities at woreda, zonal and regional levels; and
♦ information on relevant livestock research activities.
23.4 Phase II Activities
l“he following paragraphs indicate in outline some of the activities which TAMS-ULG w'ould undertake during Phase II as part of the master planning exercise:
•  review'  and  integration  of  information  collected  (sec  section  2.3,3)  between  October  1995 and  January  1996  and  the  preparation  of  representative  livestock  enterprise  models, illustrating typical livestock practice and size of unit, costs and returns, labour use etc. for use in land evaluation and in assessing alternative livestock development options;
•            assess  the  options  for  livestock  development  in  both  the  upper  and  lower  basins; including  tile  development  of  cattle,  sheep,  goats  and  game  ranching  (in  consultation with  the  wild  life  specialist)  and  the  improvement  of  sedentary  integrated  farming systems of the upper basin (in consultation with the agronomist);
•          make proposals for the control of trypanosomiasis, tsetse, and other diseases;
• 
assess the need for livestock support services including water supplies (in consultation 
with the water supply specialist), animal health controls, veterinary care etc,;
• 
prepare project profiles for priority projects; and
• 
finalising the livestock related G1S overlays.
T AMS’LTG baro-akobo river basin integrated development masterplan                 1-19Table 2 1 T Crop Areas - Upper and Lower Basins - ha
. — Uppper ------------------------------------------ ------------------ Lower Total
Cereals
Zone
rilubabor
VVollega
Katfa
Ass-osa
Total
Gambela
Basm
Maize
75.213
60,196
35 464
12.337
183.210
Sorgnum
18.497
31.521
5.250
25.290
80,558
Teff
53,774
30,669
7.326
0,301
100.070
4,100     187,310 11,860       92,418
17.333
100,070
Barley
4_754
5,333
5.109
2.137
17,333
Wheal
2,042
1.638
1.095
1.583
5 350
6.358
Mill cl
3.DO0
6.169
1,236
58
10,471
10,471
Rice
483            4B3
Subtotal
157.200
135,526
55.480
49,706
390,000
IB. 443     414.443
Pulses
Horse bean            6.421         2.334         2,730            241         11.726                           11726
Field pea
6.151         1,627         1.550            105          9.451                            9 451
Haricot bean           2,072        3,674            939         1 020           8.505                            8 505
Chickpea
1.192         2.075                              487          3.664                            3 664
Soyabean 139 139 526 765
Cowpea 26 6 32 32
Subtotal Oilseeds-
16,582         9,849         5 233         1 B53          33.517            626        34,143
Sunflower
43
43
43
'Niger Jnoug)
1 994
1.506
5.405
8,905
8,905
Unseed < tai ba}
1.091
127
60
90
1,376
1.376
Sesame
727
318
2
264
1.311
562         1,073
Groundnut
1
29
30
209            319
Rapeseed
331
192
20
551
551
Subtola’
4.105
2.143
71
5.816
12.216
851        13.067
Other Crops
Coffee
29.667
56,679
3.126
11
09,473
09,473
Fruit
210
19
69
298
290
Vegetables
1
431
432
432
Sugarcane
275
11
206
206
Chilli pepper
1.607
-96
280
2.091                            2,091
Cotton
131            131
Subtotal
31.750
56 905
3,126
799
92.580            131       92 711
Total
209.806
204.423
63,910
58,174
536 313       W.051      554 364
Source- Lower basin - SEKHOZPROMEXPORT. 1990 data Tor 1987
Upper basic. ARDCOGEOSERV. 1995. data for 1992/93
Note Administrative boundary for Gam beta. Awraja in 1987 does not coincide with boundaries of new administrative regions Total crop areas could be g eater than
r
shown
Total area of basin approx 7.500 300 na; thus 7.4% is currently cultivatedTable 2.1 2 Cropping Patterns by Zone - % Cropped Area
Crop
- —------ - =
1
UjJjJsri
Lower
Total
Zone
lllubaboi Wollega
Kaffa
Assess
Total
Gambeta
Basin
Cereals
Maize
36
29
55
21
34
23
34
Sorghum
9
15
8
43
15
66
17
Teff
26
15
11
14
19
16
Barley
2
3
0
4
3
3
Wheat
1
1
2
3
1
1
Millet
1
3
2
0
2
2
Rice
3
0
Subtotal
75
66
67
85
74
91
75
Pulses
Horse bean
3
1
4
0
2
2
Field pea
3
1
2
□
2
2
Haricot bean
1
2
1
2
2
2
Chickpea
1
1
1
1
1
Soyabean
0
0
3
0
Ccwpea
0
0
0
0
Subtotal
6
5
0
3
6
3
6
Oilseeds:
Sunflower
0
0
O
Niger (nougi
1
1
9
2
2
Linseed (talbai
1
a
0
0
0
Sesame
0
0
0
0
0
0
3
0
Groundnut
0
D
0
Rapeseed
2
0
0
0
0
0
□
Subtotal
2
1
0
10
2
5
2
Other crops
Coffee
14
20
5
0
17
Fruit
0
0
16
0
0
Vegetables
0
0
1
□
a
Sugarcane
0
0
0
Chilli pepper
0
1
0
0
0
Cotton
o
1
0
Subtotal
15
29
5
1
17
1
17
Total
100
100
1DD
100
1CO
100
100
Source Lower basin - SEKHOZPROMEXPORT 1990 data for 1987
Upper basin - ARDCO-GEOSERV. 1995 data for 1992/93
Note: Administrative boundary for Gambela Awraja in 1997 does not coincide with boundaries ol new administrative regions Total crop areas could be greater than shown
Total area of basin approx 7,500.000 ha; thus 7.4% is currently cultivatedTable: 2 1 3 Average Yields of Crops - kg/ha
Crap
-------------—------------------ Upper --------------—--- ----------------------       Lower
lllubabor
Wollega
Kaffa
Assosa
Average
Gam be la
Cereals:
Maize
1.410
1.210
1.250
900
1.279
1,340
Sorghum
1.060
1,000
950
1,200
1.073
504
Teff
690
600
440
500
628
Barley
740
680
800
900
759
Wheat
700
700
700
1.000
775
Millet
700
600
1,000
500
575
Rice
550
Pulses.;
Horse bean
690
640
670
400
669
Field pea
600
680
600
350
608
Haricot bean
520
700
810
550
633
Chickpea
700
680
400
649
Soyabean
90
Cowpea
700
n.c
Oilseeds:
Sunflower
300
300
Niger (noug)
450
430
450
447
Linseed (lalba)
430
480
300
300
420
Sesame
300
380
400
300
320
290
Groundnirl
500
500
400
403
1,090
Rapeseed
400
480
600
400
428
Other Crops:
Coffee
500
560
480
200
537
Fruit
14,000
14,000
14,000
14,000
Vegetables
6.500
6,500
6,500
Sugarcane
40.000
40.000
40,000
Chilli pepper
1,607
198
288
1,293
Cotton
450
Source: Lower basin - SEKHOZPROMEXPORT. 1990: data for 1987 Upper basin - ARDCO-GEOSERV, 1995; dal a for 1992/93Table 2 14 Total Production of Main Crops - mt
■   ■            ■            I          — -         ---------------      Lower        Total
Zone
Mlubabor
Wollega
Kaffa
Assosa
Total
Gambela
Basin
Cereals
Maize
106.050
72,837
44 330
11.103
234,321
5,494
239,815
Sorghum
19,607
31,521
4986
30,348
86.463
5 977
92.441
Teff
37.104
18.401
3,223
4.151
62.079
62,879
Barley
3.518
3.626
4.087
1,923
13,155
13,155
Wheat
1.429
1.147
767
1,583
4,926
4,926
Millet
2.106
3.701
1.236
29
7,072
7.072
Rice
266
266
Subtotal
169.814
131,234
58,631
49,137
408.816
11,737
420,553
Pulses.
Horse bean
4.430
1,494
1.829
96
7.850
7 050
Field pea
3 697
1,074
935
37
5.742
5.742
Haricot bear
1,493
2,572
761
561
5,307
5,387
Chickpea
771
1.411
195
2,377
2.377
Soyabean
56
56
Cowpea
4
4
4
Subtotal
10,392
6,550
3.529
809
21,360
56
21.415
Oilseeds
Sunflower
13
13
13
Niger (ncug)
897
648
2.432
3.977
3 977
Linseed (taiba)
469
61
20
27
577
577
Sesame
218
121
1
79
419
163
5R2
Groundnut
1
12
12
315
327
Rapeseed
132
92
11
236
236
Subtotal
1,730
922
22
2.561
5234
478
5.712
Other Crops
Coffee
14.829
31,740
1.SOO
2
48,071
Fruit
2,940
266
966
48.071
4.172
Vegetables
7
4.172
2.802
Sugarcane
11,000
440
2,808
11,440
Chilli pepper
2.582
38
2.808
11.440
83
2.704
Cotton
2.704
59
59
Source: Lower bas,n - SEKHOZPROW EXPORT, 1990 data for 1987
Upper basin - ARDCO-GEOSERV. 1995: data for 1992/93
Note: Administrative boundary for Gambela Awraja in 1987 does not coincide with
boundaries or new administrative regions Total crop areas could be greats* than shownTable 2.2.1: Grade of Hides  and Skins  Sent
to the Central Market from Bedele 1991-93
Cattle Hides
1991
1992
1993
% Grade I
70
70
92
% Grade II and III
30
30
8
Total
2.127
2.426
3 068
Sheep Skins
Sailed
17,596
19.628
10,505
Air Dried
123
2.704
11,774
Total
17,719
22,332
22.279
Goal Skins
Safted
3.299
1.129
1.037
Air Dried
2,712
7.198
3.125
Total
6011
B.327
4,162L [ L L I L l: e
E
L
L
I L
( r
rSECTION 3
WATER RESOURCES3.          WATER RESOURCES
3.1       Climatology' and Hydrology
3.1.1    General
Each   of   lhe   reports,   the   Russian   Sludy   and   the   ARDCO-GEOSERV   Study   have   presented annexes dedicated to the climatology and the hydrology of the Baro-Akobo river basin: Aux
I - Climatology and Hydrology for the Russian Sludy and. Volume IIA - Climate and Volume
II B - Surface Water Hydrology for the ARDCO-GEOSERV Study
These reports present die data collected  and  used  to  different extents,  For example the Russian study  does not include actual rainfall or runoff records,  but presents monthly  series of actual and synthetic runoff at nine gauging  stations and  dam sites for the period  from 1928  to  1987.  The ARDCO-GEOSERV report includes all monthly  meteorological data as well as mean  monthly flows al the gauging stations used for the sludy.
Similarly  the MWR provided  TAMS-ULG with  a set of meteorological and  climatic data that have been  incorporated  in  a computerized  database.  The purpose of this section  is to  review these data and to attempt to clarify inconsistancies.
3.1.2 Climatology
Annual and Monthly Precipitation
The Russian  Sludy  refers to  23  meteorological stations.  None of these records arc reported  in  the annexes.  The periods of records for these stations were compared  with  those of the 48  stations made available to  TAMS-ULG by  MWR which  correspond  to  Lhose included  in  lhe ARDCO- GEOSERV Study  The locations of the 48  sations are shown  on  Figure 3-1-1,  and  the periods of record are summarized in Table 3-1-1.
Three  of  these  23  stations  refered  to  by  the  Russian  Study  are  located  in  Sudan,  and  it  is  not possible to obtain these records at this time.
The periods of records,  locations and  elevations of the following  nine stations arc approximately the same for the two  studies: Jikawo,  Mettu,  Yavu,  Pokwo.  Tepi,  Dembi-Dollo,  Hurumu,  Bure and  Yubdo.  However,  a number of discrepencics were noted  on  the other stations; these are identified in lhe following paragraphs
The two  major precipitation  stations used  as reference by  the Russian  Study  are those located  at Gambella and  Gore for which  it indicates rainfall records of 81  years and  45  years respectively. Presently  the TAMS-ULG flics consists of 18  years of records for Gambela.  from 1971  to  1988, instead  of a period  from 1905  to  1987  indicated  by  the Russian  Study  and  19  years of records from I960  to  1965  and  from 1973  to  1988  reported  by  ADRCO-GEOSERV.  For the Gore station,  TAMS-ULG files consists of 37  years from 1952  to  1988  as it is also  indicated  in  lhe
.ARDCO-GEOSERV report, instead of 41 years over two periods front 190S io ] 936 and from
TAMS-ULG      baro-akobo river basin integrated development master plan
5-11 WATER RESOURCES
1971 to 1986 in the Russian Study.
Three stations Bonga,  Darn  and  Tata have been  installed  for the purpose of the Russian  Study and have records varying from three to five years (1983 - 1988): lhe records tor these periods are not available and  it appears that these stations have not been  maintained.  They  have not been used by ARDCO-GEOSERV.
One  station  at  Gcteha  is  shown  in  the  Russian  Study  as  having  9  years  of  records  from  1955  to 1963 but is not reported in TAMS-ULG files.
Al the bcdcllc station, the Russian Study indicates records from 1972 to 198! instead of six years from 1987 to 1992 in TAMS-ULG files, and 15 years of records from 1971 to 1986 as indicated by the ARDCO-GEOSERV Study.
At  the  Metu  station,  the  Russian  study  indicates  record  from  1972  to  1977  instead  of  1967  to 1977 and from 1985 Io 1986 instead of records from 1981 to 1988 in TAMS-ULG files.
At  Chora,  lhe  records  are  shown  to  exist  from  1973  to  1982  instead  of  1968  to  1970  and  1980 to 1989 in TAMS-ULG files and 16 years from 1973 to 1988 in .ARDCO-GEOSERV Study.
At  Abobo,  lhe  Russian  Study  shows  record  from  1976  to  1982  and  1985-86  instead  of  scattered records from 1982 io 1988 in TAMS-ULG files and in ARDCO-GEOSERV records.
Al the ltang  station,  lhe Russian  Study  indicates records from 1973  to  1976  and  from 1979  to 1986  instead  of records from 1973  to  1982  and  1985  in  TAMS-ULG files,  and  in  ARDCO- GEOSERV report.
At the Alem TcTeri station, the Russian Study shows the records to be from 1973 to 1982 and 
from 1985 to 1986 instead of a continuous period from 1973 to 1989 in the TAMS-ULG files and 
in lhe ARDCO-GEOSERV report.
Most of these discrepcncics can  be the results of shifting  years of record  to  artificially  obtain continuous records,  or or the inadvertent incorporation  of synthetic data into  a set of actual records The raw- data are used  to  establish  correlation  between  stations to  infill missing  data,  or to  evaluate rainfall-runoff relationship  and  to  extend  runoff records based  on  their correlation with  precipitation.  The fact that the periods of records are erroneous will probably  result in  poor correlation and can lead to the incorrect analyses and conclusions.
These  discrepencies  must  be  clarified  prior  to  initiating  the  process  of  infilling  or  extension  of the data.
Maximum Daily Precipitation
There is very  little mention  of maximum daily  precipitation  in  lhe Russian  Study  which  slates: "The greatest daily depth of rainfall for the last 15 years at Gambela station was recorded on June 21, 1975, it reached 181 mm." The ARDCO-GEOSERV does not report any data on the subject.
TAMS-ULG    u aro-akobo hi ver a a.stn integrated development mas ter pl an       1-23.          WATER RESOURCES
Temperature
As previously  indicated  the Russian  Study  docs not report the data collected  in  detail.  The ARDCO-GEOSERV study  however presents the available in  a tabular form.  These data arc available at 36 meteorological stations.
Potential Evapotranspiration
The studies indicate that there is no  observed  evaporation  data for the study  area.  The Russian and  the ARDCO-GEOSERV studies computed  monthly  evapotranspiration  (PET) using measured  temperature,  humidity  and  wind  data and  tables for radiation,  length  of the day.  1  lie ARDCO-GEOSERV derived the PET by use of the Penman methodology.
3.1.3 Hydrology
Monthly flows
Tlie flow gauging  stations reported  by  the Russian  Study  is corrected  from 7  stations established by  MWR and  9  stations established  by  the Russians.  The analysis performed  is based  on  the hydrological data collected  at twelve of the gauging  stations established  by  MWR and  three additional stations established  for the sole purpose of the study.  The major component of the data is the gauging station at Gambella which is reported to have records available from 1906.
The source of these records is only  referred  to  as the TAMS report dated  1977.  It is obvious that TAMS is not the originator of the raw data,  however it is likely  that these data have been processed  to  some extent A brief review of TAMS report did  not provide tlic source of die raw data nor did  it provide an  explanation  of whether or not the data have been  analyzed  or even synthetically  generated.  It is likely  that some infilling  has been  performed: further detailed review of this report and  any  other source that can  be found  will be necessary  in  Phase II of this project.  The Russian  Study  adopted  the recommendation  of 7  AMS report to  consider the data after 1928  only,  since those prior to  1928  were judged  unreliable.  Although  it probably  did  not affect their study  it is troublesome that the ARDCO-GEOSERV study  mentioned  the Gambella station as having records from 1967 to 1989.
Based  on  the  Gambella  station  and  the  twelve  other  stations  in  tlic  basin,  the  Russian  established two  correlations  by  linear  regression  (using  the  logarithm  of  the  parameters)  to  define  the  runoff volume   at   sites   of   interests.   The   first   relationship   was   established   between   the   mean   monthly runott  q  in  l/sec  per  knr  and  the  average  elevation  of  the  catchment  area  H  in  km;  the  second  is between  the  drainage  area  A  in  km*  and  the  coefficient  of  variation  of  tlic  monthly  Hows  C These equations are:
q = 8.0H17i and. C„ = 0.51 - 0.08 Lg(A+l)
The Russian Study approach to define the hydrology of the Baro-Akobo basin is based on these equations a comp‘d statistically description of the monthly runoff at each site can be
d t *nnined and used to size reservoirs as well ns compute projected energy generation, and
TAMS’L LG      HARO-AKOBO RIVER B ASIN INTEGRATED DEVELOPMENT MASTER PLAN
3-13. WATER RESOURCES
availability of irrigation water,
The ARDCO-GEOSERV report refers to river flow data at 16 stations as summcrized in table 3-1-2. The extent of data available at these stations is shown in table 3-1-3 and the locations of the stations in figure 3-1-2,
The ARDCO-GEOSERV Study utilized a different approach to generate monthly flow data at the project sites. This consisted in generating monthly lime series using regionalized parameters such as dimensionless monthly flow profiles (percent of the mean annual runoff), monthly flow coefficient of variation and coefficient of correlation between successive months and a random number generator. The methodology is referred to as the Thomas and Feiring monthly model. The regional parameter were established using gauging stations in the basin as well as some in adjacent basins,  i.e.,  the Blue Nile and  Omo  Ghibe.  The study  has been  performed  for three different size catchments based on the mean annual discharge.
The primary input to the model generating the random sequences of monthly flows is the mean annual runoff,  Using  the Geographical Information  System (GIS) a relationship  between  the mean annual rainfall P in mm and the altitude H of the rainfall station at .32 rainfall stations in the Baro-Akobo basin, was obtained by linear regression, as follows;
P = (0.60 t C ) * (760 + 055 H).
r
The coefficient C
(  is a calibration coefficient which has been mapped in the GIS for individual
rainfall gauging stations.
Based on the records of five river gauging stations in the Baro-Akoho basin (the Baro River at Gambella, the Gilo River near Pugnido. the Alwero River at Abode, the Sor River al Mettu and the Keto River near Chunks), the following relation between mean annual runoff R in mm and the mean annual precipitation at each grid cell of the basin (approximately 21 km ) was established;
2
R - 0.60 * C  ♦ P.
f
On the GIS. the basin is represented by ahout 2,000 grid cells. Although not explicitly slated, the runoff values arc presumably integrated over the entire catchment al the point of interest
The ARDCO-GEOSERV study  also  made an  attempt to  establish  flow duration  curve.  It is unclear what basic data (either monthly or daily flows, and from which stations) were used to establish regional parametric duration curve. The following table was generated;
TAMS-UL.G BARO-AKOBO RIVER. BASIN INfECjRArED DEVELOPMENT MASTER Pt.anwater resources
Percent of the
Time Flow is
Exceeded
10 20 30 40 50 60 70 75 80 85 90 95 99
Ratio to the
Mean Annual
Discharge
2.14 1.58 1.07 069 0.44 0.28 0.18 0.15 0.12 0.10 0.09 0.07 0.04
The   flow   duration   curves   generated   were   used   to   assess   the   hydro-electric   potential   of   the run-of-the river sites.
Flood Analysis
Since the available data arc limited,  the ARDCO-GEOSERV Study  took  a simplified approach  for the purpose of plannjng.  Regional observations for a number of stations within the Baro-Akobo  basin  as well as in  adjacent basins have shown  that the mean  annual maximum daily  flow in  m’/sec is approximately  equal to  five times the mean  annual flow in in^sec.  It was further assumed  that the annual maximum daily  flow is approximately  equal to the maximum instantaneous discharge,  which  is only  valid  for a sizeable catchment area. Having  estimated  the mean  annual flood  based  the mean  annual runoff at a given  site,  a probability  distribution  such  as the Extreme Value Type 1  (Gumbel) or the Log-log distribution, is then assumed to determined the peak discharge for a set return periods.
The Russian  Study  uses a different approach; having  a fairly  long  period  of records for the Gambella station,  it was apparently  possible to  perform a flood  frequency  analysis on  flood events in  the lower basin  for various duration  up  to  several months.  It is not known  if daily flows were used  but assuming  that it is the case,  the methodology  is perfectly  correct.  It is therefore surprising  that some of the extreme floods used  in  the sizing  of the projecl appear small compared  with  standard  as the Myers coefficient.  It is however,  possible that the peak discharges be underestimated  but that the volume of flood  assumed  be more correct given  the size of the catchment at Gambella and the methodology used.
For  the  purpose  of  comparison  the  table  3-1-4  shows  the  extreme  flood  used  in  both  of  the studies at each of the dam site considered,
T'KMS-Ul G BARO-AXOBO fUVFR BASIN integrated development masterplan3. WATER RESOURCES
Both  studies,  the Russian  and  ARDCO-GEOSERV have developed  the hydrology  of the Baro- Akobo  basin  with  a set of data which  has since been  extended,  ARDCO-GEOSERV did  not use the complete record  of the only  long  lime series without providing  an  explanation  and  docs not appear to  have reviewed  the work  performed  by  the Russian  Study.  It is important al this stage to  generate a consistent hydrology  for the lower and  upper BarcvAkobo  basin. For these reasons, an extensive work effort is needed to update the hydrological studies. The proposed scope of this work is set out in volume 1. section 5.
3.2      Ground water Resources
3.2.1   Introduction
Project Location and Area
The area of interest includes some 74020  sq,  km in  the west central part of Ethiopia.  It is bounded  on  the west by  the Ethiopia-Sudan,  border and  on  the north  east and  south  by  a irregularly  shaped  drainage divided  that follows through  mountainous terrain.  All surface flows in  the project area are tributary  to  either the Barn  River or Akobo  River,  both  of which  form a large portion  of the Ethiopia-Sudan  border.  These rivers,  via the Pibor River,  eventually  join then flow into Sudan as the Sobat River.
Previous  studies  of  the  project  area  divided  it  into  an  Upper  and  Lower  Basin  on  the  basis  of generally planner and upland surfaces below and above about elevation 500 meters respectively.
A large part of the Baro-Akobo  Basin  can  be traversed  on  all weather roads.  However,  much  of it is still inaccessible even  with  four-wheel drive vehicles,  particularly  in  the western  sectors during the wet season.
Scope and O/y ecftve
The general scope of this section  is to  review and  analyze existing  information  salient to  the basin  area,  then  provide recommendations for further activities to  promote better understanding of the basin and, if appropriate formulation ot an implementation program for project area ground waler development,
The major objective of this review is to  provide an overview of the area ground water resources that can  be integrated  with  other resources information  in  order to  design  a systematic regional development program.
3,2.2 General Geographic Description
The project area is irregular in  shape and  generally  contained  within  the map  coordinates of 5” 30  and  10  45  North  and  33  0  and  36  15  East.  Regionally,  the land  surface is characterized by  mountainous terrain  in  the eastern  two  thirds of the area and  a gently  westward  sloping planner surface, the Gambela Plain in the western one third.
T AMS* U LGBARO-AKOBO RIVER El ASIN INTEGRATED OF. VELGPMENT MASTER PLAN                3-A3,         WATER RESOURCES
Maximum dimensions of the project area are about 600 km north-south and 36  km east-west. 
r
'Hie total area included in the project is about 74,020 sq.km.
The major population centers in the project area include Gcmbela. Mizan Tefn and Gore. 
Numerous smaller villages are also located throughout the area. Total population in the Baro-
Akobo catchment is approximately 2,000.000.
3.2,3   General Hydrologic Features
Climate and  surface water conditions arc closely  correlated  in  the Baro-Akobo  basin.  Together they define the various hydrologic features in the project ana. Steep gradient streams originating in  eastern  highland  sectors,  with  high  rainfall,  debauch  onto  western  plains areas that have relatively  low rainfall and  moderate to  low river gradients.  Climate types range narrowly  from tropical rainy  to  warm temperate rainy.  Average annual precipitation  (all rainfall) ranges from about 600 mm in the lowlands to over 2600 mm in the highlands of the project area.
Climate and precipitation
Two  types  of  Climate  occur  These  climates  are  characterized  as  tropical  rainy  (type  A  )  and warm temperate rainy (type C).
ITte tropical rainy climate can be divided into warm temperature rainy, with dry months in winter (monsoon and upland savanna) and warm temperate rainy climate having more soil moisture than the warm temperate rainy climate.
About 65% of the basin is within the type A tropica! rainy zone The remaining 35% of the area is characterized by the type C warm temperate climate The distribution in time and space of the rainfall.  and  the extent and  density  of vegetation  and  soil cover,  contribute to  reduce potential evapotranspiration and increase infiltration to ground waler.
Mean monthly maximum temperatures show significant variations within the project area. These temperatures range from below 22’C,  in  (he highlands around  Kombolcha (Wollega) to  about
40 C, 
D         on  the  lowlands  of  Gambela  around  Akobo.  Maximum  seasonal
temperatures  in  the
highlands rarely exceed 25°C, whereas in the lowlands plains the temperature generally exceeds 36’ C during  the hotter months of January  to  April.  The mean  monthly  minimum temperatures also  have significant variations,  They  range from 14’ 168C in  the lowlands of Illubabor and western  Wollega,  respectively,  to  below 10" C in  isolated  locations in  the highlands during November-February'’
Potential evapotranspiration, as may be expected, is lowest over the highlands, then increases progressively towards and onto the Gambela lowlands. Form example, Gore (2130 masl) has a
total ET of 1263 mm/yr while Jikawo (410 mash has a total of 1545 mm yr.
The Bare-Akobo Basin has a mono-model ot single period rainfall pattern i.e.. the rainfall occurs over a continuous period  of time,  but is dominated  by  a single rainfall peak.  The seasonal rainv period decreases in duration from south to north. Thus in Keffa, the wet period ranges form
T AMS-ULG   n aro-akobo river basin integrated development master plan3-       WATER RESOURCES
V cbru  ary/March  to  Octobcr/Novcmbcr while in  the north  western  pints of the Illubabor and Wollcga regions it includes the period  from April/May  to  October-November.  I"hc highest average annual rainfall (>2000  mm) is recorded in Keffa around the Masha and Mizan Tefcri areas In contrast, the lowland areas around I tang, Pokwo and Kurmuk receive less than <1000 mm.
A comparison of meteorological parameters shows that altitude directly correlates with rainfall and  potential evapotranspiration.  Rainfall exceeds evapotranspiration  above 1150  mash  The region  above this altitude,  which  is described  as a zone of waler surplus for ground  water recharge and surface nrn-off, measures 32,682 square kilometers.
Table 3-2-1 is a summary of representative climate data within the Baro-Akobo Basin, f igure
3-2-1 depicts contours of equal average precipitation throughout lhe project area.
Sur/ace Fkater Coflrfrions
The annual surface waler runoff of the Baro-Akobo  Basin  is predominantly  influenced  by rainfall The basin is characterized by seasonal fluctuation of the runofTmuch like other basins in Ethiopia. In the subject basin, two seasons can be observed by the quantities of flows in the rivers. Such flows are closely related to seasonal fluctuations of rainfall Major rivers within the basin arc the Bare and its tributaries (Birbir. Geba and Sor5 the Alwero, the Gilo and the Akobo. The general direction  of river flow is from east to  west.  The major part of the river flows originate in (he eastern highlands al elevations of 1150 to 3000 mask Most regional streams flow to the Gnmbela Plain much of which lies below 600 mask
Stream flow in the basin generally increases from May through September and decreases from October to April. Hows reach their maximums at the end of the wet season in September, then steadily decrease. Minimum flows occur in March or April and, in general, the highest flow period lasts from June to October. During this period lhe rivers of the project area carry- up to 80% of their total annual flows. Surface stream (lows between seasonal high land low water periods are fairly consistent The greater part of annual surface runoff occurs in September while lhe lowest flow season is from January through April
Generally, lhe mean annual rainfall, potential evapotranspiration and average annual run-off in the study area correlate well, in quantity, with ground surface elevation
3.2.4      General Geologic Features
ftegjorca/ Genmorphic Ctaracferittfcj
The major morphological features of the Baro-Akobo Rivers catchment area from east to west arc high mountains, high plateaux, and lowland ranges and plains. Volcano-tectonic, 
metamorphic and epcirogenetic processes of regional dimensions have produced these features.
The high mountain ranges in coalescence, form the major surface water divides in the project area They are mostly represented by weathered Tertian basalt capped in places by resistant
T AMS-U LG  UARCVAKGiKJ RIV kit I*AS|N OMTEGkATEII DE VELC)|»MFN r MASTER PLAN3.          WATER RESOURCES
quaternary silicic volcanic rocks. Elevations in these high ranges vary from 2400 lo 3300 masl.
Hie high  plateaux  with  table land  and  denudational landforms are comprised  of basalt and granites.  Elevations therein  range from 1300-2300  masl,  and  in  general descend  gently  from the northeast toward  the south  and  west.  Next in  the westward  direction  arc the crystalline basement mountains,  known  as the Masengo  ranges,  these ranges have elevations varying  from about 800m to  1,400m.  Hie Masengo  ranges rise through  low undulating  plains,  then  gradually  plunge north and  south  to  elevations down  to  about 550m.  The Gambela Plain  is located  west of the Masengo ranges.  Il is an  extensive plain  that falls from about 475m at Gambela in  the cast to  400  masl at the Ethiopian-Sudan border.
At the northern  extreme of the Upper Baro-Akobo  Basin  are the Assosa mountains.  Elevations ill these mountains range to  almost 1000  meters masl.  The mountains and  the area south  of them, to  the edge of the Gambela Plain,  are mainly  Precabrian  age and  formed  by  granites; the Adda Formation  group  of amphibole,  chlorites,  talc,  sandstone,  greenstones and  quartzites,  and  the Ashange group of lower tertiary' basalt, tuffs and rhyolites.
The area south-east of the Gambela Plain  rises to  betw een  elevations 100  to  200  masl This area is under-hin  by  a suite of rocks similar to  that of die sector neither of the Gambela Plan.  In  this area occurrences of undifferentiated  Quaternary  alluvium finger into  and  fill small valleys and plains that extend eastward into-older upland formations.
Figure  3-2-2  depicts  the  general  location  and  distribution  of  the  geologic  formations  and  rock types in the project area, as described above.
(reologic Formation Descriptions
Rocks in  the Baro-Akobo  project area includes three general types From oldest to  youngest they are the Precambrian,  meta igneous and  metasedimentaiy  (Pi),  the lL-rtiary  igneous (Ti),  and  the
Undifferentiated  Quaternary  (Q).  The  relative  distribution  of  these  rocks  is  shown  on  Figure  4. More description of the rocks comprising the project area is presented in following text.
1. Precambrian Metaigneous and Metasedimentary Rocks (Pi).
Precambrian  rocks outcrop  south  of Lhc Gurafcrda Plateau  in  the Akobo  River Valley  and  extend  north  wards following  the Masengo  Ranges to  the extreme northern  part of the project area.  These rocks are generally  dominated  by northwesl trending  structures,  particularly  in  the area south  of the Gurafcrda Plateau.  Such  structural trends,  including  fold  axes and  foliation,  persist in  a north  and  northeast direction  from about latitude 6"45; North  According  to  the ARDCO-GEOSERV report,  these rocks have been  classified  by  previous workers as the Akobo,  the Surma,  and  the Hamar Domains,  all of which  are based  on geostructurc.  The Akobo  Domain  consists of at least 12  mappable varieties of metavolcanic and  metasedimentaiy  formations.  The Surma Domain  comprises strongly  lineated  and  layered  gneiss and  granitoid  rocks unit encompassing tectonic metaigneous, crystalline and melasedimentary rock types.
T AAlS UI G BAfM>AKO»O river basin integrated development master Pl AN3.         WATER resources
2 Tertian' Igneous Rocks (Ti)
Tertian'  Igneous Rock  formations exposed  in  the project area include the Akobo Basalt (the oldest Tertiary  volcanic yet found  in  Ethiopia),  the 1  Jndividcd  Pre Rift Volcanic succession  of basalt flows,  rhyolite,  trachyte and  ignimbrite,  the Maji Flood Basalt and Stratoid Volcanics, and the Maji Silicics.
Other Tertiary  volcanic formations include the Makonen  Basalt,  Surma Basalt, Demhidolo  Trachytes,  and  Hypabyssal Intrusive Rocks.  The rock  units named above are located,  for the most part,  in  the mountains and  plateaux  in  the Upper Basin of the project area.
3. Undifferentiated Quaternary Rocks (Q)
Quaternary'  rock  formations  exposed  in  the  project  area  include  the  Holocene  Tepi Basalt and Holocene-Pleistocene Undifferentiated Alluvium.
Tepi   formation   flows   and   pyroclastic   features   are   scattered   throughout   the southeastern sector of the project area.
The Quaternary  Alluvium Formation  is located,  for the most part,  in  the west central part of the project area and  is comprised  of rock  debris derived  from the older formations in  the Upper Basin  Area.  The Alluvium is generally  comprised of undifferentiated  stream and  lake deposits.  It is considered  that,  to  some degree,  useful quantities of ground  water are contained  within  all of the Precambrian through Quaternary' formations described above.
Table  2  presents  a  detailed  stratigraphic  column  of  the  rocks  and  formations present within the Baro Akobo Project area,
Project Area Stratigraphic and Structural Conditions
The Precambrian metaigneous and meiasedimentarv (Pi). and Tertiary igneous rocks (Ti) are
combined herein to constitute the Basement Complex Aquifer. The general regional extents of
these formations in the project area are shown on Figure 3*2-2.
Precambrian  metamorphic rocks form the regional platform upon  which  all younger formations in  the project area are based.  Subsequent to  placement more than  600  million  years ago.  these rocks were intruded,  deformed,  uplifted  and  eroded.  Tertiary  igneous rocks primarily  volcanic, were then  deposited  in  nonconformity  over the Precambrian  sequence Notable among  the volcanic rock  formations occurring  in  the project area are the Paleocenc-Oligocene-Miocene Ashangi, Group and the Miocene Shield Group,
Since the emplacement of the Shield Group the Pre-Pliocene rocks in the project region have been eroded, transported to the west, and deposited in the Sudan Synclinal Basin, in which the Gambela plain now resides. The deposited rock debris that now constitutes Lhe Pliocene Alwcro
TAMS'ULG HARO-AKOBO RIVER BASIN' INTEGRATED DEVELOPMENT MASTER Pl AN3.          WATER RESOURCES
Formation  that  is  not  exposed  in  the  project  area,  but  is  present  in  die  subsurface  of  die  Gambcla Plain. Table 3-2-2 indicates regional project stratigraphy in some detail.
Post Paleozoic structural activity,  from geologic map  information,  appears to  have been  minimal. Through  out the project area several short minor faults arc shown  to  occur in  the Precambrian formations, a few in die Tertiary volcanics and only fouraie noted to cul Quaternary deposits
3,2.5    Project Area Hydrogeologic Conditions
The geology  of the Haro- Akobo  Basin  directly  affects the ground  water environments therein. Such  environments can  generally  be divided  into,  1) The relatively  unconsolidated  lacustrine and alluvial sediments that form the Gambela Plain  in  the western  project sector and  2) The older consolidated  rock  formations in  the northern,  eastern  and  southern  highlands of the area.  For purposes of this ground  water report the former formation  is referred  to  as undifferentiated Alluvium (Q) and the latter as Basement Complex (BC).
llie Basement Complex  rocks,  in  Pre-tcrtiary  time,  were faulted,  folded  and  otherwise disrupted. During  tertiary  time extrusive volcanic rocks were deposited  in  nonconformity  over the Preexisting  disrupted  rocks and  became a younger part of the basement complex.  Pliocene sandstones and  Quaternary  Alluvium were then  deposited  on  the Basement Complex  in  the certain sector of the project area and into Sudan.
The ground  water environment of the Alluvium is that of an  unconsolidated  porous medium, while that of the Basement Complex  is one of secondary  porosity  which  has resulted  from fracturing crushing and solution during post-depositional periods of structural activity .
Occurrence and Movement of Ground Waler
Ground  water  in  the  project  area  is  derived  from  direct  infiltration  of  rainfall,  percolation  along stream courses and lateral subsurface down-gradient movement of water in aquifers.
In  the Basement Complex  rocks of the project area eastern  highlands,  ground  water is contained largely  within  aquifers formed  in  fracture fissure and  crush  zones of otherwise impermeable consolidated  rocks.  The extent,  permeability,  recharge conditions and  direction  of water movement in  such  aquifers is usually  not possible to  determine unless site specific investigations are conducted.  In  the case of the project highland  area Basement Complex  aquifers it can  be generalized  than  with  the known  annual rainfall,  surface water runoff,  and  relatively  small amount of ground water use, the aquifers will be totally replenished essentially every season.
Ground  water within  the Basement Complex  rock  highland  is usually  contained  within  localized aquifer environments.  However,  ground  waler movement over all is,  in  general from cast to west,  and  encompass the area of some 18,800  km sq.  The alluvial plain  and  environs vary  in elevation  from approximately  475  meters al Gambela to  400  meters at the w estern  extent of the project area,  a distance of about 160  km.  Ihcse dimensions result in  a rather flat average surface gradient of 0.5 m'km from east to west across the plain.
-j-
[ l_rG          HARrt-AKORO RIVER BASIN INTEGRATED DEVELOPMENT SLAV TER PLAN
5-113.         WATER RESOURCES
Ground  water within  the Gamhcla Plain  is derived  primarily  from infiltration  of surface water flows from the numerous streams that cross it.  Secondarily,  direct percolation  oi rainfall and subsurface inflow from Basement Complex  aquifers in  hydraulic continuity  with  die plain Alluvium probably  also  occur.  In  general,  ground  water movement in  the sediments of the Gambcla Plain  is form cast to  west and  its gradient probably  nearly  parallels that of the surface gradient of 0.5  m/km Although  phreatic (unconfined) conditions appear to  prevail continuously throughout the alleviated  area of the plain,  it is probable that semi-confined  and  confined conditions occur locally  in  deeper aquifer (s) beneath  the plain.  This condition  appears to  be confirmed  by  aquifer testing  described  in  the Russian  Report fl990.) wherein  three separate aquifers were identified as
I I The Aquifer in .Alluvial Deposits, Longshore Bars and Holocene Deltalic Sediments.
2) the aquifer in Holocene Lacusirine-AIluvial Deposits and
3) the Regional Aquifer,
The Alluvial Deposits Aquifer I) and  The Holocene lacustrine-Alluvial Depost is 2  ) are phereatic and  frequently  interconnected.  The Regional Aquifer is also  usually  phreatic,  hut is considered  semi-confined  to  confined  in  the vicinity  of Russian  Report lest well designated  Bore Hole 104  This Bore Hole penetrated  a 30m thick  sandstone in  the interval of 90  to  120m depth that had  a tested  transmissivity  of I ZOmVday  and  storativity  of 2.4  xlO'.  It was presumed  that this aquifer represented only part of the Regional Aquifer in the vicinity of Bore Hole 104.
Ground  water moving  down  gradient across the Gambcla Plain  within  phreatic aquifer (s> in  all probability,  reaches and  is effluent to  the Haro.  Akobo  and  Pibor rivers.  It has been  suggested that the confined  Regional Aquifer,  as penetrated  at Bore Hole 104,  may  continue at depth beneath these same rivers into Sudan.
Water Bearing Characteristics of Aquifers
Four aquifer system have been recognized in the project area To summarize they are as follows:
The Basement Complex  Aquifers that are related  to  localize fracture and  crush  zones resulting  from geologic structural disruptions in  consolidate rock.  These aquifers arc found  only  in  the Basement Complex  rocks and  intermontane valleys of the project area highland. In the Russian Report these aquifers were included in the Rcginal, Aquifer.
The  aquifer  in  Alluvial  Deposits.  I  .ongshore  Bars,  and  Holocene  Deltaic  Formation.  This aquifer is herein referred to as the Holocene Alluvial Aquifer.
The  Aquifer  in  Holocene  and  pleistocene  Lacustrine  Alluvial  Deposits,  is  herein  referred to as the quaternary Lacu-Alluvial Aquifer.
The Regional Aquifer This aquifer includes the Pliocene deposits of the Alwero Formation  which  is considered  to  include an  artesian  aquifer penetrated  at Russian  Test Bore 104. This aquifer is termed the Regional Aquifer herein
TAMS-ULG    baro-akobo rivfr basin in tegrated development master i-ijus
3-12WATER RESOURCES
J. The Fosemenr Compter Jgwr/er.
Only  a small amount of information  and/or basic daLa arc available regarding  the Basement complex  Aquifer(s) in  the project area.  Such  information  indicates borehole and  spring  production  to  range from less than  1.0  1/s.  To  more than 
18.0  1  /s.  This  range  in  production,  however,  is  based  on  only  a  few  data  points in   the   large   project   area   underlain   by   Pre-Quatemary   Basement   Complex   rocks. Most  spring  flows  from  the  Basement  Complex  arc  in  the  lower  range  ol  less  than 
1.0 1/s.
Because the Basement Complex  rocks contain  only  aquifers having  secondary, they  do  not lend  themselves to  accurate analyses of water bearing  characteristics, such as transmissivity, storativity etc,, as do those of primary porous media.
Aquifers in  the intermontane valleys of (he Basement Complex  area are most frequently  of limited  area,  depth  and/or capacity  and  therefore can  usually provide for only relatively small water supply demands
2 The Holocene lluvial Aquifer
This aquifer is comprised* for the most part, of holocene Alluvium deposited 
along the alignments of river courses. Such aquifers are generally perched, 
contain a significant quantity of silt and clay, and are in direct hydraulic
contmuiiy with the rivers. Water bearing characteristics of the Holocene Alluvial
Aquifer arc generally poor with reported low permeability (0. Im/day) and well
yields of 0 1 to 1.0 1/s. Water levels in this phreatic aquifer arc usually less than 
5m below ground surface.
3. The Quaternary Lack-Alhn ial Aquifer:
Phis aquifer occurs throughout the Gambela Plain  generally  westerly  of Meridian 3411 East.  The aquifer ranges in  depth  to  about 30m.  Deposition  of this alluvial aquifer has occurred  from Upper Pleistocene throughout Holocene lime.  It is effectively  separated  from the underlying  Regional Aquifer by  lacustrine-clayey deposits,  but is open  to  infiltration  of water from the surface.  Permeability  is reportedly  low,  ranging  from 0.01  to  2.0  rn/day  Well production  is also  indicated to  be low,  on  the order of 0.01  to  0.5  1/s.  Phreatic water levels in  this Aquifer range from ground surface to about 7m depth
4 The Regional Aquifer:
Because of the wide distribution  of the Regional Aquifer,  i.efrom the project area highlands to  the western  extent of the Gambela Plain,  it has several environments of occurrence Consequently  its water bearing  characteristics range broadly.  This aquifer has been  described  in  the Russian  Report to  include water associated with the Basement Complex rocks, but for purposes of this report thewater resources
Regional Aquifer is defined  to  include only  the Pl io-Quaternary*  sedimentary deposits beginning  at the ilubador Plain  and  continuing  beneath  the Gambela Plain  to  and  beyond  the Ethiopia'Sudan  Border,  ["he formation  comprising  this Regional Aquifer is predominantly  (he Pliocene AJwero  Formation  and  Lower to Middle Pleistocene alluvial deposits of the Gambela Plain  Because of the varied sedimentary  environments in  which  the Regional Aquifer occurs,  its waler bearing  characteristics,  naturally,  also  vary  .  Transmissivity  in  the sedimentary formations has been tested to range up to 120 m* /day. while test well production has ranged form 0 01 lo 2.5 Lk Water levels in this Regional Aquifer are known to vary from about 13.8 to 30 Im below' ground surface.
In  general,  geologic conditions and  test information  in  the projet area lend  lo  confirm that although  .Alluvial aquifers are abundantly  present,  they  consistently  have poor to  moderate waler bearing  characteristics.  As a consequence individual wells constructed  therein  would  ordinarily be expected to produce only modest quantities of water from relatively deep pumping levels.
It appears,  however,  that the Regional Aquifer,  comprised  Largely  of lhe Pliocene AJwero Formation,  could  produce significant quantities of water.  Table 3-2-3  lists information  contained in  lhe Russian  Report regarding  aquifer testing  conducted  in  lhe Alwero  Formation  of the Regional Aquifer.  Column  7  of Table 3-2-3  indicates the recommended  individual well production,  to  fully  capture Alwero  Formation  ground  water,  to  range from 3.0  to  20  l/s.  Such recommended  production  may  be possible considering  the lithology,  thickness and  hydraulic characteristics of the Alwero Formation, but this is yet to be proven,
Aquifer Recharge and Discharge
The recharge of aquifers in the project area is by percolation of direct rainfall, infiltration through 
the bottoms and banks of surface streams and down gradient migration of ground water. Aquifer discharges result primarily from effluent flow io surface water courses and evapotranspiration. 
Only a relatively minor quantity of water from springs and wells is known to be consumptively 
used in the project area.
High  seasonal rainfall in  the elevated  areas,  where the Basement Complex  Aquifers) occurs, appear to  completely  recharge ground  water annually.  Surplus waters from mountainous and high  plateau  zones,  move down  gradient as subsurface migration,  surface flow and/or rising water flow.  After fully  recharging  the Basement Complex  Aquifers),  these surplus waters continue westward  and  subsequently  recharge the downstream regional.  Quaternary  lack- Alluvial, and Holocene Alluvial Aquifers.
The fact Lhat recharge and discharge occur is reflected in seasonally changing water levels in all aquifers.  Such  water level variations,  as indicated  in  the Russian  Report data,  arc usually  less than  5  meters,  although  some may  fluctuate more than  13m.  such  as at Bore Hole 102.  The
seasonal  water  level  change  in  a  possibly  regional  artesian  aquifer  penetrated  at  Borehole  104 
was  recorded  to  be  only  0.3Im.  This  small  variation  suggests  low  water  discharge  and/or  use  in the aquifer forebay and restricted down gradient outflow from lhe aquifer.
TAMS-ULG    0ARO-AKQBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN                   3-M3.          WATER RESOURCES
Considering  the quantity  of potential recharge waters and  the relatively  small seasonal fluctuations of water levels in  the project area aquifers,  it is probable that all basin  aquifers arc effectively  recharged  fully  during  each  rainfall season.  Deepest water levels in  wells are indicated to usually occur in April-May and the highest in October-November.
Given  the current paucity  of data in  the project area,  it is not possible to  calculate a valid hydrologic balance of the Baro-Akobo  basins.  From all existing  evidence,  however,  it appears that ground  water recharge potential greatly  exceeds that naturally  occurring  there al present An availability  estimate presented  in  the Russian  Report indicates the potential recharge ol ground water to  be 1284  x  10s mVycar.  This estimate,  as noted  in  the same report,  is only  a general one which is probably excessive.
In  the course of the Russian  investigation  (.1990) and  the ARDCO-GEOSERV project (1995) waler samples were collected  from rivers,  springs,  wells and  test bore holes throughout the Baro- Akobo  project area.  These water samples were analyzed  for inorganic chemical content to determine the chemical characteristics of project area w aters.
Appendix  2.8  of the Russian  Report lists 126  water quality  analyses and  the ARDCO-GEOSERV report describes the results obtained  from 273  sampling  points.  As would  be expected,  water quality form the sampled sources and locations of the large project area vary considerably.
In  the Gambela Plain  vicinity  (the Lower Basin) most surface and  groundwaters arc suitable for many  potable and  irrigation  uses.  Total Dissolved  Solids (TDS) of tested  ground  water ranges from 72  to  955  mg  1.  Of note is the quality  of the water from the artesian  aquifer (Alwero Formation) encountered  tn  Bore Hole 104.  This waler was found  to  be of excellent quality  w ith a JDS of 313  mg/1  A pH of 7  0  and  generally  low in  all constituents,  including  rare elements
in  the Upper Basin,  ARDCO-GEOSERV (1995) sampled  and  analyzed  water from 173  springs. 59  hand  dug  wells,  and  five deep  bore wells.  With  the exception  of a few highly  mineralized springs ranging  to  2307  mg/1  IDS,  most spring  and  well water occurring  in  the Upper Basin  is chemically  well suited  for both  drinking  and  irrigation  purposes.  This,  naturally,  must be confirmed al each source location before the water is put to use,
Il should  be noted  that the discussion  presented  above relates only  to  chemical characteristics of the water samples.  No  information  is presented  herein  relates to  the bacteriological or hazardous constituents content of the waters.
3.2.6   Use and Production uf Existing Springs and Wells
Available information  regarding  the use and  production  of existing  wells in  the project area indicates dial currently  functioning  wells serve primarily  as billagc or individual household  water supplies Most wells are hand dug and seldom reach depths of greater than 20 meters
Present day use of water resources on the Gambela Plain is graphically described in the Russian
TAMS-ULG baROakobo river basin integrated development master flax,
*15water resources
Report as follows:
"Present-Day  Use of Water Resources,  The present day  water consumption  in  the river basins located  in  the area under study  is insignificant and  therefore impossible to  calculate.  The total water consumption  from surface water sources,  ground  water sources,  street taps and  springs makes up  about I 8  - 2.5  mem yearThus,  as can  be seen  from the above,  the water resources of the Baro-Akobo Valley Rivers arc practically in a natural state,..A
Judging  from (he above quote,  die Russian  work  completed  in  1990  appears to indicate that little well production  of ground  water is currently  occurring  within  the area of the Lower Basin, including I he Lambda Plain.
The Upper Basin  Study  of the Baro-Akobo  project area conducted  by  ARDCO-GEOSERV (1995) identified  183  springs,  59  hand  dug  wells and  21„ deep  bore wells.  Developed  springs in the Upper Basin are estimated to produce about 2.0 million mVyear to the local population. Hand dug  wells are estimated  in  the ARDCO-GEOSERV Report to  produce about 675.000  mVyear of
ground  w ater
r
to  the  population  of  the  Upper  Basin.  Machine  bored  wells  (deep  bore  wells)
number  21  in  15  towns  of  the  Upper  Basin.  Production  ranges  from  0.3  to  8.2  1/s  from  wells constructed from 41 Io 185 m in depth.
Annual production  from the 21  Upper basin  deep  bored  wells has been  estimated  by  ARDCO- GEOSERV to  be about 660,00  mVyear.  Total production  from ground  water resources, including springs, hand dug and bored wells is estimated to be about 3,335,000 m  year.
J
Comments in  the ARDCO-GEOSERV Report indicate that most developed  springs,  hand  dug and  bored  wells are inadequately  sealed  or not otherwise protected  and  maintained  in  a sanitary condition.
Table 3-2-4  represents a water supply  conditions survey  of woredas (counties) in  the project area conducted  by  ARDCO-GEOSERV for their 1995  report.  This survey  included  54  woredas and included  information  regarding  major sources of water supply,  town  of supply  location  and population  opinions of waler supply  adequacy  It can  be noted  that,  by  anJ large,  the local populations did not consider their water supplies to be adequate.
Information  regarding  production  in  the Upper and  Lower Basins of the project area indicate that only  a very  small amount of the total ground  waler available is utilized  by  the regional population.  Springs and  wells that are properly  constructed  and  maintained  offer the prospect of safe, healthy and dependable water supplies to effectively all people within the project area.
3,2.7  Future Ground Water Development Potential
All data information  reviewed  and  analysed  in  the course of the preparation  of this evaluation indicates that ground  waler is available in substantial quantity in almost all sectors of the project area.  Such  waler has been  developed  tor local population  water supplies from springs,  hand  dug wells,  and  deep  well bores V»  ater from these sources is used  primarily  for drinking  and  other household  purposes .Although  some irrigation  of private gradens occurs.  only  a few,  significantly 
TAMS*ULG haro-akobo rivfr basin tntegkatf.d development master plan                   3-lt.3.          WATER RESOURCES
large ground water irrigation projects arc known to be presently active.
In  the ARDCO GEOSERV Report of 1995.  Volume Il.C.  Hydrogeology,  section  7.6t Irrigation Water Requirement,  10  locations of self-help  small irrigation  sites were identified.  Such  sites included  a total area of 1100  ha.  Gross potential irrigable area in  the Upper Basin  has been estimated io total some 109,000 ha.
No mention of existing ground water irrigation projects, or the potential for same, was noted in 
the Lower Basin Russian Report of 1990. However, four future potential water well supply areas were defined (Jikawo-Baro, Baro-Alwero. AlweroGilo, and Gilo-Akobo) which reportedly 
should provide bored wells ranging in production from 1.5 to 20.0 1/s.
Water supplies can  be derived  from ground  water contained  within  essentially  all rock  types in the basin,  Although  there appears to  be plentiful waler developable for village and  individual households,  the potential for high  rate ground  wratcr production  for irrigation  of crops is not considered  feasible throughout most of the project area.  This is largely  because of expected  low production  capacities and  low specific capacities form wells constructed  into  low permeability' formations.
Locations for Ground Water Development Potential
There is some ground  water development potential in  effectively  all areas of the Baro-Akobo Basin.  Howcvct. there is only  one vicinity  that has demonstrated  potential for the construction of relatively  shallow (170m) high  capacity  wells (5-101/s) and  having  good  ground  water quality. The test wells identified  in  the Russian  Report as Bore Holes (test wells indicate potential water bearing  zones between  depths of about 60  and  170m in  both  electric and  lithology  logs.  All of these wells arc located  in  the vicinity  of map  coordinates 8*45'  North  and  34BOO'  East.  If agricultural conditions are considered  suitable,  a test-production  well construction  program should be initiated in this vicinity.
Although  currently  available information  indicates that there are few project area locations where high  capacity  irrigation  wells can  be constructed,  because of poor aquifer conditions,  a large element of benefit could  be derived  by  the local population  if dependable and  sanitary  wells are provided  to  village communities If water wells 0,2  to  1.0  1/s capacity'  with  sanitary  construction and  maintenance could  be provided  to  village and  rural dwellers,  the health  of the people and efficiency of the workers in the project area could be advanced significantly.
Proposed Ground Udler Jest-Production Program Jar f/ie Pong Vicinirv
On   bases   ol   information   available   from   previous   studies   of   aquifer   conditions   in   the   Lower Basin,    Regional    Aquifer.    Alwero    Formation,    a    test-production    program    is    recommended    in the   Itang   vicinity.   The   I   tang   vicinity,   as   herein   defined   includes   the   area   contained   within coordinates   8“10'   and   8fl15'   north,   and   34’   10'   and   34“   10'   and   34“20’   east,   kang   is   located about  50km  west  by  road  of  Gambela.  Gambela  is  the  largest  population  center  on  the  upne Gambela Plain and is accessible by road from Addis Ababa mosi of the year.
TAMS-U LG       HARO-AKOB0 RIVER BASTN PmEGRATLD DEVELOPMENT MASTER PLAN3. WATER RESOURCES
An  initial test-production  program in  the Rang  vicinity  could  verify  the water bearing characteristics of the Alwcro  Formation  while providing  water to  a potential irrigation  sector on the upper Gambela Plain.  The specifics of this proposed  project cannot be defined  until an  actual site is selecicd  on  the basis of agricultural suitability,  surface topography,  site accessibility, property availability, and availability of well construction equipment and materials.
At this time,  however,  it is expected  that the maximum depth  of test-product ion  wells would  be 250m,  with  300mm diameter blank  casing  and  well screen  placed  in  500mm diameter bore holes. Gravel envelopes would  be installed  in  the annulus between  the 300mm casing  and  500mm bores of the wells.  Well construction  of the type recommended  would  require medium heavy  drilling equipment.
The cost of wells of such  design  and  construction  would  be on  the order ot 2000  Biir per meter, exclusive of equipment costs Therefore the cost of a 250m deep  well of the type recommended would be about 500,000 Birr.
3.2.B Project Area Ground Water Policy
Discussions with  authorities of the Ministry  of Water Resources Development indicate that there is currently  no  codified  national policy  or legislation  regarding  ground  water rights or control. There is some indication  that draft language for such  a policy  has been  prepared,  but is still in the formative stage and certainly not yet adapted by the Ethiopian Government.
Ground  water policy,  as described  by  authorities,  is presently  formulated  on  a case by  case basis by  the project respective provincial governments.  No  written  policy  upon  which  the provincial government base its ground water development judgements is known to exist.
3,3 FLOOD PROTECTION PROJECTS
33.1   The Flooding Situation
The  Russian  Study  identified  the  two  following  reasons  for  the  flooding  problems  of  the  lower basin, the Gambela Plain area:
•           flooding resulting from overflow of river bank; and,
•            flooding resulting from rainfall and poor drainage.
The flooding  resulting  from the river overflow is staled  as the consequences of wo  possible phenomcma: a river hanklull capacity  too  small and  the backwater effect from the Pibor River and the Sobat River.
The extent of flooding in the Gambela Plain varies from year to year. For each flooding situation the Study  has estimated  the probability  for the land  to  be flooded.  The total area of the lower basin  subject to  repeated  flooding  is 9720  km: The Russian  Study  has estimated  the percentage
□f land flooded to be as follows:
TAMSULG DARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT master planwater resources
Recurrence Intervals
Flooded Area
resulting from
Flooded Area Resulting from
River
Rainfall
Total Flooded Area
2-year
8%
20%
28%
10-year
37%
58%
95%
50-year
100%
-
100%
33.2     Flood Control Potential Benefits
During  the 198R flood  (estimated  lo  be equivalent to  the 50-year flood),  which  was observed  in detail by  die Russian  experts,  the town  of Itang  was entirely  under water and  a considerable part of Gambela was also  inundated  The consequences were quite severe particularly  for the town of Gambella where several administrative buildings,  houses,  hotels were inundated.  The power station  was also  flooded  and  became inoperative for one month.  Severe difficulties followed  with regard to supply of food and drinking water for the people affected by the flooding.
The Russian  Study  states that the most fertile pastures are those located  in  area subject to frequent flooding,  as a consequence this land  is essentially  used  as pastures for cattle 11  also states that the lack  of flooding  makes the land  unusable for perennial pastures.  For that reason, the study  does not put any  economic value on  the damages resulting  from flooding  of pastures. It also states that no capital has been invested in the flooded pastures
There is an attempt to assess the losses resulting from the 1988 flood, by valuating the grain 
distroved: it is estimated as a possible loss of up to 4.0 to 4.5 million birrs.
3.3.3 Proposed Protection
The Russian  Study  proposes two  types of flood  protection; first by  the construction  of dikes, drains and  pumping  stations for the protection  of Gambella and  Itang  for example The protection of Gambella consists in  building  2.4  km of dikes along  the right bank  of the Baro  River and  its right tributary  the Jejebe River with  a five-meter wide road  on  the crest.  Portion  of the dike is replaced  by  a concrete retaining  structure to  accommodate existing  houses.  The dike would  be supplemented  by  1.4  km canal to  divert the runoff from the elevated  unprotected  areas.  Similar diking is proposed for the town of Itang.
he second  approach  proposed  by  (he Russian  Study  consists in  providing  flood  control storage capacities to  the TAMS dam to  store a volume of water such  that all incoming  flood  up  to  the
100-ycar  flood  can  be  control  and  released  at  a  not  exceeding  the  bank  full  capacity  of  the  Baro River of950 mVsec, The storage required at the TAMS dam is 3 95O million cubic meters’.
t
A similar scheme is proposed  on  the Alwrero  R.  using  the Chint reservoir.  This reservoir with  130 million  cubic meters,  however controls only  25% of the catchment area and  cannot store enough to prevent the Alwero River from overflowing under the 100-year event.
TAMS-U LG       BARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN1 WATER RESOURCES
3
Finally  on  the Gilo  River,  the Gilo-1  reservoir is considered  as a flood  storage providing  a capacity  of 1,460  million  cubic meters to  ensure a release which  does not exceed  the river capacity downstream of 280 m /scc during the 100-ycar event
The recommendations of the report arc extremely  unclear with  regards to  the benefit of flood control.  While the Study  indicates that the flow can  be controlled  in  some cases,  and recommends that further studies be undertaken, to assess the advisability ot flood control.
3.4 Irrigation Projects
3,4.1 Overview
A review of the previous studies was carried out, including:
For the Lower Basin,  the Studies carried  out by  Selkhozpromexport (Russian  Study) in 1983-1990,  as published  in  the Baro-Akobo  Master Plan  Study  of Waler and  Land Resources of the Gambela Plain,  Final Report Draft,  1990  (Volumes II,  Main  Report,  VII (Annex  7),  Irrigation,  X (Annex  16).  Infrastructure and  Social Amenities and  XII, Drawings
For  the  Upper  Basin,  the  Survey  and  Analysis  of  the  Baro-Akobo  Basin,  Final  Report, ARDCO-GEOSERV, May 1995 (Volumes H-E, Irrigation, and IT-C, Groundwater).
Figure 3-4-1 shows the locations of irrigation area in the previous studies,
3.4.2  General
The Russian  Studies are very  extensive and  include climate,  water resources,  land  resources, agriculture,  irrigation  engineering  for several projects,  cost estimating,  construction  planning  and economics.
Land Resources, Irrigable Areas
In  this study,  a total area of about 2  million  hectares was investigated  and  about l.l million hectares were found to be suitable for irrigation based on a field and laboratory- soils survey and the Soviet land  classification  system.  This included  about 380,000  hectares in  the eastern  part of the plain  at elevations above 425  m (non-inundable) and  738.000  hectares in  the western  part of the plain,  below 425  m elevation  (inundable).  It was found  that most of the lands in  the eastern  part are suitable for development whereas the lands in  the western  part would  require protection from inundation by dikes and pumping out of excess surface water in the rainy season. For this reason  it was recommended  to  start irrigated  farming  in  the eastern  part of the plain.  It was also  recommended  to  postpone the introduction  of irrigation  in  the western  part until such time as positive results are obtained  from an  experimental plot proposed  by  EUROCONSULT at Papeo.
T AMS-ULG baro-axobo fuver basin tntegrateo development master h an3.          WATER RESOURCES
Potential Irrigation Projects
Several potential irrigation projects v>cre identified as follows:
Total Area - ba
Irrigable Area
Gross
tkL
A. Non-Inundabie
Baro River Basin
1.  Right Bank, Itang-Lare
2.   Left Bank
Alwero River Basin
3.    Abobo
4.    Alwero (tinder construction
Gilo River Basin
5.    Right Bank, Ubcla-Pashula
6.    Left Bank
88.800
16,000
20,900
12,000
42,400
33.000
79,000
15,000
18,800
10,400
39,000
Sub-Total
213,700
30,000 193,000
B. Ipundubk
Bara River Basin
7. Left Bank, from Itang 
Town to Adura River
121.71K1 
110.000
Sub-Total
121,700 
110,000
TOTAL
335,400 
303.100
The   water   resources   for   the   above   would   be   supplied   by   the   Baro,   Alwero   and   Gilo   Rivers. Surface water quality is suitable for irrigation.
Several irrigation schemes were proposed, as described below:
Irrigation on rhe Burn River Basin
The hang-Lare area in  the Bare River Basin  would  be irrigated  from the Bang  dam Initially 
a nghl bank canal would rmgale a nel area of 37,000 hectares by gruvtty l  u,
nc
' Jv
additional storage upsimam of Itang, this area could be expanded. On the righ^banit jte gravity area would he expanded to cover a net area of 51.900 hectares. A pumped scarce on the nght bank would rrngare an additional net area of 28.000 ha Thus toe roLl surfl« 
area rmgaled on the nghr bank could grow to 79,900 hecures. In addirion. a pumped scheme and a left bank canal could imgare a total net area of 15,000 hecrares on toai bijk Th?? Ihe long .erm rhe total ne. surface area urigated would reach 94.900 hectares. Furthe^ta
' ,0.000 hectares on toe .eft bank by means of. g~ “ £ 
£
TAMS-ULG baro-akobo river basin INTEGRATED development master plan
3-213. WATER RESQIFRCES
excess water back  to  the Baro  River.  This way  the ultimate net area irrigated  downstream of [tang  from Baro  River water resources regulated  at ltang  and  al other not yet determined  storage upstream could  grow io  about 200,000  hectares.  However,  the timing  and  sequence of development of this ambitious project was not defined in the Russian Study.
Four main  unlined  canals with  a total length  of 228  km and  combined  capacity  of 300  m’/sec were proposed  for the ltang  Lane Irrigation.  Additionally,  two  major pump  stations were proposed  for the upper areas.  The project would  include extensive distribution,  drainage systems, and food protection dikes and pump stations for the flooded areas.
Future dam alternatives under consideration  after construction  of the ltang  Dam are: Gambela, Bonga and  TAMS dams.  The construction  of any  one of these dams would  be extremely  costly. A conveyance between  any  of these dams and  ltang  would  be very  difficult,  may  require an additional 70-100  km canal and  its cost would  be prohibitive.  Thus only  gravity  conveyances from the [tang  Dam site and/or pumping  appear feasible for additional irrigation  using  the flow regulated at those dams.
The main advantages of developing irrigation in the ltang area are:
Accessibility-
Abundant land, water and hydropower resources
Possible navigation downstream of Gambela
Irrigation in the Al war a River Basin
The Russian Study proposes two irrigated areas in the Alwero River Basin.
The first area (Alwero  Project),  covering  a net surface of 10,400  hectares,  is located  downstream of Alwero  some 20  km from the Abobo  village.  Water would  be supplied  from the Abobo  d^m and left bank gravity conveyance canal presently under construction.
The second  area (Abobo  Project) is located  along  the Gambcla-Abobo  road,  near the existing Abobo  state farm.  The net irrigated  area would  be 18,800  hectares.  At the time of the Russian study  about 5,000  hectares were under rainfed  cotton  cultivation.  The rest was covered  by  forest Waler for irrigation  would  be supplied  from a dam on  the Chiru  river,  a right tributary  of the
.Alwero,  A  25  km  long  earth  canal  with  a  discharge  capacity  of  24  m’/sec  would  convey  water from Chiru dam to the Abobo project area on the right bank of the Chiru river.
Irrigation in the Gilo River Basin
The Russian  study  proposes to  ultimately  irrigate a total net area of 69.400  hectares in  the Gilo River Basin,  including  39,400  hectares on  the right bank  and  30,000  hectares on  the left bank Several storage sites were considered.  The Gilo-21  dam was proposed.  Water would  be conveyed  by  two  main  gravity  canals.  The right bank  canal would  be 42  km long,  designed  for 50  m/sec.  The left bank  canal would  be 34  km long,  designed  for 39  m’/sec.  The canals would be concrete lined in view of potential seepage losses along the proposed route. The area is
T AMS-ULG haro-akobo fuv i r bask ktegrated develormlnt mas it.r flan3.          WATER RESOURCES
generally accessible.
Two  planning  oplions were considered,  both  in  the Ubela-Pachala area,  on  lhe right bank; A minimum area for diversified  agriculture covering  a net imgated  area of 13.500  hectares and  a maximum option  which  in  addition  to  the above would  irrigate a net area of 18.000  hectares.  In both  options,  the regulating  dam and  the right bank  conveyance would  be built during  the period of 1991-2010. The left bank irrigation was not included in the 1991-2010 plan.
First Stage Development
From these schemes in  the Russian  studies a first stage development plan  to  be implemented  from 1991  to  2010  was selected  The proposed  first stage development includes the 10,400  ha Al were Project presently  under construction  and  the development of two  areas at Ubela-Pachala including  13,500  ha and  18.000  ha on  the right hank  of the Gilo  River.  The first one constituting  a minimum scenario  would  be devoted  to diversified  agriculture while the second  one making  up  a maximum scenario  would  be devoted  entirely  to  sugar cane and  would  be part of a sugar mill agro-industry.  The mill would  be located  at Gambela.  This project (Ubela-Pachala) would  also  require the construction of the Gilo-21 dam and hydropower station.
On-Farm Irrigation Systems
The Russian  study  proposes gravity  earth  on-farm irrigation  systems including  furrow irrigation  for tilled  crops (cotton,  sugar cane,  maize) and  strip  irrigation  for close-growing crops (groundnuts, crotaline). Overall imgaiion efficiency is assumed to be 60%.
Quantities and Costs
Quantities and costs were detmnined on the basis of the following:
-  Layouts and preliminary designs of the main canal systems on I 50.000 scale however longitudinal profile for Abobo was studied on 1.10,000 scale);
-   Geotechnical investigations along main canals.
-   Preliminary design of representative plots at L 2,000 scale.
-   Costs were determined for machinery , labor and other inputs for dams, irrigation 
systems, and other related works such as housing, agro-industries, hydro plants, roads.
lhe Gilo project was preferred ovct the liang project on the Bare River because of:
-   Better soils;
-     Ilie  Gilo  21  reservoir  would  inundate  a  deep  canyon  where  land  is  not  fanned  whereas lhe hang reservoir would inundale 8.450 hectares of the most developed flood plain
y Yf S-ULG        b aroakobo river basin integrated development master pi an
1213.         WATER RESOURCES
funds reaching as far as Gambela Town;
-  The unit cost of hydraulic works would be lower in the Ubela PashaJa (7300 Birrs)
than in Ilang (9000 Birrs);
-  Gilo 21 would produce 20% more power than hang
Total water demands for 60% efficiency  and  a diversified  agriculture orientation  were
estimated in the Russian Study as follows;
Wet season 
4098 m’/ha
Dry season 
8955.jn’/ha
Total annual
13,095 m’/hn/year
The above was based on 80% rainfall probability.
Sugar  cane  demand  was  estimated  al  24.300  m’/ha/year.  The  above  includes  all  losses, from the intake to field application.
Thc agricultural orientation proposed for irrigated agriculture in the Lower Basin includes
a)     A rotation;
Grain crops
Cotton 
Green manure crops
Green legumes, sesame, groundnuts 25%
b)     vegetables
c)      sugar cone
25%
25%
25%
In  a  diversified  agriculture  scenario.  90%  of  the  land  would  he  in  the  rotation  and  10% in vegetables. Sugar cone would be grown in an area devoted io this crop only
3.43 Upper Busin Studies
Sixteen  putcnliid  irrigation  areas were identified  by  ARDCO-GEOSERV in  the Upper Basin valleys nt elevations ranging  from 450  to  1600  meters.  These areas are distributed  throughout the study  urea.  Access is poor or lacking  in  many  of these areas.  Topography  and  generally  lack of infrustniciurc and  population  are other constraints for development of irrigation  to  several of the urea.'
The  areas  were  identified  on  the  basis  of  topography  (less  than  8%  slope)  and  availability  of surface water resources. Soils data is scarce and based on a region-wide reconnaissance along
TAMS-ULG    ItAKU AKnilO Hl VI K UAStN INirORATTIl OEM IIIPM1ST MASTTR PLAN
33.          WATER RESOURCES
selected  corridors where aerial photo  interpretation,  limited  sampling  and  field  inspection  was carried  out and  subsequently  extrapolated.  A tentative land  suitability  classification  for irrigation based  on  the IJSSCS criteria is given  in  the ARDCO-GEOSERV report.  It is recommended  that this should  be reassessed  based  on  additional soils survey  work  as discussed  elsewhere in  this report.
A list of die project areas,  source of water supply,  average altitude,  slope,  total gross area (based on  topography),  approximate net area (based  on  estimated  peak  water demand  per hectare and estimated 75% probability available dry season water supply) is given below:
TAMS-ITLG
BAJtO-AKOnO RIVER BASIN INTEGRATED DEVELOPMENT MASTER ManWATER resources
Tiihie; 3-4-1
Trrinntian Areas in the Upper Basin
(ARDCO-GEOSERV)
Potential
Irrigation
Site
Water
Supply
Source
(Rivers)
Altitude
(m)
Slope
(%)
Total
Gross
Area
(Ha)
Potentially
Non-
Irriftablc Area (Ha)
1 Koji
Koji, Gioue
475
<1
6000
288
2. Snko Giida "•
Sjko Guda
550
2-5
4600
3010
3. Beko
Beko, Meni
1150
2-6
6000
544
4. Kiln1 2 3"
Ou
1050
2-6
5600
4994
5 Lalo Kolu
La-u
550
2-6
9000
19
6. Baro "> '»>
Bare
475
<2
2000
2000
7. Blrbir«'» «
Birbir. Kelo
1150
4-7
8000
8000
1 Fani
Fani
1250
3-7
1200
212
9. Abwcro
Alwero
650
3-7
5500
240
10. Guv
Guy, Leman
550
4-6
1800
110
11. Godarc
Godarc
750
2 6
3300
651
12. Achitni
Acham
950
<2
4300
312
13 Awavi
Awayj
600
2-1
5000
137
14. Bebeka
Gctcheb
j 950
4-6
6000
424
15, Guracha
Gumcha
650
2
2000
144
16. Gumcro1’1 w
Gumeru
1600
<8
4000
1467
17. AkoboU1*
Akobo
150
<8
5000
4528
IS. Akobo 11111
Akobo
650
<5
30000
2842!
TOTALS
109300
51671
NOTES:
(1)          The S.iko Gud», Kiln, Dwo. Birbtr. Gunicro. Akobo jhcs art located tlowcunraar of ptrttnriat hydro power Rlrrngfl dams.
(2)         Th* WrMr and Oumnro ■ lies miy be Ln conflict wiih downstream hydropaurr development
(3)          |-he limiting A*10* m the caw of die Boro tile Is land, available flows with or witliom regulation exceed the Lrri gallon requirements.
TAMS-VlX* 11 Alto AKMIU1HIVLH IUMN IN RW FED niATI.OFMUX'T MASTER PLAN
1-7*3.          WATER RESOURCES
Only  surface water is considered  as a potential source.  The above estimates on  water availability arc based  on  available data on  river flows.  The only  storage sites considered  are those identified for hydropower purposes Other storage dam sites may  exist; however,  these have not been identified.
Possible   diversion   locations   were   suggested   However,   the   study   docs   not   include   conveyance, distribution, drainage, bank protection, access roads, nor cost estimates.
Waler quality for irrigation based on seven samples appears to be excellent
Irrigation System
The ARDCO-GEOSERV study presents a typical layout for a surface irrigation canal- drainage road network. The unlined canals are sized for the peak flows for 21 hours irrigation, an assumed 80% net area vs. gross area, and an overall irrigation efficiency of 43%. This efficiency is arrived at by assuming 90% conveyance efficiency. 80% distribution efficiency and 60% field application efficiency.
Irrigation Water Demands
For the Upper Basin. ARDCO-GEOSERV estimated the irrigation water demands for four typical project areas located at different elevations. assuming several crops, as follows:
Group
Altitude
(m)
Net .Annual Water Requirements
(m  ha)
3
Gross Annual
Water
Requirements
(m’ ha) (1)
1
450 -650
6847
15.900
II
750 - 950
5717
13.300
Ill
1050-1250
6219
14.500
IV
1600
7047
16.400
(1) For 43% overall efficiency
TAMS-l LG
RIVER BASIN INTEGRATED DFV| I Ol’MENT M VSTFIt AN
i 213. WATER RESOURCES
Crops considered were:
Gimip             Crops
I                      Banana. Cotton, Groundout
II                      Cotton. Sorghum, Maize
III              Sorghum. Maize, Coffee
IV              Tea. Rice. Onion
Existing Irrigation
The  ARDCO-GEOSERV   report  (Vol.   II-C,  Groundwater)  notes  the  existence  of  10  self-help small irrigation sites in the Upper Basin. Such sites comprise a total area of 1100 hectares.
3.5 Hydropower Projects
3.5.1    General
A   wide   range   of   hydropower   projects   has   been   investigated   in   the   Russian   study   and   the ARDECO-GEOSERV study.
Some of the projects are medium lo  large single propose schemes with  regulation  storage.  Others are multi-purpose projects,  usually  involving  irrigation  as well as hy  dropower and  occasionally with  flood  storage potential.  Also  included  are run-of-lhe river schemes,  located  in  the upper basin.
In  lhe course of this review,  project data sheets have been  prepared  for each  project using  the information  contained  in  the Russian  and  ARDGO-GEOSERV reports.  These data sheets are given in Volume 3.
The   location   of   ull   the   projects   except   the   very   small   run-of-the   river   schemes   are   shown sc hematic ill I y on Figure 3-5-1
3.5.2    Lower Bisin Study
The Russian  Study  identified  17  projects for hydroelectric,  irrigation  and/or flood  control purposes These ranged  from the 275-metcr high  TAMS dam.  with  519  MW of installed capacity to the 23 meter high Mei flood control dam in die Alwrro basin with 2 MW of installed capacity.  One of their sites,  the Abubo,  on  irrigation  dam with  no  power,  is presently  being constructed  on  the Alwero.  Some of their proposed  projects are not.  strictly  speaking.in  the lower basin,  and  one.  die Birbir project.is included  the ARDCO-GEOSERV study  as a diversion site, with u storage site much further upstream.
l AMS-VLG UAkO AMIIUIHIVIK MMN INIKrk A11 ll til V| l OPMCNt MAS11 R PlAN3.          WATER RESOURCES
The schemes out actually fall in the lower basin are.
Project             Rated Head
Installed Capacity
(M)
(MW)
Baro River system
Itang
5.5
Gambela              88
Bonga
124.8
Tams 220
Alwero Rivet System
Abobo                *
16
258
372 519
Chiru
33
Mey
23
Dumbo ng
37
2
0.2
3
Gilo River System
Gilo-3                   198 Gilo-2                 27 Gilo-21               26 Gilo-1                   782
14.4 16 18 72
On  the Baro  River System,  the Itang  Dam is mainly  for irrigation,  with  hydropower as a secondary  purpose.  The other three projects on  the Baro  River,  the Gambela,  Bonga and  Tams projects,  are large hydropower projects which  also  have some flood  storage potential as well as providing  improved  regulation  for downstream irrigation  systems.  The Russian  study  concluded that of the three, the TAMS project was the strongest economically.
On  the  Alwero  River  System,  the  Aboho  dam  is  a  single  purpose  irrigation  project  and  the  other three projects are primarely flood control projects, but all have hydropower components.
Similarly  the  projects  on  the  Gilo  River  are  primariy  for  irrigation  projects  although  they  have hydropower components
3.5.3 lipper Basin Study Schemes
The ARDCO-GEOSERV study  identified  11  run-of-river projects and  9  projects with  storage. Of the 11  run-of-river projects 9  have capacities,  of less than  350  KW.  apparently  without any possibility  for future expansion  The other two  at 2.5  and  6  MW arc on  the Sor Ris er downstream of a storage site identified  by  the Russians and  could  be expanded  to  an  ultimate capacity  of 80MW (or 96 if a powerplant were provided at the storage dam).
T AM S- UI.G    baroakowj rtver sastn tshtora tfr oevelofmixt master n.AN3.         WATER RESOURCES
The nine projects with  storage actually  incorporate 15  power plants,  by  adding  diversions downstream of the storage sites which  take advantage of the regulated  flow.  These pow erhouses range in size from 10 to 220MW with a total potential of over 1000 MW. Based on the available hydrology,  and  the storage provided,  the installed  capacity  shown  for some of these schemes appears loo high This wall be reviewed in the Phase II Studies.
The 9 storage projects in the upper basin are as follows :
Project
(M)
Rated Head
Installed Capacity (MW)
Baro
DIV-1 80
50
DIV-2 360
220
Geba
DIV-1 275
100
DIV-2 160
80
Birbir
DIV-1 130
70
DIV-2 250
220
Kushu
DIV 500
50
Gumero
DIV-1 120
15
DIV-2 180
22
DIV-3 200
25
Sor
DIV 200
80
Beko
DIV 260
100
Sako Gudu
DIV 500
25
Sese
DIV-1 HO
10
DIV-2 130
20
The Bibir,  Koshu,  Gumcro  and  Sako-Guda projects arc all associated  with  irrigation  projects.  It will be noted that the Boro, Gcba, Biribir. and Ciumero projects comprise upstream storage darns to regulate llow, feeding alternative downstream diversion systems which develops the head.
3,5.4 Dam SUc Review
A large amount of data on  the projects has been  assembled  in  the project data sheets given  in Volume 3 The Ibllow ing sections gives information on the geological setting, the proposed dam types, and construction materials. The locations olTitll the sites ire shown on Figure 3-5-2,
TAMS-VLG H*xo amuiu HivimiAsiN inteomateo development master ruesWATER RESOURCES
Baro Dam Site
The Baro  dam site is in  the upper reaches of die Baro  basin,  about 75  km upstream of the junction  with  the Birbir and  30  km almost due south  from the town  of Gore.  An  earth  fill
dam was proposed  lor this site The dam is founded  on  organic clay  over residual clay overlying  basalt which  is underlain  by  amphibolite and  gneiss.  The residual clay  on  the hill slopes will probably  be suitable as a construction  material and  the basalt will provide rockfill.  The proposed  dam has a maximum height of 35.0  meters and  the fill volume about 420,000 cubic meters
Geba Dam Sire (ARDCO-GEOSERl^
The Geba dam site is on  the Ge ba River about 120km upstream of its junction  with  the Birbir and  60  km east of Gore.  A rockfill dam was proposed  for this site The dam is located  on  alluvium overlying  basalt.  The alluvium will have to  be removed  Basalt is suitable as rock  fill Clay  on  the hill slopes is suitable as impervious fill Sand  and  gravel is found in the Geba flood plains.
The  proposed  dam  has  a  meters  maximum  height  of  45  m  and  a  fill  volume  of  about 1.690,000 cubic meters
Geba Dam Sire (Russian SfutfyJ
lhe  Geba  dam  site,  as  proposed  by  the  Russians,  is  on  the  Geba  River  3km upstream of its confluence with the Birbir River and 75km from Gambcla
The dam is located  in  a steep-sided  valley  on  granite The granite is weathered and  fissured  to  a depth  of 10m I n  weathered  granite is dense and  hard  and suitable for rockfl II-
Thc proposed  dam has a 258m maximum height and  an  820m long  crest.  The crest width  is 12.0m Normal free board  is 5  35m The average side slopes are I on  2.0  upstream and  1  on  I 75  downstream The fill volume is estimated  to  be 30.474,600 cubic meters.
Birbir Dam Site ( .4/HXTKGEOSERV)
The Birbir dam site,  as proposed  by  ARDCO-GEOSERV.  is on  the Birbir River about 65km upstream of its confluence with  the Baro  and  70km ENE from Demhidolo A rock fill dam was proposed at this site
The dam is located  on  biotite granite overlain  by  residual day  and  fluvial fine sand.  Five test pits were located  at the dam site which  reached  to  a depth  of 4m. A geological map  at 1  -2000  scale and  a cross section  through  the pits is given  in the report.  The clay  and  rock  arc considered  to  be suitable as construction materials. Imperv ious fill also is found in undulating hills and hill slopes at the
yf
f;         RAk< ^AkOBO RIVER BASIN INTEGRA TED DEVELOPMENT MASTER PLANWATER RESOURCES
approach to the dam site. No deposits of natural sand and gravel were found in 
the vicinity of the dam site.
The proposed dam has a maximum height of 46 m and a fill volume 1,450,000 
cubic meters.
Birbir Dam Site f Russian Study)
The Birbir dam site,  as proposed  by  the Russians,  is on  the Birbir River 13km upstream from its confluence with  the Haro  and  75km northeast of Gambela.  The dam is located in  a mountain  valley  on  granite and  clay  shale interbedded  with  quartzite.  Weathering extends to  20m.  The granite is suitable as rockfill.  The proposed  dam has a maximum height of 185 meters and a fill volume of about 19,433,000 cubic meters.
Koj/ur Dam Site
The Kashu dam site is on the Kashu River, a tributary of the Kilu River and 85km
upstream of its junction with the Akobo It is 22km SSE of Mizan Teferi.
At  the  dam  site  clay  overlies  volcanic  flows  and  pyroclastic  rocks,  The  clay  on 
the  hill  slopes  at  the  dam  site  is  residual  clay  and  is  probably  suitable  as  a  fill
material.  The  volcanic  flows  may  be  suitable  ns  rockfill  but  pyroclastic  rocks
would be unsuitable. Sand and gravel may occur in the Kashu flood plain.
The proposed dam has 30.0 meters maximum height and a 250 meters long crest
The  dam  crest  width  is  10  meters,  normal  free  board  is  3.0  meters.  Slope  ratio:
the  upstream  slopes  are  1:3.0  and  the  downstream  slopes  ore  1:2.5  The  fill
volume is estimated to be 310,000 cubic meters
Gumero Dam Site
The Gumero dam site is on the Gumcro River 35km upstream of the junction with Birbir
and 28 km northwest of Gore. A rockfill dam was proposed for this site. The dam is
located on basalt overlying schists. The basalt valley slopes will probably be suitable for
rock fill. Gentle slopes outside the sleep valley rim have soil suitable for impervious fill
No sand and gravel was found in area lhe proposed dam has a maximum height of 31 0 
meters and a fill volume of about 863,000 cubic meters.
Sar Dmu ,%c
The Sor dam site,  which  was also  studied  by  the Russians and  ARDCO-GEOSERV,  is on  the Sor River about 30km upstream of the Geha and  10km northwest of Metlu.  a rockfill dam was proposed  for the site The dam will be founded  on  fissured  weathered gneiss,  30m thick,  underlain  by  hard  gneiss.  The gneiss will be suitable as rockfill The proposed  dam has a maximum height of 69.(1  meters and  a fill volume of about
1.572.000 cubic meters,
TAMS-ULC lv«u.AMHMi«lvm HASIM wrFtMATEDIWVajOfMBHTMASTTJt n ANwater resources
ileko Dam Site
The Beko  dam site is on  the Beko  River 25km upstream of its confluence with  Gilo  and 35km Northwest of Mizan  Teferi.  A rockfill dam was proposed.  The dam is located  on basalt overlying  gneiss.  Basalt and  gneiss will both  provide suitable rockfill.  Residual clay  immediately  outside the steep  valley  slopes is suitable for impervious fill.  Natural sand and gravel was not found
The   proposed   dam  has   a   maximum   height   of  123.0  meters  and  a  fill  of  about
8,069,000 cubic meters.
Sako GuJa Site
The Sako  Guda dam site is on  the Sake Guda River 35km upstream of its confluence with Raro  and  25km north  of Gambela.  A rockfill dam was proposed  The dam is located  on basalt underlain  by  granite Granite upstream of the dam site will provide both  suitable rocklill and  riprap.  Residual clay  on  gentle slopes on  both  hanks will provide impervious fill Natural sand and gravel were not found
The proposed dam has a maximum height of 58.0 meters and a fill volume of
about 210,000 cubic meters
Sese Dani Sire
The Sese dam site is on  the upper Scse River 45  km upstream of its confluence with  the Geba and  60km northeast of Gore A rockfill dam was proposed  The dam is located  on residual clay  overly  ing  basalt Basalt is not exposed  al the dam site.  The clay  is abundant and  both  the clay  and  the rock  arc considered  to  be suitable as construction  materials. Basalt is not exposed  but should  be av  ailable on  the deeper hill slopes.  Natural sand  and gravel were not found-
The  proposed  dam  has  a  maximum  height  of  43.0  meters  and  a  fill  volume  of
about 2,540.000 cubic meters
Gf/o 7 Dam Stfe
The Gilo  1  dam site is on  the Gilo  River approximately  75km upstream from the village of Agcnga,  in  the flood  plain,  and  95km WNW from Mizan  Tefen.  Il is praciically impossible lo  reach  this dam site Ihc site is composed  of hard  gneisses.  weathered  to  20* 25m The gneiss would be suitable as a rockfill
The  proposed  canhfil)  dam  has  a  maximum  height  of  a  lti5m  and  a  fill  volume of about 5,205.000 cubic meters
Tams Dam Site
T XMS-l EG raro-akob*> river jiastm iNTBOXATrn developm! nt master n an3- WATER RESOURCES
The  TAMS  dam  site  is  on  the  Baro  River  40km  upstream  of  Gambcln.  The  dam
is  located  on  gneiss,  granite  gneiss,  schist  and  granite,  A  thin  colluvial  deposit
overlies.  The  rocks  orc  weathered  and  fissured  to  35m  depth  Unwcathcrcd  rocks
arc   dense   and   hard   and   suitable   for   construction.   The   proposed   dam   rias   a maximum height and a fill volume of about 83,094,000 cubic meters
Gilo 2 Dam Site
ITie Gilo 2 dam site is on the Gilo River approximately 33km upstream from the
village  of  Agenga  and  85km  south  of  Gambela.  An  earthfill  darn  was  proposed 
for  the  site.  At  the  dam  site,  basalts  overlie  gneisses  with  clay  overburden  up  to 
10m  thick.  Weathering  of  the  bedrock  occurs  to  15-20m.  The  gneiss  would  be
suitable  as  a  rockfill  and  the  clay  would  probably  be  suitable  as  impervious  fill
Five boreholes were sunk at the site
The  proposed  dam  has  a  maximum  height  of  40m  and  a  fill  volume  of  about
1,595.000 cubic meters
Bonga Dam Site
The  Bongn  dam  site  is  on  the  Baro  River  27km  upstream  from  Gambela.  A
rockfill dam was proposed for this site. At ihe dam site there is a Lhin covering of lacustrine alluvial sand  and  loam overlying  gneiss,  granite and  schist Rockfill
and  cohesive  material  are  available  for  construction  Weathering  of  the  bedrock extends to 10m. A cross section shows three boreholes.
The  proposed  dam  has  a  maximum  height  of  165m  and  a  fill  volume  of  about
86,003.000 cubic meters,
Gambela Dam Site
The Gambela dum site is on  the Baro  River 13km upstream from the town  of Gambela A rackfill dan  was proposed  for this site At the dam site Quaternary  deposits overlie granite gneiss,  gneiss and  granite Weathering  of the crystalline rocks extends to  30m depth The crystalline rocks would be suitable as rockfill.
Hie  proposed  dam  has  a  maximum  height  of  65m  and  a  fill  volume  of  about  61,000.000 cubic meters.
/king Dam She
The I lung dam site is on the Bare River 10km upstream of the town of hang in the flood plum,  and  26km downstream from Gambela At the dam site lacustrine-alluvia! deposits of clays und  foams overlie intrusive rocks at depth  A cross section  of the dam is given showing  five boreholes.  It is considered  that this would  be an  earth  fill dam and  clay  is probably available us core material.
TAMS-ULG HAXi) AMMioKFvrn own iKnr.n<n nm vfi opMtNr masterm_an
vx<3.          WATER resources
The proposed dam has a maximum height of 13m and a fill volume of about
1535.000 cubic meters,
Dumbong Dam Site
The  Dumbong  dam  site  is  located  on  the  AJwero  Riv  er  26km  upstream  of  the
Abobo   dam,   currently   under   construction,   and   55km   south   of   Gambcla.   An 
earthfill dam was proposed  for the site.  The dam site is located  on  gneiss intercalated  with  milonite and  breccia.  Overburden  is negligible.  On  the surface
the  rocks  arc  badly  weathered  and  fissured  to  35m.  The  gneiss  would  be  suitable
as rockfill. A cross section al the dam site shows nine boreholes
Ihc  proposed  dam  has  a  maximum  height  of  48  3m  and  a  fill  volume  of  about
477.000 cubic meters
CWn/ Dam 5rre
The Chiru  dam site is on  the Chini River 15km upstream of its confluence with  the AJwero  and  42km south  of Gambcla.  An  earthfill was proposed  at this site.  Al the dam site thin  Quaternary  deposits overlie crystalline rocks.  A badly  crushed  zone 800m wide and  up  to  76m deep  which  is severely  fissured  occurs at the dam site and  geological conditions are not favorable A cross section at dam site shows three boreholes.
The  proposed  dam  has  a  maximum  height  of  50m  and  a  fill  volume  of  about  3  J  54.000 cubic meters
3,5.5 Hydraulic Structure
The ARDCO’CiEOSERV spillways for the storage dams are*  apparently  without exception, ungated  with  low unit discharges.  Surcharge on  the spillway  crest is normally  only  15*2.0 meters.  From their sketches,  it would  appear that the chutes are the same width  as the spillway, and  that no  energy  dissipation  structure is planned  Even  with  only  15  to  2.0  meteTs surcharge, the effect of routing  the design  inflow flood  is very  marked,  resulting  is a reduction  of as much as 50% in the outflow No information is given on spillway capacity for the power diversions.
The Russian  spillways,  with  one exception  I Dumbong),  are provided  with  vertical lift gate*, generally  5-7  meters high  and  up  to  12  meters wide Current practice indicates that radial gates would  be a more appropriate choice The chutes taper downstream of the gate structure,  resulting in  a unit discharge at design  flow of 80-00  ems meter In  all but three cases,  it was apparently intended  to  end  these chutes in  flip  buckets I tang.  Gilo  2  and  Gilo  3  had  stilling  basins.  In  these projects also,  the effect of routing  the design  flood  through  the reservoir.  aJuwing  a nsc above normal pool ot 1*2 meters, was very marked, resulting in a decrease in the outflow of up to 50%.
Outlet Works
TAMS-ULG baroakoho rafr basin iwnriMri-t>n»-vEi.upmi stmaster piWATER RESOURCES
Based  on  the only  two  detailed  drawings available (Birbir and  Kashu),ARDCO-GFlOSERV intended  to  combine the diversion  conduit and  outlet works for the storage dams in  a single structure,  with  a gate tower located  upstream of the dam over one conduit while the other(s) would be plugged wilh concrete after the diversion period ended
The Russian  projects generally  show no  low-level outlets per sc.  apparently  relying  only  on  the
power  intake.  The  exception  is
the  TAMS  dam  where  a  note
t
indicates  that  the  diversion  tunnel
will he used ns a low level outlet. However, no gates or gate house is shown
3J.6 Power InMcitations
For most of the ARDCOGEOSERV projects,  the power intakes arc canal he ad  works structures associated  with  diversion  dams.  Presumably  they  would  have low-pressure slide gates with automatic or remote control.  In  three cases,  however,  Beko,  Sako  Guda and  Sese,  the power diversion is from the reservoir and ramp or tower intakes would need to be provided with higher’ pressure operating gates
The Russian  intakes are generally  tower intakes,  with  vertical lift or roller gates located  below the maximum drawdown  level.  At the TAMS dam and  at Birbir,  Ge ba and  Sor,  the intakes art located on the abutments, with inclined ramps and wire rope hoists.
Power waterways for the ARDCOGEOSERV projects are mostly  canals and  penstocks,  with enlarged canal sections at the upstream end of the penstocks to provide daily or weekly pondage. Some projects also have tunnels and/or syphons.
The Russian projects generally have conduits or runnels under or around the dam but in some the intake structure leads directly  to  steel penstocks.  At Irang.  Gilo  3  and  Dumbong  the powerhouse is located within a concrete section of the dam.
No in formation is available on the powerplant structures planned by ARDCO-G EOSER V. The Russian powerhouses are all of the indoor type, and are equipped with Francis turbines except
for Gilo 2, which has Kaplan turbines, and Iiang, which has bulb turbines,
3.5.7    Existing Projects
Al the present time there arc only rwc existing hydroplants in the basin One is the Sot plum near Menu  with  5  MW Of capacity.  Il saves Metro  and  Gore,  and  numerous small communities between  Gore and  Gambols,  and  Denn  dole Fhe other plant is the Dem bi powerplant in  the Pcbckii coffee plantniicn near Mi/jn I'cferi. Il has a capacity of 800k w.
3.5.8    Project Costs
A very preliminary review has been made of the cost estimates pre sen led in the two reports White no formal expluination of the methodology used is given in the Russian report, they made
| 5MQ survey maps oftheir sites and quantity estimates, and estimated unit prices. The two latter will be reviewed and updalcd in lhe course of the Phase 11 studies.
TAMS-l'U;
wvt« imsjn iniig^h tintwtuuwst mamwhas3.          WATER RESOURC ES
The ARDCO-GEOSERV studies used  cost estimating  formulas based  on  a limited  range of major projects parameters and  were restricted  to  the existing  1:50,000  scale topography,  v'ith  the exeption  of the Bibir site,  where they  were able to  do  a ground  survey  at a scale of 1:2000. Although  cost estimating  formulas are sometimes used  for very  preliminary  screening,  with  a manageable number of site,  it would  appear that some of the cost formula used  ,  have not given an  accurate enough  assessment of the relative cost for the range of projects under considcrartion. A more systematic computer based  cost estimating  system will be undertaken  in  Phase II,  It is clear drat more work  needs to  be done on  estimating  project costs before reasonable comparisons can  be made between  projects.  The ranking  of projects given  in  the ARDCO-GEOSERV report must be regarded as very preliminary and should no be used for screening purpose
3.6 WATER SUPPLY PROJECTS
3.6.1 Overview
The following studies/reports were examined
1.           Survey and Analysis of the Upper Baro-Akobo Basin Final Report. ARDCO- GEOSERV, May 1995.
Vol I - Main Report
Vol II C - Hydrogeology
2. 
Baro-Akobo Master Plan Study of Water and Land Resources of the Gambela
Plain Final Report Draft. Selkhozpromexpon. 1990 (Russian Report>.
Vol X Annex 16 Infrastructure and Several Amenities
3. 
Gambela Needs Assessment Survey Region 12 Group Report. UNDP Addis
Ababa, August 1993
The ARDCO-GEOSERV Report covers the Upper Basin.  The present waler supply  situation, presented  as part of the Hydrogeology  is based  on  a socio-economic survey  earned  out in  1994. The report covers the present water supply  and  a demand  for the urban  and  rural population. Projections for the period  from 1^95  to  2025  and  discusses the current status of water supply based  on  the survey  and  the potential sources of waler supply.  However 3  Warcdas appear fo  be left out: Meko, Sachi and Tirmatid
In  addition,  ihe ARDCO-GEOSERV report provides information  on  existing  industries with limited  assessment of their water supply  requirements as well as estimates for the present livestock  water supply  requirements.  Sewerage is not discussed  in  this report.  Specific plans arc not developed.
The   Russian   Report   is   based   on   1988   conditions   in   the   Gambela   area.   It   covers   lhe   needs perceived al that time for the waler supply anti sewerage m tbc lowns of Gambela and Itang, the
TAlMS*ULG RaROaXOBO RIVER BASIN INTEGRATED DEVfioPMENT MAST! RPt an3. water resources
water  and  sewerage  for  Giuribela  Awraja  for  die  period  from  1990  at  2010,  It  also  includes  the
water  supply  and  sewerage  for  the  rural  settlements  and  agroindustry  proposed  in  the  1990 
Master Plan  of the Gambcla Plain  and  considerations on  watering  of pastures for animal husbandry.  The study  includes several schematic diagrams for the water supply  of Gambela, settlements and  watering  points An  estimate of capital investment requirements is provided  for the proposed  waler supply  and  sewerage systems based  on  1988  prices for three development stages; 199M995. 1996-2000 and 2001-2010. However, this Master Plan is mostly outdated.
The Gambela Needs Assessment Survey  carried  out by  VNDP in  August 1993  includes a description  of the conditions prevailing  at that time and  provides specific recommendations for improving  the living  standards of the population.  However,  this report does not include estimates or water needs,  nor a development program.  Nevertheless,  the report is useful because of its diagnostic and  practical recommendations although  there is not much  in  terms of quantifying  the proposals. Generally speaking, the recommendations are very basic measures.
3.6.2   Institutions]
Until recently,  the institution  responsible for water supply  was the Water Supply  and  Sewerage Authority.  At present,  this entity  has been  put under the newly  created  Ministry  of Waler Resources.  However*  details of its organizational structure are not known.  Reportedly,  a regional brunch  of the waler sewerage responsible for operations in  the Baro-Akoho  basin  is located  at Jimn.
The ARDCO-GEOSERV report provides no indication as to the institutional organization 
dealing with water supply in the Upper Busin. The report indicates that from the survey it
appears that Governmental and/or NGOs have made several attempts to provide 50% of the total
urban population with piped and presumably safe water. However, because of the absence of
efficiently organized institutions to follow up the required repair and maintenance works, most
schemes are now totally abandoned or function below design capacities. This is especially 
evident where the schemes are based on pumps that require special technical skills for operation.
The Russian Study indicates that:
data on using drinking water standards was supplied by the Water and Sewerage Authority of Ethiopia; and
data on present stare of the population water supply of the Gambela Awraja was
supplied hy the Water Supply Authority in the Gam be I a and Gino cities.
The UNDP report on Gambela utiles the following;
"The main  organization  involved  in  (he abstraction  of groundwnier is the Ethiopian  Water Works Consirwiion  Authority  (EWWCA).  Adequate information  in  their findings has been  obtained  from the Jlmma Zcnnl Office.  Some information  hw been  obtained  from UNHCR*  Mcknncycsus.  and  WSSA (tho  Water Supply  tmd  Sewerage Agency) as well". In addition, the World Lutheran Red Cross is also mentioned in the UNDP report.
TAMS-VW*       IIARO-AMUI!JR!V| Ft flASIN FNTT CkATI D DnVFl.OI>MRCT M/S1TR H AW
in3.          WATER RESOURCES
The  above  provides  an  indication  of  the  organizations  that  in  one  way  or  another  arc  involved in water supply activities in Gambela.
3.63 Population
The population  data used  in  the ARDCO-GEOSERV study  for the Upper Basin  was population data and  housing  census reports al the Wereda and  published  by  the Central Statistical Authority (CSA) in  1984  (census Preliminary  Report and  Supplements,  CSA,  1984).  At the time of writing the ARDCO-GEOSERV report in  early  1995,  die 1994  population  and  housing  census was being processed by the CSA.
The analysis or the 1984  census (Census Supplement- 1,1984) estimated  annual population growth  rale of 2.9% for the year 1984-1985  with  an  increment of 0.01% for each  succeeding year.  According  to  the same statistical report,  the overall urbiin  growth  rate estimate for year
1984-85  in  Ethiopia was 4% per year with  increments of 0.12  % for the succeeding  years.  I'hese growth  rates for rural and  for urban  population  were adopted  by  ARDCO-GEOSERV for estimating  the population  for urban  and  rural areas in  10  sub-basins for the period  of 1995-2025 at 5-year intervals.
The ARDCO-GEOSERV Projection is presented in Table 3-6-1, below.
T AMS-IILG baro-akobo river basin integrated development master Plan
1-J93          water resources
Table 3-6-1
Population Growth Projections in the Upper Baro-Akobo River Basin
CATCHMENT 1995 2000
URBAN POPULATION
2005                    2010                     2015                    2020                2025
YABUS
17731
23243
DE GA
31340
43449
61717
84874
MIDDLE haro
6005
7871
10613
14713
20900
11C68
28741
14506
19559
UPPER BIRBIR
27116
3851B
81641
107004
529T0
144279
11DB20 37528
69163
LOWER BIRBIR
200023
284126
52503
68814
390733
92765
SOR & GE BA
128635
182721
71918
94261
*27096
251283
510178
328094
BARO
6387
176202
8372
250286
344199
11268
15649
UPPER AL WE RO1'
22229
0
0
30570
0
449410
39915
LOWER ALWERO'1
0
0
0
0
UPPER GILO
0
0
0
29169
38231
0
51548
0
71465
0
0
UPPER AKOBO
29986
101514
39301
139602
52992
73456
182278
TIRMATID1"
0
104356
0
0
143511
187381
0
0
0
0
TOTAL
306410
401603
RURAL POPULATION
YABUS
159317
184962
DE GA
541501
215255
85921
750718
251119
100912
1956368
293540
117439
72863
1456481
343560
137006
t9l4775 401968
MIDDLE BARO
53928
62609
160205
167440
65003
219307
UPPER BIRBIR
590128
585145
797350
99396
116294
136065
LOWER 8IRBIR
239373
277904
930207
1087715
323420
1272633
377304
1488988
SOR & GE BA
*50277
522755
441192
606373
709733
516197
603954
BARD
137102
159170
829909
185240
970999
216102
1136075
UPPER ALWERO"
0
0
252694
0
0
295653
0
345916
LOWER ALWERO''
0
0
0
0
0
0
0
0
UPPER GILO
274004
318109
370210
431B90
0
UPPER AKOBO
290993
TIRMATIO’11
337033
393 IM
S05D20
4585E-6
59007?
536333
691329
0
0
0
3063328
627513
0
734194
_______________ TOTAL
2282071
2649401
0
3597036
0
0'
NOTE: (l) These sub-basins are located in the Lower Basin, Source; ARDCO-GEOSERV, May 1995
TAMS’ULG BARO AKOW) WVERBASW [NTLGRATTD
DEVELOPMENT MASTI R PLAN
4-103.          WATER resources
The projection  given  in  Table 3-6-1  show that the urban  population  estimated  at 306,410  in  1995 would  gTowto  1.914.775  by  2025.  The rural population  would  grow from 2.287.071  in  1975  to 5,757.798  in  2025.  The total population  would  grow- from 2,588,482  in  1995  to  7,672,574  by 2025.
Likewise,  ARDCO-GEOSERV estimated  the livestock  and  poultry  population  in  the Upper Basin.  ITie estimate for each  catchment area and  type of livestock  and  poultry'  is given  in  Table 3-6*2.
TABLE 3-€-2
Livestock Population by Catchment
L
L
CATCHMENT
LIVESTOCK
OXEN             COWS           CALVES             SHEEP              GOATS           HORSES         MULES          DONKEY S
POULTRY
[
TABUS
1&341              234&T                9272                  14726              12305                    13                  231
2D34               44269
□EGA 22311 1D3H7 4554 17351 4721 5761 166 971 8996
MIDDLE BARO 23248 19045 176 14601 790 5401 410 572 34902
UPPER B'iRBIR
14S107              78309             1B434                   79535              19635              15713                2920                6525             2D1110
| LOWER BlRBlR                          39414              35453              12409                  28194               51 S3                2990                  650                1640               &B362
SOR A GE BA
118220              81264              2637B                  67285              30674              10325                4607                5069             142230
BA RO                                      53439                5739                2041                   50555              22931               11333                 601                     37             24 IDG-
UPPER ALWERQ'' 952 221 136 52
LOWER ALWERO”-1 952 221 136 52
BOB
6OB
UPPER GILO
67699
39654
14260
70171
37961
322?
1378
267
161350
UPPER AKQBO
27371
36995
15935
32287
3414
5577
1015
18
907 IE
TIRMATlD’*1 ' TOTAL
52D054              33H53            103641
374146            137922              63464              11978               17333             777333
NOTE:(l> These arc located in the Lower Basin
c
r
T AMS-I1 G       BARO-AKOBO RIVER BASIN INTEGRATED DEVEl Ol*MI NT MASFFR PLAN
Ml3. WATER RESOURCES
A  detailed  discussion  on  livestock  and  livestock  population  is  found  in  the  livestock  section  of this report.
The total population  in  the Gambela Awraja in  1988  given  in  the Russian  Report is 137,000 people.  This would  indicate that the population  living  in  the lower basin  is small as compared to that in the upper basin. The population growth projection is given in Table 3-6-3 below.
TABLE 3-6-3
Population Growth Projection Gambela Awraja
Rural
Urban
Total
1990
1995
2000
2010
139,620
161,230
186,900
251,190
5,310
6,130
7,110
9.550
144,930
167.360
194,010
260.740
Source: Russian Report, 1990
A  preliminary*  1995  estimate  in  the  ARDCO-GEOSERV  Report  indicates  the  existence  of  15 
towns with  papulations exceeding  3000,  10  towns with  population  exceeding  4.000,  7  towns with  population  exceeding  5JOO,  3  Lowns with  populations exceeding  10,000  and  2  towns with 
populations exceeding 20.000. Tne towns w ith populations over 5,000 arc:
Asosa
7.406
Dembi Dola
52,413
Metu
24,293
Mugi
8,148
Gore
8,4] 4
Mizan Teferi
10,238
Tepi
8,414
From the UNDP report,  the town  of Gambia's curr-ent population  is estimated  to  be around 9,000.  However,  from ARDCO-GEOSERV Report,  out of about 50  towns and  villages in  the upper basin,  the majority- of the populated  centers are villages with  populations comprised between  2,000  and  3,000.  Information  concerning  the population  served  by  water and  sewer services is scarce at best.
According  to  ARDCO-GEOSERV.  "The actual number of urban  people who  receive safe (piped) water is estimated  al 29% of the total.  The remaining  71% are presently  using unprotected rivers, dug wells and springs for drinking and washing”.
TA MS-I1 LG  BAKO-A KOBO RIVER B AS IN INTEGRATED DEVELOPMENT MASTER. H AN
Ml3.          WATER RESOURCES
The above would  indicate that about 89,000  people (out of 306.000) are served  by  piped  water in  the upper basin.  However,  from the survey  conducted  by  ARDCO-GEOSERV it would appear that only  some 30.000  would  receive an  adequate supply  and  that most of the rural population  has to  obtain  waler from far away  sources.  Women  and  children  are usually  given the task of bringing water home from natural springs and spend several hours a day an this task.
The ARDCO-GEOSERV report provides no information concerning sanitation in general or
sewerage. Ilowever, this is believed to be non existent in the upper basin.
The   Russian   Report   provides   the   following   estimates   for   the   Gambela   Awraja   water   supply,   in 1986.
Centralized waler supply
8%
(ll,000j
Waler service pipes
1%
(1,300)
Water from laps or outdoor installation
7%
(9,700)
Sewerage
The Awraja has no sewerage system except for Baro-Akobo section in Gambela township.
The UNDP report estimates the population  served  by  water supply  in  the Gambela Awraja at about 5% (10.000)*  lite refugees and  settlers that receive some improved  supplies are not included in the above estimate.
With  respect to  sanitation  the UNDP estimates that the situation  is even  worst; only  1% of the population  would  have any  means of waste disposal and  this would  be mainly  in  some institutional buildings such as schools, hospitals and probably some hotels,
3,6,4 Available Water Resources
At  present  surface  water  resources  utilization  is  limited  Information  on  surface  waler  resources availability is found in both the ARDCO GEOSERV and the Russian Reports.
ITc rivers in  the project area generally  flow from the highlands in  the eastern  portion  of the basin at elevations from 2000  to  3000  m to  the plains in  the western  portion  of the basin  at elevations ranging  from 450  to  500m on  its eastern  extremes to  about 400  to  435  m in  the most westerly part of the basin.  The main  rivers in  the study  area arc the Baxo,  the Gilo  and  the Akobo  rivers. These rivers discharge into  the Pibor on  the western  border with  Sudan.  The Pibor and  the Baro form the Sobat River wliich  is a major right tributary  of the White Nile in  Sudan.  The low lands below about elevation 425m arc subject to flooding.
Flow   measurements
r
are  carried  out  in  some  16  river  gauging  stations  throughout  the  study  area
The  periods  of  records  vary  from  a  few  years  to  long  term  The  stations  with  the  longest  periods of records arc Gambela (since 1905) and Sor (since 1967).
TAMS-lfLG HAROAKOflO RIVER BASfN INTEGRATED DEVELOPMENT MASTER PLANX WATER RESOURCES
Data  presented  in  Table  3-6-4,  below,  was  abstracted  from the Russian study in order to illustrate the magnitude of river flows in the project area
TABLE 3-6-4
Estimated Available Flows
River
Site
Drainage
Area
sq.km
Mean
Annual flow- mJ/s
Monthly flow 95% probability 
m’/s
Birbir
Birbir
Ge ba
Bare
Baro
Barn
Baro
Baro
Gilo
Gilo
Gilo
Alwero
1
2
1
I
2
Gambcla
3
Itang
1
2
3
Abobo
6,840
13,230
6,220
20,970
22,290
22,700
22,740
24,420
7,570
9,640
8,820
630
110
243
132
375
385
386
389
412
115
125
126
20
3,9
7.9
4.0
12.5
13.0
13.2
13.2
14.2
6.1
6.9
7.0
0.5
TAMS-ULG is reviewing  the available data for planning  purposes.  However,  it is apparent that because of the abundance of resources, there should not be a concern as to their availability, with respect to demand for water supply of the population and livestock in the foreseeable future.
In  terms of quality,  it appears that the chemical and  physical characteristics are appropriate for use in  water supply,  however,  from available reports,  notably  the Russian  Report and  the UNDP report, it seems that, because of the absence of sanitation, river waters in Ciambela and elsewhere are contaminated  by  human  and  possibly  animal wastes.  Nevertheless,  the results of bacteriological tests have not been  found  in  any  of the reports reviewed.  More discussion  on surface water resources can found in the hydrology section elsewhere in this report.
At present groundwater is the main source of potable water supply used in the basin. This is due to  the presence of numerous of shallow aquifers in  the vicinity  of many  population  centers and throughout the rural areas.  The latter is evident by  the existence of many  springs and  dug  wells. In  addition,  a limited  number of drilled  wells are present.  Well yields reported  50  far are modest and ranging from 1 to 4 liters per second. There is no data concerning flows at springs.
Several water bearing formations and a regional aquifer are present in the project area, Waler
TAMS-ULG b.aro-akobo river basin Integra ted development master plan
l-*4WATER RESOURCES
quality  data available in  the ARDCO-GEOSERV report suggests that by  and  large groundwater in  the basin  is of good  quality'.  However,  there is concern  about the risks of contamination  due to the absence of sanitation and protection of springs and wells,
A  detailed  discussion  on  groundwater  resources  can  be  found  in  the  hydrogeology  section  ot  this report.
3.6.5 Present Water Supply and Sanitation
The   present   water   supply   and   sanitation   conditions   are   discussed   in   the   ARDCO-GEOSERV, Russian and UNDP reports.
ARDCO-GEOSERV carried  out a socio  economic survey.  As part of this survey,  the current water supply  conditions were examined  in  54  Weredas including,  about 22  townships.  The examination included the following :
, Source of water supply
. Existence of piped w ater
. Adequacy of the supply.
I he wavs in which existing problems may be overcome were also suggested.
I lowever the data was presented in quantitative form and no qualitative data was presented.
Table 3-6-5  presents an  assessment on  the existing  water supply  as reported  by  ARDCO- GEOSERV for 54  Weredas.  As can  be seen  the water supply  is indicated  as adequate in  the case of 20  Weredas and  as inadequate for 34  Weredas Many  problems are suggested.  Other assessments indicated in die same report arc as follows:
Most urban centers have potential alternative domestic water supply sources.
The  present  level  of  supply  is  estimated  at  10  liters  per  capita  per  day,  on  the average.
The   demand   is   hampered   by   the   long   travel   distance   to   collect   and   transport water.
Overall,   there  is   practically   no   shortage  of  water  supply  for  drinking  or  other purposes in the rural areas.
Only a fraction of the rural population (8%) receive potable drinking waler.
Thanks to  the abundance of rainfall and  the low density  of population  most spring are not polluted  and  the quality  of water in  the rural communities in  the area is far better than in many other places in Ethiopia.
TAMS-17LG RARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MAS TER PLAN3. WATER RESOURCES
The latter assessment would appear to be in contradiction with the results of the survey * since in most cases* the water supply is assessed to be ’inadequate11.
In   addition,   ARDCO-GEOSERV   indicates   the   presence   of   several   small   scale   industries, including;
I print shop
8 saw mills
73 coffee pulpers
I 7 coffee washers
10 furniture shops
27 bakeries
1 pastry'
15 sweater works
2 construction materials
There is no  indication,  however,  as to  how these industries are supplied.  Never! lie I ess, it is suggested  that the largest consumer of water may  be the coffee pulpers,  most of which have a capacity of 3 tons of coffee per hour and suggests a total yearly requirement
□f about 1.12 million cubic meters of water (36 liters per second) per pulper.
The  list  of  industries  per  Wereda  and  per  type  of  industry  ,  as  reported  by  ARDCO- GEOSERV, is given in Table 3-6-6.
The  ARDCO-GEOSERV  Report  has  no  indication  as  to  existing  sanitation  infrastructure including, sewerage, cesspools* septic tanks, in the Upper Basin.
The  Russian  Report  provides  the  following  assessment  on  the  water  supply  and  sanitation prevailing in the Gambela Plain in 1988.
TAMS-ULG HaRO-AKOIK) river, basin integrated development master planTable 3-6-5
Water Supply Conditions in each Woredas
lOUrtn q« WMw Supply
Manna to Orerronaa tha Problem
Warned
WtW
tombolas
Rteor
Stora^n
Rasarrofr
Spring
Lteto
Hand
Pump
Other
Loe alter
Vhjkh oT Supply
fa wptte
avpply
Yeo-MHo-O
SpafnQ
□ml
Trwratenp OteXic
CXogtep
Wette
etrxtlve
Maaawra
1
•
*
2 Aswan
1
a
-
-
o
1 3 KuntxA
-
a
4 $«r«
■
1 5 teonai
a
I
6 O«Xj tan
a
a
1
7 Him
■
a
a
1
8 Bve
■
X
a
0
a
X
• Mtwu
a
a
a
Maftj W
Sor -«w»
0
a
10 Hururnsj
C
a
•
a
Huroriv Town
Ground Waler
0
a
a
•
M becno
■
*
X
0
a
-
12 Ate
a
*
X
0
a
a
13 Dau
a
x
a
0
a
a
Htteinj
a
a
X
£><-« Tvwi
r.iDund Waler
0
a
a
a
a
a
Supai 6 Ayr
OtKrd 4 <4»nQ
r
0
a
a
16 V»yV
a
a
a
C.tocnd Watte
0
a
■
a
a
0
a
■
X
a
Sprmp
1
1"
—   ■
a
0
-
a
-
a
0
a
a
| 21 Anten__________________
■
a
a
—
Kkap Town PambDoo Town
f>aund Wtea»
ixrxrid Watte
0
■
a
a
•
-—______
■*Jor Sourcw aF Water Supaly
piped Water
Mmm to OlW*W tfw PreMem
WHlf
| WM&tee
Wear
StOrapWD
Ml
RMftYcdf
Spring
LaU
Hxnil
Pump
Off*
Loetelon        ,|:; ri'
loutcta Sopp^r
fa water
•W*V
rfpqtMie
VrXlNoeO
Spring
0ml
Tnveflhg
U«y
Dlauncew
CMflglng
Wrili
HUp
front
NGcrt
Adm*nl-
etrvdv*
tfeemire
Men* Vjj£i
R
Tum Town
1
Amo
K
■
0
50 Itan
Jt
■
■
D
i
M|
II 51 S%aVa
T
1
+
X
D
i
»
■ II
52 Own.
■
1
S3 GcWrya
*
1
0
X
i
*
M □Oder*
1
X
-
■
1
■
-
Source: ARDCO-GEOSERV, May 1995
■r:TABLE 3-6-6
TYPE AND NUMBER OF INDUSTRIES
Numii.tr ah-J r?iiaorim*4iirl«
W*r*di MniUnn
WmAi
StwMB
CdtfM
Pulpw
■ y' '• ■-■ :*
r
Clift*
W*WW
BMnr
Fully I Swifter
Wfli*
-■'1
CcMwVuriJon.
MUirtU
1 &<7
.
J
r
,
.*
*
J Kiirmuh
r
4 Sira
_
.
1 Np«
*■
• DMulh*
-
-
♦
i     7 Hrtj
,
1
-
*
|i 1 Bura
,
4
f 1 M«lu
r
I
2
■
I
I lOHurum
1
1
*
.
►
+
WM
1
1
l> Ah
1
1
1
2
*
13 M
3
4
T
HDvrtnu
llfrupi t Mi
»
a
w
tlTiy.
1
*
■1
b
+
--
1? Dwranl
*
*
*
11 Siyrfin
3
*
11 lilirwi
1
,
1
-
*
.
i-------------------------
*
! G HARO-AKrinti INA" '•>. ItASININTFCHATFO HFV!!* ni'MCN'T
O’ '.*■I
h
h
•
R
•
•
*
•
*
•
R
R
•
-
V:      ’’
£
i
•
•
•
•
•
•
•
•
•
•
•
•
«
-
•
I [- I
I L L L L
«
•
•
»
•
*
I-
•
•
r
z
1a
i
i
i ►- 1ib
E
□
-
•
•
•
•
•
•
•
•
•
••
•
•
1-
J
!l ’
x;
•
•
l;
<
•
■1
*
•
*
•
£
c
t
*-
•
*
•
R
•
•
•
••
lb
-
•
•
»
••
•
•
•
•
[
•
R
•
I
J1
J
-
•
♦
••
•
•
•
s
E JJ i
•
♦
•
•
1
I
R
J
a
1
■
i
i E
• >
a
i
i
1
fa
i s R* i 0
j
i R
t
[
[
I
1
I1
R
l
3
R
J
i
V
R
i TS
3 n
J
O
R
R
£
1
i
*
i
E
P X
i
□
i
c
fi
I
: 5
■
9
1
R
1a
uV*
a
>
9
•
•
9
1
J
1
•
• X
E
9
I7
<
£ EZ
s-
2 u
t- Li Jr i
i
i
II
L
"The analysis of the data and material supplied by the Ethiopian construction and exploitation^
organizations, engaged in waler supply and occupied with the material on actual inspection of water*"
supply in the Awraja settlement, showed that absence of engineered water supply and sewerage
ssys ystteemmss i iss one    one of  of t thehe m maajjor or f feeaattur ureess pr  preevevent ntiing f   ng frromom i immpr prove  ovemmeent nt t thehe l leevevell oi  oi l liivi ving   ng
of the population.
The considerable water resources of the Gambcla Plain arc used very poorly for water supply. The major part of the rural population receives water mainly from surface and
groundwater  sources,  only  the  Gambcla  and  Itang  cities  and  resettlements  arc  partially  supplied  wii
water   suitable   for   drinking   by   its   chemical   composition.   But   its   organoleptic   and   microbiologic:
indexes do not correspond to (acceptable) drinking water standards.
pollution.   The   water   intake   is   of   simplified   type:   submersible   pump,   submerged   directly   near   the
bank, supplies water to the chlorination plant and (without treatment) the water is delivered to watef
supply  network  The  reservoir  envisaged  on  the  Gila  mountain,  is  not  under  operation  that’s  why1
constant breaks of water supply take place there. In the dry' season, when water level in the Baro River drops, the submerged pump is replaced by a centrifugal pump. The city has several waler
which cannot provide water to the whole population. People who live in the outskirts of the city use
w-atcr directly from the Baro River. The city’ has no sewerage system. The above written, as well as-
absence of sanitary control for water quality in the city create pre-conditions for possible epidemics J.
-
L 
The  level  of  water  supply  in  the  Gambela  plain  estimated  in  the  Russian  Report  is  23.22  liters  per
capita   per   day   (about   10%   would   be   for   local   industry).   Further,   the   same   report   provides   thf
following assessment of the present (1988) conditions:                                                                                            I On the average for the Awraja only 8% of the population is supplied with water frorf
centralized waler systems. 
One (1)% of the population has water service pipes.
Seven (7)% of the population uses water from outdoor water taps.
S
Ninety-two  (92)%  of  the  rural  population  takes  water  from  the  water  sources  manually  or uses primitive water lifting means,
r
rhe number of "holes" and wells identified included
[
r
TAMS-ULG     BARO-A KOBO RIVER BASIN INTEGRATED DEVE1OPMENT MASTER P|_AN. 11 pumps
♦ 72 manual pumps
The length of existing water pipes is 12.5 km
Concerning water quality, the Russian Report provided the following information.
In  the wet period  due to  the Pibor River backwater,  water of the rivers overflows tire flood plain  west of 34*15  Meridian.  As a result,  waterlogging  takes place thus impeding construction  of waler supply  systems.  Comparatively,  high  water turbidity  up  to  195  mg/l is a characteristic of rivers.
The   highest   degree   of   water   turbidity   is   observed   during   the   rainy   period   and   under maximum flood, when turbidity values exceed 500 mg/l.
The  lowest  degree  of  turbidity  is  in  the  dry  period  (November-April)  and  makes  up  to  10 mg/l iter
From the point of view of salinity, the w'ater of the Gambela plain rivers can be
characterized as hydro carbonaccous-sodium water. Water is fresh and its mean value of
salinity is up to 0.5 g'l and, according lo the value of total hardness (less than 1,5 
mg/equiv.,1), it is referred to as "very soft water".
By  concentration  of  hydrogen  ions  water  changes  from  slightly  acidic  to  slightly  alkaline. pH is between 6 15 and 7.9,
Indices  of  water  quality  show  that  the  water  of  the  Gambela  Plain  rivers  is  suitable  for domestic water supply but on condition that it is treated by traditional methods.
The  UNDP  report  gives  a  detailed  account  of  conditions  prevailing  in  the  Gambela  plain.  This  can he summarized from the following quoted verbatim from this report:
"'Gambela is well endowed  with  water resources.  Its rivers produce a yearly  flow of over 20  million (billion) cubic meters of run  off.  Groundwater is available almost everywhere with  the water tabic ranging  from 0-I 50m.  rhe average daily  rainfall is about 1200mm.  Despite (he presence of these precious resources,  the people suffer from acute shortages of safe drinking  water.  This is due to  the fact that most water sources are not developed,  are highly  polluted  and  far away  from the users- Feiching  the available water is associated  with  hardship  and  difficulties.  The tedious task  is left to women and children.
fhc sanitation problem is even worse than the water supply. There is practically no sanitation
TAMS-ULG BAROAKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER I't AN
355facility  in  the entire region,  except a limited  number of pit latrines located  in  institutions like schools,  clinics and  a few enlightened  individuals'  residences.  Due to  the absence oi potable water and  appropriate sanitation  facilities,  most of the diseases in  the area are reported  to  be either water bome or waler related.  Lack  of improved  water supply  and  sanitation  facilities among  other factors have greatly  contributed  to  the backwardness of the region  and  are believed  to  be hindering  progress and sustained development of the regional economy 11.
Unquote
Further, the UNDP reports the following concerning the Gambela waler treatment plant in 1993:
gwofe.
" The treatment plant imported from abroad for Gambclu town which is seating idle while the people*- arc suffering from acute potable water shortage is a good example of the problems associated wiihL complcx systems".
llie UNDP report estimates the average consumption rate of the people in Gambela Region al belovL 10 liters per person per day. It also indicates that, in almost all cases, the people consume raw water without any type of treatment.
With respect to sanitation the UNDP report indicates there is almost complete lack of sanitation facilities except for a small number of latrines as previously described. It also states that humarjT excreta is the chief cause of pollution of the water resources In the town of Gambela, people are seen defecating in almost the center of town. Neither NGO's not the governmental organization,^ consider the inclusion of at least safe excreta disposal scheme in their programs.
L
L
The  present  water  supply  situation  as  reported  mainly  by  ARDCO-GEOSERV  is  presented  graphically  orp Figure 3-6-1
3.6.6 Present Tariffs
C
No  indication  on  tariffs  or  the  tariff  structure  for  water  supply  with  piped  water  systems  and/or  with  treated water supplies w as found in any of the 3 reports
3.6.7 Water Demand Projection!.
The   World   Health   Organization   (WHO)   recommends   the   following   daily   supply   rates   for   developing^ countries.
150  liters  per  capita  per  day  for  cities  and  towns  with  developed  distribution  systems  and house connections and houses with sanitary' facilities and to cover town needs.
50 liters per capita per day for rural villages with primary water systems and sanitation
TAMS-ULG r aro-akoejo river basin integrated development master plan
3-56
[
[
I?facilities
20  liter  per  capita  per  day  as  a  minimum  to  meet  the  water  needs  of  the  population  with common (group) water taps and elementary group sanitation schcmcs.
Minimum waler quality standards arc also recommended by WHO
ARDCO-GEOSERV,  taking  into  account the present status,  local conditions and  the primary'  rural character
of  the  project  area  (Upper  Basin),  has  proposed  adopting  the  following  criteria  over  the  period  from  1995 
to 2025.
Urban areas
Starting with about 30 liters per capita per day in 1995.
2 liters per capita per day increment every 5 years.
Ultimate level of 56 liter per capita per day.
Rural areas
Starting  with  the  recommended  minimum  of  20  liters  per  capita  per  day  and  allowing  for a modes! increase over the planning period.
Attain 30 liters per capita per day by 2025.
Based  on  the population  growth  projections discussed  in  section  3  6.3,  and  on  the above referred criteria ARDCOGEOSERV has estimated  the w-ater consumption  requirements for the urban  and rural population  in  the Upper Basin  for the period  of 1995-2025  at 5  year intervals.  The result of the projection is given in Table 3-6-7, below;
TAMS-ULG   &ARO AKObO HJ VER BASIN INTEGRATED DEVELOPMENT master plan
3-57Table 3-6-7
Projected Water Consumption Requirements
in cubic meters per year * ’
2
1995                   2000                   2005                 2010                 2015                 2020                2025
UURBAN
N POPULATION
201204               291160               433446             663459           1040508           1579849          2277497
YABUS                                          68134                 98597               146780             224670             352352             534992             771240
DEGA
I MIDDLE BARO
I UPPER BIRBIR
LOWER BIRBIR
125571
181713               270513             414063             649379             985981           1421382
926276 1340406 1995442 3054342 4790149 7273094 10484827
595688 862013 1283265 1964241 3080536 4677313 6742773
815964 1180771 1757796 2690586 42*9669 6406909 9236142
SOR & GEBA 72470 104870 156119 238965 374771 569032 820311 1 BARO 0 0 0 0 0 0 0 « UPPER ALWERO*" 0 0 0 0 0 0 0
W LOWER ALWERO1'
I UPPER GILO
330944 478905 712938 1091265 1711441 2598556 3746055
340209 492313 732699 1121819 1759358 2671310 3050937
UPPER AKOBO                                  0                        0                         0                       0                      0                       0                      0 H TIRMAT1D1"
3476466            5030752            7489202         11463415         17978167        27297041         39351167
TOTAL
1 RURAL POPULATION                1297544            1583242             1936540           2374425           2918104           3588358          4412560
U YABUS
B DEGA
MIDDLE BARO
UPPER BIRBIR
LOWER BIRBIR
SOR & GEBA
707918 863791 1056544 1295446 1592069 1957748 2407419
439216              535925              655515             803739             987774           1214653          1493645
4806424 5864721 7173421 8795454 10809375 13292162 16345210
1949552 2378811 2909638 3567557 4384432 5391484 6629843
3667228 4474692 5473210 6710797 8247388 10141715 12471144
1116612 1362471 1666504 2043330 2511197 3087989 3797263
BARO                                                  0                        0                        0                      0                       0                       0                      0
UPPER ALWERO"                             0                        0                        0                       0                      0                       0                      0
LOWER ALWERO'"               2231596            2722959            3330582          4083684           5018738           6171482          7588997
UPPER GILO
2369963            2891791            35370BB          4336384           5329914           6554132          8059537
UPPER AKOBO                                 0                        0                        0                      0                       0                       0                      0
TIRMATID”
TOTAL
10586059          22678407          27739046        34011321         41798996         51399727        63205623
Notes:These areas are located in the Lower Basin, Pl Losses are not included Source: ARDCO-GEOSERV, May 1995
TAMS-ULG    BARO-AKOBO RIVER BASIN INTEGRA TED DEVELOPMENT MASTER PLAN
3-58In  addition.  ARDCO-GEOSER V made an  estimate of present water requirements of livestock  and poultry.  The estimated  requirements shown  a Table WS-8  are based  on  the following  unit consumption per head in litcrhead/day (according to U.S. Standards):
l/head/d>y
Cattle Goats Sheep Equines Poultry
45
3 
21
12
0.1
Table 3-6-8
Livestock Water Requirements by Subcatchment
in cubic meters per year
CATCHMENT
CATTIE
SHEEP
GOAT
EQUINES
POULTRY
YA0US
DEGA
033317
10749
13473
A11M4
9977
12565
5169
MIDDLE BARO
$97556
30300
10555
873
UPPER BIRB1R
4G21651
27957
57330
1615
325
1276
21500
LOWER 0IRHIR
110192
1434822
20581
SOR & GE BA
5681
3710111
4911&
24002
33558
BARO
1005522
87604
36905
UPPER ALWERCK"
25109
15636
52433
161
7340
2495
5191
679
145
LOWER ALWERO '
,,
15630
227
161
UPPER GILO
148
1997493
227
51224
UPPER AK060
41591
22
22
1337011
21339
23569
sm
TlRMATID'1’
TOTAL
3735
0
28951
0
O
0
15688663
273126
151024
393215
GRAND TOTAL
3311
0
25372
16532402
NOTE. P!Tha» arm ar© located in th* Lower
TAMS-ULG    liARO-AKO BO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PI
i-S9The   Russian   Report   proposed   several   alternatives   for   mean   daily   waler   supply   and   sewerage rates per person as presented in Table 3-6-9, below.
Tabic 3-6-9
Mean Daily Water Supply and Sewerage Rates per Person
Rate in liters/capito/day
Type of development
Water Supply
Sewerage
1. Building of houses equipped with inner water supply and sewerage systems without hath
125-160
125-160
2. Building of houses equipped with inner water supply and sewerage systems and bathrooms
with water heaters
160-230
160-230
3. Building of houses equipped with inner water supply and sewerage systems and with central water heating systems
230-350
230-350
4. Water use from water taps of
outdoor installation, public
and individual toilets
30-50
25
5. Water use from local sources
20-30
-
6. Green Plants watering
20-50
Types 1.2  above include piped  distribution  systems.  The rate of water demand  covers water discharges for domestic and  drinking  needs of local industry.  Choice is to  be made depending on design stages.
The  daily  per  capita  rates  for  water  and  sewerage  adopted  in  the  Russian  study  for  the  period of 1990 to 2010 at 5 year intervals are presented in Table 3-6-10.
TAMS-ULG R ARO-AKono RIVER HASTN INTEGRA TED DEVELOPMENT MAS EE R Pl JUN
J-6CTable 3-6-10
Adopted daily per capita water supply and sewerage rates
(liters per capita per day)
Year               Wafer Sewerage
1990 
33.9 
1.8
1995 
40.6 
9.1
2000 
52.4 
2.1
2010 
110.0 
79.0
2. Outdoor taps
0.264
0.55
0,92
2.38
3. Local Sources
1.397
The resulting water and sewerage demands projected in the Russian Study are presented in Table 3-6-11, below.
Table 3-6-11
Water and Sewerage Demand Projections
(in million cubic meters per year)
Water                               1990        1995          2000       2010
1. Piped waler systems 0.132             0.555         1.36      7.661
Total Water              1.793       2.479         3.597   10.469
1.374
1.317
0.428
Sewerage
1990
1995
2000
20/0
L Integrated Systems
0067
0,397
1.133
6.566
2. Local system
0.026
0.152
0,354
0.952
Total Sewerage        0.093       0.549        1.487    7.518
Water consumptive me          1.70        1.93          2.11     2.95
In  addition,  the Russian  Report projected  the water and  sewerage requirements for various planned  settlements,  agribusiness (agro-industrial) developments.  The total requirements estimated for these demands are given in Table 3-6-12, below:
TAMS-ULG BAROaKOBO river BAStN INTEGRATED development master plan
1-61Table 3-6-12
Settlements and Agro-Industries.
Water and Sewerage Demand Projections for Planned
(in million cubic meters per year)
I
Water Suppty
Sewerage
1995
2000
2010
1995
2000
2010
1 Alwero irrigation
Settlements
2. Ubeia Pacha la
Irrigation Settlements
3. Ralnled lands
Settlements
4 Gambela Agrolndusthes
Sub Total"•
1 5. Sugar Stale Farm Settlements
1 6 Sugar Refinery
Sub Total Sugar Compie*
0 28
0
026
2
0 54
0
2
0
056
0
052
Q
1 08
0
2
0
056
2 20
052
112
4 38
1.00
28 1Q
29 10
013
0
009
2
0 22
0
2
0
0 26
0
018
2
044
0
Q
0
0.26
1.10
0.18
O.$5
209
0 50
IS.Sfl
20.10 L
Grand total *
054
1.08
3348
0 22
044
2219            |L
<n    Minimum  Development  Alternative
12)   Maximum  Development  Alternative
p ' Demands in 1990 were zero
The present (1990) water demand for livestock and poultry was estimated in the Russian Report as shown in the following Table 3-6-13.
5 r r r
[i: [  r. tr [
c
r
TAMS-ULG    BARO-AKOBO RIVER BASIN INTFGRATED DEVELOPMENT MASTER PLAN
[
r
r
1-62
IL, iG
JI ■Table 3-6-13
Pre,ent Water Demand for Livestock nnd Poultry
(Lower Basin, 1990)
Waler Demands
Animals
Quantity 1000 head
Unit Daily water
Demand l/d/hcad
mVday
Million mVyear
1. Cattle
2. Sheep
3. Goats
4. Poultry
321
77
103
139
70
8
6
1
22,470
616
618
139
8 20
0.22
0.23
0.05
Total
-
-
23,843
8.70
3.6.8    Costs
Investment Costs
No information on costs is given in either the ARDCO-GEOSERV in the or UNDP reports.
The Russian Report provide some data on unit investment costs (1998) but these were difficult to interprete and in any case are outdated.
3.6.9    Water Supply Projects
No specific project are indicated in either the ARDCO-GEOVERV or in the UNDP reports.
In the Russian Report a water supply scheme for Gambela is suggested. The scheme includes the following.
Intake works comprising an inlet, a gravity tine, an intake well, a suction line and a flushing line;
A low head pumping plant and a low pressure conduit; Treatment works (filtration plant);
A treated water reservoir at ground level, and
TAMS-ULG baro-akobo river basin integrated development masti r plan
3-€3A   high   head   pumping   plant   from   which   treated   water   would   he   fed   to   a   pressure water line and elevated water tank (not shown in the flow diagram).
The project would  be designed  for the 2010  population.  The average design  flow would  be 4.38 million  cubic meters per year (12.000  cubic meters per day).  The quantities and  the estimated  cost of the scheme were not identified.  The size and  length  of pipes in  the scheme and  the capacity  of the storage nor the distribution  characteristics were indicated  and  their cost was not estimated.  O &. M costs were not provided.
In  addition,  the Russian  Report recommends the installation  of combined  (storm plus sanitary) sewage system and  a biological sewerage treatment plant.  However,  other than  the schematic diagram for the sewerage treatment plant,  and  some estimates for the composition  of waste waters, no  indication  is given  on  the project characteristics,  its estimated  investment cost or the O & M costs.
Likewise several sketches arc given in the Russian report for:
1.          Groundwater (well) based wutcr supply intake treatment, pumping and elevated tank for proposed settlements (2 alternatives)
2.           Schematic diagram for a wastewater treatment scheme for settlements.
3.           Schematic diagram for watering points (3) alternatives for pumping from wells to reservoirs and gravity supply to a watering pond, based on use of solar energy.
Again, no other information or project specifics was found.
T.A-MS-VLG BARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
1-MSECTION 4
SOCIO-ECONOMICS4            SOCIO-ECONOMICS
4.1         Introduction
The  Baro-Akobo  basin  has  been  the  subject  of  four  studies  conducted  for  the  purpose  ol  water resources development.
1.   The  Baro-Akobo  Master  Plan  Study  of  the  Water  and  Land  Resources  of  the  Gambcla Plain,  dated  1990,  was  conducted  by  Selkozprom  Export  of  the  USSR  and  referred  to as  the  Russian  Study  and  covered  the  Lower  Basin.  The  Russian  study  makes  only  the briefest   reference   to   the   socio-economics   of   the   area,   merely   identifying   the   tribes which   inhabit   the   plains.   It   does   include   a   brief   mention   of   the   epidemiology   and likely effects of the proposed project
2.    The   Reconnaissance   Study   Report   on   SocioEconomics   and   Environment.   dated   June 1990,  was  based  on  field  work  carried  out  by  the  Environmental  Conserv  ation  Studies and  Research  Desk  of  EVDSA  at  the  request  of  the  Boro-Akobo  Basin  Master  Plan Project  Office.  The  report  contains  a  limited  amount  of  basic  data,  including  Awraja based  estimates  of  population  and  some  statistics  on  health  and  wraier  supply.  Il  is refcred to below as the Reconnaissance Study
3.    The  Socio-Economic  Report  of  the  Upper  Baro-Akobo  Basin  prepared  by  an  EVDSA in-house  learn  which  was  appointed  to  start  work  in  June  1993  under  comprehensive terms  of  reference  dated  May  1993.  Their  work  was  intended  to  complement  a  study by  ARDTCO  (sic)  of  the  Upper  Basin,  which  covered  the  upper  basin,  but  .specifically excluded     environment     and     socio-economics.     The     study     was     abandoned     after publication  of  an  inception  report  following  desk  studies  in  Addis  Ababa  and  a  10  day site   visit   in   June/July   1993.   The   report   is   accompanied   by   unfinished   tables   which show  clearly  that  a  very  thorough  appraisal  had  been  planned.  The  team  was  however disbanded  by  the  time  the  report  was  published  in  1994.  It  is  referred  to  below  as  the EVDSA Study.
4.    The   Survey   and   Analysis   of   the   Lepper   Baro/Xkobo   Basin   by   ARDCO-GOSERV dated  May  1995,  This  report  includes  a  population  estimate  by  warcda.  using  1984 census  data  updated  to  1990.  It  also  contains  data  and  analysis  of  a  limited  scope survey   conducted   in   892   households   in   29   locations.   The   survey   was   focused   on energy   requirements,   but   gives   some   useful   additional   information   on   age   structure, occupations,   household   income,   time   spent   in   fuel   collection,   energy   prices,   travel times  by  transport  mode,  and  time  spent  on  agricultural  activities.  It  is  referred  to below as the AG Study.
Further information  for the preparation  of tins preliminary  report on  data availability’ has been obtained  from Master Plan  Reports for the Rift Valley  and  Omo-Gibe Basins,  a UNDP sponsored 
•‘Gambela  Needs  Assessment  Survey",  the  “USA  Statistical  Abstract  1992*  and  the  'Report  on the  National  Rural  Nutrition  Survey,  Core  Module,  March  1992*  as  well  as  other  publications as available (see bibliography)
TAMS-ULG DARO-A KOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTERPLANJ SOCIO-ECONOMICS
Maps are not included in this report since, al the lime of writing, the project GIS system has not been set up. The GIS system will assist the preparation of more detailed presentations or the data currently available, and that yet to be collected in the course of the study.
4.2       Local Government Administrative Structure
4,2.1 Current Structure
The Ethiopian  Government administrative structure has undergone many  changes in  the last decade.  Following  a proclamation  issued  in  January  1992  by  lhe Transitional Government,  the administration been based on a four tier system with representation at the following levels;
National (Central Government)
Regional (14 Regional Sell-Govemmcnts)
Awraja
Were da
lhe regional government comprises, at a minimum, the following organs:
Council directly elected by lhe people.
Executive Committee
Judicial Organ
Public Prosecution Office
Audit and Control Office
Police and Security’ Office
Services and Development Committee
These organs are equally  established  at wereda level.  The wcreda administration  has the power to  prepare,  determine and  implement within  its own  areas plans concerning  social services and economic development; as well as to  implement laws,  policies and  directives of the Central Government and the National/Reginnal Self-Governments.
4.2   2 Boundary Changes
The 1984 census was conducted at a time when the country was divided into 14 regions, and 
further sub-divided into awrajas and wereda. At this lime the wereda was the lowest tier of
administration. In 1987 Lhe regional boundaries were redrawn into 27 regions, and subdivided 
into awrajas, and the wereda tier of administration was abolished. The awrajas were the centres
for local government, courts, prisons, police and taxation. Schools, hospitals and line ministry 
offices were usually located in awraja towns. These awraja boundaries form the basis for much 
of Lhe data collected in the preparation of the project.
[n  1992,  lhe boundaries were again  changed,  with  14  new regions based  essentially  on  linguistic grounds.  These regions do  not correspond  with  the 14  regions abolished  in  1987  The wereda has again  become the lowest tier of government,  with  provision  for election  of local government officials.
tams-ulg BaRO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
4-24            SOCIOECONOMICS
Table 4-1  shows the history  of the administration  of the warcdas in  the project area throughout these changes,  llie project will need  to  have accurate maps showing  the regional and  wereda boundaries under the present administration  for G1S use*  so  1  hat statistics collected  al different periods can be assimilated and compared
4.23 Associations
Following the overthrow of Haile Selassie in 1974 by a provisional military administrative
council known as the Dergue, a program of socialist reforms was initiated. Up to 30*000 small
organizations were formed within a few years. All Ethiopians above the age of 18 were obliged 
under the law to belong to cither in rural areas, Kebelc Peasant Associations (KPAs), organised 
on an area of approximately 800 hectares, or, in towns, Kebele Urban Dwellers Associations
(KUDA’s).
These associations arc basic,  grass roots units of formal political*  social and  economic association.  They  had  the function  of bridging  between  the slate and  the people for the implementation  of national laws*  policies,  plans and  development projects undertaken  by  the stale,  lliey  were discredited  due to  their role in  conscription  and  corvee labour during  the Ikrgue.  but after many  changes in  leadership  they  were again  recognised  as having  an  important role to  play,  and  1992  they  were reinstated.  It remains to  be seen  how they  function  politically now' that the w ereda level of government has been reinstated.
r
Apart from their political role,  the associations also  act as forums for the community  discussion and  action  in  expounding  community  needs to  the administration  and  projects financed  by International Agencies and NGO's,
Each  Kebelc has an  elected  committee accountable to  its general assembly  and  a number of sub committees for such  activities as social and  developmental matters.  Other single or multi-purpose organizations,  such  a women  s and  youth  groups,  operate under the umbrella of die Kebele associations.  The individual kebcles arc organised  hierarchically  to  form associations at all administrative levels (wereda,  awraja,  regional and  national).  To  the extent that they  are still operational,  the Kebele can  obviously  play  an  important role in  assisting  communities to  have a voice in  development.  They  could  also  play  an  important role in  disseminating  information  for health and agriculture.
4.2.4     Other Institutions
Producers  co-operatives  were  created  under  the  Dergue  but  were  resented  and  disbanded  before its collapse. They can no longer be considered relevant factors in future development.
Service Co-operaiives were indirectly  responsible for the provision  of services,  agricultural inputs and  consumer goods to  farmers.  They  have surviv  ed  in  many  areas but seldom provide the type and  quantity  of services needed  Until an  efficient market-oriented  distribution  system has developed, they probably meet some basic needs and therefore still have a role to play
State Farms face an uncertain future. It is unlikely that they will be preserved and a decision is
TAMS-ULG          HARO-A KOBO RIVER BASIN INTEGRATED DEVELOPMENT MaSTFR PLAN
4.34           SOCIO-ECONO MI CS
awaited on whether they will be privatised or turned over to local communities.
Private Investors/Inslitutions arc beginning  to  appear tn  the agricultural sector,  although  al present mosi activity  is concentrated  in  Addis Ababa The investors obtain  concessions from “unutilised" areas,  which  can  mean  the dispossession  of existing  settlers or pastoralists from communal grazing land.
4 J       Ethnic Groups
The  Reconnaissance  Report  and  the  Russian  study  identity  the  ethic  composition  of  the  peoples living in the upper and lower basin as follows,
Region          Agriculture                Other Sedentary'        Semi-
Nomadic
Ilubalor         Oromo. Amhara|          Mazanger
Uncertain
Kefa              Oromo, Amhara, Kefa, Gimira Shuro/Tama (Mesengo)
Shekocha, Anwak
Mazanger, Shuro
Me'enil
Sheko
Bench
Welega          Oromo. Amhara          Mcscngo, Mao Asosa
Gambele
Anwak (48%) Malanger
Gidame, Begi. Dul
Nuer (38%)        Anfilo. Rome Masongo
L.onge, Opuo
In  addition,  then: are settler groups from other parts of Ethiopia,  mostly  practising  agriculture or engaging  in  trade.  By  agriculture is meant farming  and  animal husbandry.  "Other sedentary* includes farming,  hunting,  bee-keeping  and  root cultivation.  Semi-nomadic includes seasonal pastoralism and shifting cultivation.
The information  available is inadequate to  asses the importance of ethnicity  and  lifestyle in relation  io  integrated  development.  Ethnic and  cultural values must be taken  into  account when considering  different interventions.  However,  way  of life may  be more important that ethnicity al a planning  level Ethnicity  will however affect lhe detailed  implementation  of plans for settlement, methods of collection and dissemination of information.
An  anthropological  study  is  planned  to  assess  die  effects  of  the  proposed  project  developments This will include a brief description of the peoples, and a description and analysis of land tenure.
tams-ulg    BARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
4-44           SOCIO-ECONOMICS
livestock  and  pastoral ism.  the relationship  between  people and  proposed  parks in  the lower plains,  in- and  out migration  issues,  the relations among  groups,  and  gender issues.  The report should  also  cover the presence of any  cultural artefacts or belief systems which  should  be considered when planning particular developments and any resettlement.
4.4       Population
4.4.1 Census
The last population census for which results are available was conducted in 1984 and results are available from the Central Statistical Authority.  A census was carried  out in  May  1094  and  the full results are not yet available. However, some Irasic preliminary statistics were obtained by the project from the CSA.  These have been  updated  to  1995  estimates using  the 1984  census data and  the rable of population  projections as set oul in  Table 4-2.  Overall population  is projected to almost double (factor 1.97) in the period 1995 to 2015.
The best methodology for estimating population is included in the A-G Report and this has been used with updated (1985) figures in Table 4-1 All the 71 wcredas (68 named, of which 3 are divided by regional boundaries) completely or partly contained within the basin area are included, and the pru portion of each wereda lying within the area has been used to estimate the rural population on a pro rata basis. The urban populations have been included on a yes/no basis depending on whether the principle town does or does not lie within the basin.
A summary by region is also set out in Table 4-3. which shows;
Region % Area
6
8
% Population
8
4
31
60
11
26
25
12
35
7
Total
100
too
72,315 km:
2.91 million
These estimates are satisfactory for planning purposes at this stage, but should be revised when recent census results are fully available and new wereda boundaries are entered into the GIS. In particular,  the high  proponion  show'n  for the urban  population  (18%) of Region  6  (Gambcla) deserves investigation.
Possibly excluded from these estimates is the huge influx of refugees from Sudan who straddle the border.  A master plan  for the development of the area would  not normally  include such people as part of the planning process, but the fact of their existence will undoubtedly affect the implementation of any plan and the possibility of a semi-permanent refugees population in this area should not be overlooked.
T AMS-ULG BARO-AKOBO JUVFR BASIN JNTEGRATEDDEVELOPMENT MASTER plan-1         SOCIO-ECONOMICS
According  to  informalion  supplied  by  UNI ICR,  the bulk  ot the refugees who  arrived  in  the 1980 s departed in 1991. but since 1993 there has been a steady return due to on-going unrest in Sudan.  Il is unclear to  what extent these people are fully  reflected  in  the 1994  census.  The UNHCR Project Manager advised that there arc al present (October 1995) two main camps at:
Fumido,  50  km from Abobo,  where there are 29-30.000  refugees,  mostly  Nuer.  Some 200 ha of rainfed  crops arc grown.  Water supplies from 16  tubewclls 40-60  m deep  fail to provide adequate supplies, and water has to be trucked in.
Bong,  40  km  from  Gambela.  where  there  are  16,000  refugees,  mostly  Uduk,  Some  300 
ha  of  rainfed  crops  are  grown.  Boreholes  20  m  deep  provide  some  water.  In  both  camps, livestock are provided by UNHCR.
Population  densities in  the lower plain  should  be treated  with  extreme caution  Much  of the area is seasonally  flooded  and  the Anuak  practice seasonal migration  in  search  of better soil to cultivate as well.  Despite their attraction  as areas with  low population  densities inviting development, there are major constraints to be overcome before these areas can be settled
4.4.2     Demography
None of the reports give any  significant indication  of the demography  of the area.  However, there is a significant amount of data available from the Population  Analysis and  Studies Centre The Rational Family and  Fertility Survey Repart (CSA 1990).  This and  Other documents will be used in the analysis of population trends to be carried out under this study.
4.4.3     Population Support Capacity Studies
In 1988, FAO published the results of an assessment of the population support capacity (TSC)
as part of their Master Land Use Plan for Ethiopia. The PSC attempts to determine the maximum number of people the area could support, subject to meeting requirements for food (cereal
equivalent), bio-mass fuels, and oxen far agricultural inpuls, and recognizing the constraints
imposed by factors such as lite prevalence of vertisols and tsetse fly.
The results of this study  arc available in  graphic form,  and  from these Tables 4-6  and  4-7  have been  prepared,  bused  on  projections for the years 1995  and  2010.  The index  tabulated  indicates the theoretical number of people which  the region  could  support (the PSC) exceeds the projected population. Under the base case, the index is positive in 1995 for all areas except Gambela By
2010.5   out of the 11 areas will be negative. The study also examined the effects of relaxing some of the constraints, and it s apparent from Table 7 that only by eradication of tsetse fly can all the regions remain positive. This is confirmed in Table 8, where a comparison is made of alternative interventions  and  the  optimum  combination  of  all  possible  interventions  In  almost  every  case, the eradication of tsetse contributes the most to the optimum solution.
4.5        Health
Health facilities in the area are extremely limited. Table 4-4 lists the medical facilities listed hv
TAMS-ULG      baro-akobo river basin integrated development master plan
4-44            SOCIO-ECONOMICS
region  in  1992,  based  on  CSA figures.  This Table shows very  different estimates from project documents.  In  principle,  the facilities per capita as compared  with  lhe rest of the country  outside Addis Ababa can  be compared  once population  by  regions has been  estimated,  in  order to determine whether the situation  in  the basin  is significantly  worse than  elsewhere.  However,  the diiferenccs between  Tables 4-4  and  4-5  show that,  to  be meaningful,  such  a comparison  would have to be made using better data than that available from CSA.
The Reconnaissance Report lists lhe health  institutions and  staffing  levels by  awraja in  the llubabor,  Kefa and  Walega regions (Asosa was avoided  for security  reasons),  but the Russian report gives no  listings for Gambela.  The findings are shown  in  Table 4-5.  The Reconnaissance Report notes the main diseases (as %) as follows:
Disease
llubabor
Kefa
Walcga
Helminthiasis
15
10
17
Skin diseases
10
6
10
Rheumatism
9
6
6
Gastric diseases
8
4
4
Anaemia
8
3
Respiratory diseases
6
7
9
Malaria
5
5
6
Dysentery/Diarrhoea
5
6
2
Malnutrition
4
Sexually transmitted
3
2
Unspecified fever
■>
3
Influenza
4
Eye diseases
3
Total (op 10
73
51
60
The Reconnaissance Report notes dial these is no  diagnostic facility  in  the basin.  Laboratories arc found  in  the hospitals and  health  centres,  but 5  (35%) of them have no  technicians.  All hospitals have X-ray facilities and all health centres delivery facilities.
The Russian  Report contains a section  on  the epidemiological situation  in  the lower basin,  and thus a comparison  is possible between  the two  areas.  Salient points from the two  reports are tabulated below.
As the Russian  Report makes clear,  any  irrigation  scheme will have io  make extensive provision for medical facilities and  preventative medicine on  a very  large scale,  and  as this is likely  to  be expensive,  further investigation  is required.  Additional data should  be collected  from refugee camps and  UNHCR staff to  assess the likely  effects of disease on  settlers.  In  particular,  the reports of the development of resistance of malarial parasites to  cutTent drugs should  be checked as Ethiopia has been included in die region where this has been reported.
TAMS-l'LG          B.ARU-AKOBO RIVER BASIN INTEGRATEDDEVEljOPMEM MASTER PLAN
4-74 SOCIO-ECONOMIC S
Point Malaria
Upper Basin
One of the leading causes of morbidity Ma I ana transmission is mien sc after the rainy seasons. but near swamps and HEP projects, transmission is throughout (he year Control is attempted with DDT spraying once or twice year.
Oncho-
ccriasis
Lower Basin
Cerebral malaria is lhe prevalent parasitary disease It is wickspread in the plain and is associated with a high mortality rate
Peoples living m lhe area will have developed resistance to malaria. If non-immune settlers are brought In Io irrigation schemes, mortality rates will be high during the lime (many years) that it takes for immunity to develop. Dry season storage will dramatically increase (he incidence of malaria within a radius of 5 km Spraying against anoph t line mosquitoes with persistent insecticides, treatment of infected people and chemical prophylaxis will all be necessary Important
Yellow fever
T rypano- somyiais
Endemic in lhe lower lying parts of the Upper Basin, below 2000m Infection rates of up to 90% have been reported, The vector is lhe Simulium fly which breeds in fast flowing and shaded streams.
Endemic over (he ccniral part of the basin, in the form of massive outbreaks A fatality rote of 30% among 100,000 coses was noted in 1959-62. Ng outbreak has occurred since 1^89. Evidence of sub-clinical infection The monkey population provides a reservoir for infection.
No recorded case* of human infection
Schisto
somiasis
Hclmintho ses
Reported b) the Instiruie of
Pathobiology has prevalence of up to 5%  rn lower areas, particularly Bcbcka coffee plantation.
The most leading cause of out-patient visits.
Common among cattle, it is also found in the human population, but the rate is low at one case per year. Tsetse fly population may increase with the increase in moist vegetal ion, but this is uncertain (since scrub clearing will reduce lhe available habitat)
Not menu uned
Other
Skin infections, respiratory diseases, rheumatism, gastritis and gastro enteritis, diarrhoea, sexually transmitted diseases and dysentery feature in lhe lop ten most common
Account for 30% of all infectious diseases.
Geohelminthosis will be spread by poor sanitation practices,
and it is difficult to change cultural practices io avoid this.
Close monitoring of lhe health of lhe settlers will be
necessary
Intestinal diseases listed art bacterial and amoebal
dysentery,  typhoid  fever  and  infectious  hepatitis,  transmitted through water, vegetable and fruit.
TAMS-ULG BARO-AKDBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN4            SOCIO-ECONOMICS
The data above does not permit an  integrated  master plan  to  make recommendations which justify  particular types of intervention.  An  analysis of community  health  is required  to  identify the relative importance of interv entions to:
•    increase basic nutritional levels
•    increase cash crops
•   change cultural practices
•    improve water supply to urban and rural areas
■ improve sanitation
Much  data  is  available  from  the  Report  on  lite  National  Rural  Nutrition  Survey.  Core  Model  1992 
produced  under the National Nutritional Surveillance System.  The purpose of the NRNS is to provide information  that will assist in  incorporating  nutritional considerations into  a wide variety of long-term planning  and  policy  decisions The report surv  eys 609  Farmer Associations in  22 regions,  of which  39  were in  llulabor.  34  in  Kefa and  25  in  Welcga,  with  household  coverage in each  region  of 948.  851  and  624  respectively.  Some indicative findings are shown  below. Overall,  Ethiopian  infants (up  to  the age of 2) show levels of stunting  which  arc amongst the highest in the w orld.
f
Rank Position of Regions in Basin for Incidence of Stunting and Wasting
(I = Region with least problem, 22= Region with worst problem)
Factor
llulabor
Kefa
Welcga
Average
National average
Stunting
17
11
8
12
64%
Wasting
16
12
19
16
8%
Stunting = low height for age Wasting — low’ w eight for height
r
Results quoted are for ages 6 to 59 months.
These results indicate that the basin  area has generally  a bigger problem with  child  nutrition  than the rest of the country.  Other correlations made in  the survey  indicate that areas with  lower altitude have more severe problems with stunting and wasting.
The Ministry  of Health,  supported  by  WHO,  has undertaken  a series of studies under the name
Regional  Health  Profiles  and  5  years  Health  Plan  The  one  for  Gambella  is  dated  October  1992 
and  others  have  been  collected  by  the  Consultant  for  Asosa,  llulabor,  Kefa  and  Wolega.  These will be reviewed in the next phase.
4.6        Education
The  Reconnaissance  Report  notes  that  literacy  Tates  have  not  yet  been  worked  out  on  the  basis of the newly (1987) reconstructed regions. It notes that in 1980 there were in the uppeT basin'
TAMS-ULG           u AKO-AKOBO wver basin INTEGRATED DEVELOPMENT master plan4           SOCIO-ECONOMICS
(41) kindergartens 701 (360) elementary schools
186,265
students
84 (112) junior secondary' schools
18.767
students
16(27) senior secondary schools
13.921
students
Total
218,953 (173,922)
students
Figures in  parenthesis arc those quoted  by  the AG Report There are major differences for which there Eire no  immediate explanations.  Attendance is poor and  drop  out rates are high,  especially in the coffee picking season.
No  reference  to  educational  facilities  are  made  in  the  Russian  Report*  but  the  Gambela  Needs Assessment notes that in Gambela Lhere are
74 primary (elementary) schools of which 31 functioning
I senior secondary school
I non-functional teacher training institute.
The  report  notes  that  school  attendance  is  low,  but  even  lower  among  females,  whose  drop-out rate is high.
4.7         Rural   Infrastructure
47 1 Roads
The  Russian  report  describes  the  history  of  development  of  road  construction  in  the  plains,  and lists the following gravel surfaced roads:
Routes connecting to the plain
Surface
km
(Gambela - Assaba)
Asphalt
(1681)
Gambela - Addis Ababa
Gravel
804
Gambela - Bure
Asphalt?
55
Routes within the plain
Gambela - hang
Gravel
45
Itang- Lara-Jikawo
Gravel
65
(under construction in 1990)
Gambela - Abobo -Lake Tata -Agengo
Gravel
115
Abobo - Dumbong
Gravel
22
Abobo State Farm - Ukuna - Kijang
Gravel
12
Lake Tata - Pashala
Gravel
29
Abobo State Farm - Perpang
Gravel
15
Abobo - Gog
Earth
??
The existence of a few gravel surfaced inter-farm roads was noted. The roads are described as liable to waterlogging, washout and rain damage. The earth road from Abobo to Gog is
TAMS-ULG       BARD-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTFR PT AN
4-104            SOCIO-ECONOMICS
impassable   in   the   rains.   ITic   report   notes   the   difficulties   experienced   by   people   living   in   the western part of the plain due to the absence of roads.
In  Gambela,  the Needs /Xssessmenl report notes that the level of communication  is poor.  There are no  asphalt roads in  the area,  and  all weather roads arc limited  to  Gambela and  Abobo wcredas.  There are only  38  vehicles in  the whole region,  of which  only  4  are privately  owned. There are no  taxis or buses,  or even  animal drawn  vehicles.  There is also  a lack  of local radio  and TV transmission
Proposed roads under the Russian Master Plan was based on:
Gravel surface
Vnsurfaccd
Irrigated lands
7 km/'ODO ha
30 km/’000 ha
Rain fed lands
6W‘000 ha
25 km/” 000 ha
which led to an estimate of total road lengths (km) of:
Length of roads, krn
Gravel surfaced
Unsurfaccd
Development
1990-
1995-
2000-
1990-
1995-
2000-
Alternative
1995
2000
2010
1995
2000
2010
Minimum
Irrigation
97
134
228
406
562
979
Minimum Rainfed
97
134
500
406
562
2087
Maximum            97                    134                 354                 406                 562                  1507
The  inter-farm  and  main  roads  would  he  8m  wide  and  the  farm  roads  6m  wide.  Other  proposed roads were:
Route
Surface
Km
Abobo - Gog -Gilo 2.1
Gravel
44
Gilo 2.1 Dam - Pashala - Ubcla
Gravel
42
The report reviews transport requirements to  export produce and  provide for trips by  the new inhabitants at the rate of 5  trips per year per inhabitant plus at 100  kg/h  plus 1000  kg/h  of personal goods Grain  is laken  al 50% of yield  and  other produce at 100%,  while sugar exports were calculated directly.
TAMS-ULG UAROAKOBO RIVER BASIS INTEGRATED DEVEI dement MASTER PLAN
4-J4           SOCIO-ECONOMICS
From these calculations,  plus some allowances,  a design  figure of 715.000  vyr.  was estimated for the route Gambela - Agenga.  over a period  of 200  days,  to  arrive al 2300  trips a day  of 5t vehicles, for which an asphalt surface would be needed.
A requirement for 500 km of 8m wide all weather roads built on 1.5 m of fill on the plain was also  calculated,  with  a gravel surface 30  cm thick.  In  addition  23  bridges between  50  and  200m long  and  31  storm water culverts would  be needed.  No  further calculations were done on  costs, but it was appreciated  that this would  be very  expensive as road-building  conditions are difficult and borrow pits scarce.
National statistics were reviewed  to  verify  what data is available,  and  a summary  is given  in Table 4-8  of the trucking  capacity  and  activity  in  the public and  private sectors.  These figures demonstrate the huge dominance (75% oi tonne-kilo metres) of the public sector in transportation, for the 1990/91  figures available This situation  may  change dramatically  if Government policy towards encouraging the private sector is implemented.
4.7.2     Airports
The airports in  the region  are identified  in  both  reports,  but no  figures arc given  for movements or numbers of passengers. It is believed that air freight is a minor activity through these airpods. If irrigation projects are proposed for the production of high value crops, there may be a case for up-rating  handling  facilities to  allow' export of perishables from the region.  Such  considerations need not be of importance at this stage of planning.
4.7.3     Water T ransport
Gambela  w as
r
a  pon  from  1907  to  1964,  serving  Khartoum,  a  voyage  of  7  days  during  the
navigation  period  of  2-3  months.  Nowadays,  the  only  navigation  is  the  142  km,  3  day  trip  to Jikawo. which transports some 50-60 tons annually.
The Russian  Report reviewed  the possibilities for developing  commercial transport and concluded  against it.  However,  the potential for navigation  based  on  a minimum of I 5m water depth  for tourist traffic was recognised,  especially  if upstream regulation  of the Baro  was implemented.  *fhc Gilo  and  Alwera rivers are not considered  to  be of interest commercially,  but might well also  be of interest for tourism if regulated  for hydropower.  This potential is not reviewed in the Russian report.
4.7.4     Electrification
Power  supplies  come  from  small  diesel  power  plants  of  15-30  kW.  There  are  no  lines  over  15 
kV. In Gambela itself, there used to be one 194 kW, two 250 kW and one 710 kW sets connected by a two 315 kVA step-up trans formers to a 15 kV overhead line, A 5 MW hydro-electric power station  was completed  al Sor in  1990,  and  connected  to  Gambela via a 66  kV transmission  line. Studies are likely  to  be completed  by  EELPA before end  1995  on  upgrading  this station  to  10 MW in  the future.  Diesel generation  in  Gambela town  was discontinued  when  the power line from Sor was completed.
T AMS“LXG daro-akouo river dasin integrated development master plan
4-124            SOCIO-ECONOMICS
4.7.5      Water Supply
The situation  in  urban  areas is described  elsewhere- The Needs Assessment report notes that Gambela has no  organised  water supplies.  Water in  almost all rural areas is carried  long distances from polluted sources (rivers and poorly managed wells) by women and children.
4.7.6      Postal Services
The CSA publishes a list of Post Offices and  Post Boxes by  region,  which  is reproduced  in  Tabic 10  h  is apparent that Ilulabor and  Welega are very  poorly  served  in  comparison  with  other regions of the country', but for some reason Kcfa is well served.
3.8       Commerce
Very  little  commerce  is  reported  in  the  region,  although  the  A-G  Energy  Survey  listed  a  fair number of very small scale establishments. These consisted of:
1           Tea Plantation at Gumero (860 ha)
2           Coffee Plantations at (6537 ha) and Tepi (4.44] ha)
8          Saw Mills
15        Coffee Pulpers
37        Grain Mills
61         Bakeri es (approx.)
539 Food Preparation Services (Hotels, cafes etc J
The list was compiled  for the purposes of estimating  energy  requirements,  and  is not therefore comprehensive Traders,  for example,  would  not appear in  such  a list.  However,  the list is remarkably  sparse for a population  of almost 3  million  people,  and  indicates the very  low level of economic activity in the region,
TAMS-llLG BARO-AKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
4-11SECTIONS
ENVIRONMENT5.          ENVIRONMENT
5.1        Background and Objectives
A major study  was undertaken  by  the Russians over a period  of some eight years,  concentrating on  the Lower Basin.  This resulted  in  a Master Plan  (Selkozpromexport.  1990) projected  over a development period up to 2010,
[n 1990 a short study was undertaken on the Upper Basin (EVDSA. 1990), but was limited in its environmental content.  The most recent study,  that of ARDCO-GEOSERV (1995),  was also restricted largely to the Upper Basin (above the 500 tn contour).
The objectives behind  the inception  Report on  the environment are to  review these previous studies of the basin,  in  particular those of Sclkhozpromexport (1990),  hereafter referred  to  as the Russian  report,  and  that of ARDCO-GEOSERVE  (1995).  The intention  in  these reviews is primarily  to  identify  gaps in  the database which  will constrain  the preparation  of an  integrated Master Plan; at this stage of the Inception Report, no recommendations are proposed.
The Russian  report,  although  presenting  a detailed  account of the agroecological status of the lower basin,  assumed  a political structure and  development process that has subsequently  been superseded.  Consequently,  many  proposals have since become impractical under new political structures.  Further,  ail reports are now out of dale,  particularly  in  those sectors where in migration of peoples has impacted on the resource base.
5.2        Structure and Approach of Environmental Section
There are at present problems caused  by  the absence of clear national policies in  many  critical sectors,  including  development of water resources,  environmental impact assessments,  wildlife and  tourism policies,  and  the National Conservation  Strategy.  Although  government has recognised  the need  for such  policies,  many  of the key  documents are either still in  draft form, or remain to be developed.
.As  to  the  absence  of  policy  documents  and  clear  definition  of  the  nature  of  the  Master  Plan,  it was  simply  assumed  that,  following  a  major  shift  in  government  focus  towards  devolution  of authority,   "gaps”   in   the   data   would   be   identified   largely   from  a  perspective  of  "bonom-up" planning, and that projects in Phase II would involve some degree of 'stakeholder participation".
To  supplement this approach,  general environmental guidelines are available through  the Science and  Technology  Priority  sectors,  as spelt out in  the National Science and  Technology  Policy  of Dec 1995 (refer Annex 1).
By definition, any environmental review must operate "cross-secto rally". Thus, to some extent, the environmental Inception Report will "mirror" other components of the Inception Report as
a whole, and reference to other sections of the Inception Report will be made, and some repetition will be inevitable. The Inception Report submission on Environment should be rc.-uj
in conjunction with those submissions on Wildlife and on Tourism (Sections 1.6 and 1,7
TAMS-ULG       HAIO-AKOBO RIVER 0ASFN INTEGRATED DEVELOPMENT MASTER, PLAN
5-15.           environment
respectively).
This section  is structured  to  give an  overview of the major environmental features of the project area (Section  5.3).  and  then  continues by  summarising  the main  forms of land  use in  as far as they  impact on  the environment (Section  5.4).  The environmental concerns arising  from these varied  forms of land  use are then  listed  (Section  5.5); from this summary  arc extracted  the key environmental issues present or likely to arise in the project area (Section 5 6).
For these principal concerns,  some mitigatory  actions will need  to  he considered  in  the final Master Plan.  To  meet these mitigatory  needs,  the report concludes by  highlighting  those areas where additional data will be required (Section 5.7).
5 J Environmental Perspectives of the Project Area
The abiotic features,  which  clearly  are major determinants of the biological components of the environment are addressed  elsewhere in  Ihe Inception  Report; it is the biotic components of the environment which are reviewed here.
53-1 Natural Vegetation
The characteristics of the natural vegetation are determined largely by the above abiotic factors. 
On the basis of elevation. Ethiopia has traditionally been divided into five agro-climatic bells: the Bcrha (relatively dry and hot, below 500 m); Kolla (dry to moist, between 500 and 1500 m); Weyna Dega (moist to wet, between 2300 and 3200 m); and Wurch (moist to wet, cold belt
above 3200 m). In this contest, "dry” is seen as receiving rainfall below 900 mm, "moist" as
between 900 and 1400 mm, and "wet" in excess of 1400 mm annually.
in  refinement of the above,  ARDCO-GEOSERV distinguished  11  agro-ecological zones for the upper Basin,  based  primarily  on  temperature and  altitude,  whilst FAO have recognised  eleven zones for the project area overall ic for both  basins.  This is expanded  on  in  the Inception  Report submission by the TAMS-ULG Land Use Planner.
Upper Basin
Vegetation  surveys  of  the  Upper  Basin  appear  to  have  been  of  two  types:  those  that include mention of dominant species, and those which refer simply to structural features.
In   the   former   category.   Chaffey   (1980)   identified   economic   tree   species,   and   EVDSA (1990) produced a very generalised vegetation map of "climax vegetation".
Within  the second  category  ,  previous mapping  by  FAO (1:1  000  000) gave 16  vegetation cover types for the Upper Basin,  as well as distinguishing  six  classes of land  use This classification  is summarised  by  ARDCO-GEOSERV (Vol 111  A).  Ihus,  the FAO mapping kept land  use and  broad  vegetation  communities as qilitc separate categories For example,  ’perenma1 crop  cultivation" was FA0 land-use category,  with  "lowland bamboo bushland an example of a vegetation cover type
TAMS*ULG baro-akobo fuver BAsrN integrated DEVELOPMENT master plan
5-2ENVIRONMENT
In  their agricultural appraisals,  ARDCO-GEOSERV then  combined  these two  attributes into  a single unit,  which  they  termed  a *  landscape -ecology  unit"- For example, '^Intensively  Cultivated  Open  Woodland” was a unit combining  the two  attributes.  In  this manner, 84 units were identified, which were then grouped within agro-ecologicaj zones.
Almost half of Ethiopia's surviving  primary  forest is said  to  be located  in  the BaroAkobo basin.  The dense forests are found  in  the Weyna Derga,  the Dega and  the upper K*olla zones.  A brief overview is given  by  EVDSA (1990),  and  further reference in  the Inception  Report is found  in  Section  1-5  of this report.  Of note is chat EVDSA commented on the almost total absence of any silvicultural management of these forests.
£owe r florin
The Russians employed  a complex  system of vegetation  classification.  Essentially,  they produced  a vegetation  map  (which  was not made available to  the consultants); according to  the text this "geobotanical map” was based  on  hydrological,  hydrogeological, geological and  soil factors.  This approach  allowed  for the designation  of three main zones:
*  seasonally  flooded  grasslands  and  savanna  of  the  low  plain  on  the  western  zone  of  the project area;
* savanna forests of the inclined plain (425-550 m);
* savanna and forests in the elevated (> 550 m) eastern zone of the project area.
These in  turn  were subdivided  into  six  vegetation  types,  as shown  in  Table 5.1.  For each type,  dominant grass,  shrub  and  tree types were recorded.  The Russian  report also covered  the hydrophyte community  in  some detail.  In  general,  the Russian  section  on vegetation,  although  painstakingly  detailed  in  parts,  is a rather ’rambling" account, requiring  more consolidation.  As no  internationally  accepted  phylosociological classification  system was adopted,  there are conceptual ambiguities and  overlap.  This problem is encountered  in  the forest section  as well.  In  sum,  there are three problems with the Russian approach to vegetation classification and mapping:
*  lack  of  compatibility  with  other  systems  adopted  in  Ethiopia  (and  indeed  elsewhere  in Africa);
*    employment   of   an   agrocentric   perspective,   in   that   classification   and   mapping   were sometimes approached in terms of the potential of the land for clearance for agriculture;
♦ overlap of categories employed will complicate any efforts at repeatability. 77ie Bdro-zikoio Aorin ar a wWe
The various approaches to vegetation classification are summarised in Table 5. i
TAMS-ULG       BARO-AKCi&O RIVER BaSEN integrated development mas e er plan
535.
ENVIRONMENT
Table
5.1:
Summan   of   vegetation   classi
fication 
ad
opted   by   vario
us
authorities
in 
the
project
area. 
(Note
that  the  ARDCO-GEOSERV
listing 
is
incomplete,  but
the
diversity 
of
terms
listed 
serves nevertheless to illustrate the problem of incompatibility).
FAO SYSTEM
RUSSIAN REPORT
ARDCO-GEOSERV
•  Cultivated land
•  Perennial crop cultivation
•  Perennial crop cultivation and disturbed high forest
•  Disturbed highforest
•  Dense Mixed high forest
• Dense Woodland
• Open woodland
•  Dense woodland & Lowland and Bambo
•   Lowland bamboo
Bushland
• Wooded Grassland
• Bushed & Shrubbed
Grassland
* riverine grassland under long-term flooding
* savanna under recurrent short-term flooding 
•savanna with insignificant
forest
•deciduous forests
•piedmont savanna forests mountain forests
• Intensively cultivated Open Woodland
• Moderately cultivated open woodland
• Sparsely cultivated Disturbed forest Land
• Scattered Cultivated Disturbed Forest/Woodland
• Moderately cultivated Disturbed forest/Woodland
• Scattered Cultivated Dense woodland
•  Dense Wood and Undisturbed Forest Land
•  Bamboo/foresl &
Woodland
• densc/Undisturbed Broadleaf Forest Land
• Intensively 
cultivated/very Disturbed Forest Land
--
What are the implications for these divergent systems of vegetation classification?
Firstly,  there is the question  of repeatability.  Can  the same units be recognised  in  the field,  and  mapped  accordingly  ie are the definitions and  distinctions rigorous enough  to allow for repeatability? In  the ARDCO-GEOSERV study,  it is unlikely  that the same system of classification  can  be adopted  to  locate the same vegetation  types,  such  is the lack of firm definition.
The implication  here is that there will be problems using  the ARDCO-GEOSERV data xs a benchmark  against which  to  measure rates of loss of specified  "landscape-ecology units", as it is unlikely that the same vegetation types could all be "relocated" in the field,
TAMS-ULG haro-akobo river basis integrated development master plan
5-»5.          ENVIRONMENT
Secondly,  there is the question  of compatibility.  For example,  is the FAO category  in Table 5.1  "perennial crop  cultivation  and  disturbed  high  forest' consistent with  the ARDCO-GEOSERV category of "scattered cultivated disturbed forest "?
53.2 Floodplain Ecology
Basic to  both  vegetation  and  wildlife on  the Gambella plains is the flood  plain  ecosystem.  The key  feature of this ecosystem is obviously  the annual flooding  regime,  and  this subsumes some important ecological processes:
- change in the nature, extent and productivity of associated pastures;
-   provision  of  spawning  grounds  for  fish,  as  well  as  breeding  zones  for  a  variety  ol aquatic invertebrates and herpetofauna;
-   deposition  of  a  layer  of  nutricnt-rich  silt  which  acts  as  a  fertilizer  to  enrich  pastures when water levels decline;
-   provision  of  vast  areas  for  the  overwintering  of  numerous  Palaearctic  migratory  bird species, especially waders.
Any  disturbance  of  this  system  will  impact  on  fish  productivity,  on  bird  habitats,  and  on  the productivity of pastures,
The ecological data from the Russian  report for the floodplain  is restricted  largely  to  plant species lists,  together with  descriptions of phenologicai processes for key  species and  communities.  For example,  they  recorded  clear sequences of flowering  for "cphemerals’’ (presumably  this refers to annual species), for legumes, and for grasses.
Whilst the Russian  report gave due recognition  to  the critical conservation  needs of rainforests and  catchments,  they  perhaps gave inadequate attention  to  the role of flooding  in  nutrient recharge,  in  the breeding  of fish,  amphibia,  and  a of a range of invertebrates on  which  the fish resource depends,  as well as the role of wetlands in  bird  migration  and  biodiversity.  The report shows little consideration  for ecosystem processes like nutrient cycling  (for example,  the role of termites on  the floodplain  was seen  solely  in  terms of the extent to  which  they  were an "obstruction" to  agriculture,  with  no  apparent appreciation  of their role in  nutrient cycling; indeed, termitaria were seen as being composed of "absolutely nonfertile sterile soil").
In  addition,  the Russian  report has a perspective of "damage" due to  flooding,  measured  in  terms of the loss of crops from inundation.  As such,  they  envisaged  two  approaches to  "flood  control”: the construction  of dykes in  susceptible1'  areas,  and  the construction  of large reservoirs in  the upper teaches of the river.  In  sum,  the annual flooding  event was viewed  largely  from an agricultural perspective,  and  ignoring  the ecological benefits and  implications of this major annual event.  They  did  however,  in  the section  on  seasonally  flooded  pasrures,  acknowledge the role of flooding in nutrient deposition floodplain "meadows '.
TAMS-ULG 0ARO-AKOBD RIVER. BASIN INTEGRATED DEVELOPMENT MASTER PLAN
S-J5.           ENVIRONMENT
These  concerns  arc  should  not  necessarily  be  seen  as  criticisms,  in  that  the  Russian  terms  ot reference  catered  largely  Tor  a  very  specific  applied  aspect  of  ecology:  agriculture.  Nevertheless, to  the  extent  that  one  wishes  to  understand  floodplain  ccodynamics  of  the  Baro-Akobo  interfluve, the  report  is  of  limited  value.  Indeed,  there  are  some  naive  ecological  interpretations  eg  that  fires can  be  partly  beneficial  in  that  "a  lot  of  harmful  pests  and  rodents  perish  partly  in  this  fire..'' Such  notions  are  inconsistent  with  modem  understanding  of  the  impact  of  fires,  as  well  as  with current thinking on the maintenance of biodiversity as an environmental objective
533 Biodiversity and Endemism
The above provides a very  brief overview of some aspects of the ecology  of the project area Recent developments in  major projects worldwide have led  to  an  expectation  that a component on "biodiversity" be included, and a separate section thus follows here.
"Loss of biodiversity" has numerous implications,  and  a useful starting  point is the link  between biodiversity  and  the potential source for wild  crop  species,  Ethiopia is particularly  well-endowed with  traditional varieties of crops,  as well as (heir wild  relatives.  Tcff (Eragrostis tef) and  ensete (Ensctc ventricosunfl are both  examples of fully  domcslicated  endemic crops.  However,  there is surprisingly  little documentation  on  the diversity  of the Ethiopian  flora,  although  the Ethiopian Flora Project is an attempt to meet this shortfall.
Knowledge of the medicinal uses of wild  plants is increasing  in  many  regions of Africa; this "ethnobotanical" approach  indicates the considerable dependence placed  by  rural peoples on natural medicinal products.  However,  there arc few data for Ethiopia on  the species used,  nor on the volume of the trade.
Endemism (a measure of the extent to  which  a taxon  has evolved  only  in  that area under consideration,  and  thus a reflection  of "uniqueness") is a further concept normally  addressed under environmental concerns.  In  the project area,  the level of floral endemism is relatively  low. This is largely  because floristically  the project area merges into  the Sudanian  floristic region, which  itself extends westwards to  the Atlantic (White,  1979).  For example,  some of the dominant montane fores: trees of the upper basin  (J um^rus proccra,  Podocarpus gracilior.  Olea africana. and Cd Lis afrkana) are found in other montane regions of Africa.
In  the lower basin,  this relatively  lowcndemicity  is again  demonstrated,  as Sudanian  elements are common  eg  the savanna woodland  genera of Termi nalia and  Combrctum are widespread  in sub-Saharan Africa.
5.4       Current patterns of land use
From  an  environmental  perspective,  an  understanding  is  required  of  the  extent  to  which  current land use patterns have modified the natural resource base. In particular, one needs to understand:
- nature and rate of change of specific natural resources;
-  impact of such land use practices on these resources, as well as on '’ecosystem
TAMS-ULG BARO AKOBO RIVER flASfN IMTCRated development master plan5- ENVIRONMENT
processes" (a term which includes nutrient cycling, hydrological function etc);
- the ability of various ecosystems to show resilience in the long term to such impacts, and the states to which ecosystems will tend in the absence of resilience;
■ future trends in the nature and intensity of land use.
Land Use in the project area is discussed in both the Agronomist's report (which deals largely with cropping), in that of the livestock specialist, in that of the forestry specialist, as well as in that of the Land  Use Planner.  Salient features of land  use as they  relate to  environmental processes and impacts are summarised below.
5.4.1 Crop Production
With regard to cropping, almost all the permanently cultivated land is presently found in the Upper Busin, as the plmnspeoplc have little experience of intensive farming, apart from exposure to State Farming exercises; most farmers of the Upper Basin are engaged in mixed fanning with crop rotation, chiefly involving maize (Zea mays), left (Era frost is tef) and sorghum (Sorehum spp.). The predominance of a particular cereal (or group of cereals) for any specified area is in the main a function of agroccology (for example, sorghum is a drought-resistant genus, and will be expected to occur in more arid zones); however, a subjective impression from the consultant's field trip is that cultural factors also operate, in that different regions, apparently under similar soil and climatic regimes, may be growing a quite different suite of crops.
in the Lower Basin, subsistence cropping may take place either as floodwaters recede, or is concentrated along the banks of rivers. Most farmers appear to plant maize in the wet season, and sorghum in the dry; tillage is unusual, and sites are usually cleared of weeds by burning. In addition, some groups practice shifting cultivation. State Farms have been established tor cotton, coffee and palm oil production.
5.4-2 Livestock
For Livestock,  the recent figures appear most unreliable for both  Upper and  Lower basins However, the relative distribution of livestock for the two basins may be of more value than disputed absolute figures: the majority of livestock arc in (he upper basin, and only 15 % in the Gambcla Plain (report of the Livestock Specialist).
Constant reference is made in most reports reviewed to the role of tsetse fly (Glossina spp .) as a constraint in livestock distribution and productivity, even for the Upper Basin (eg ARDCO- GEOSERV). Many of these statements appear to be carried over from report to report, and the impact of trypanosomiasis in fact lacks quantitative data, and is often speculative. The definitive survey (Langridge, 1976) of tsetse distribution in Ethiopia is now about 20 years out of date, yet reference to  this source to  define distribution  still continues.  As the fly  responds rapidly  to changes in habitat, which have been pronounced over this time period, there is no way these data can be used to infer present distribution of the fly.
TAMS-ULG BARO-AKOBO RIVER D ASIN INTEGRATED DEVELOPMENT MASTER PLAN
5-75.           ENVIRONMENT
Livestock  performance is also  clearly  a function  of the grazing  resource.  For the Lower Basin, the Russian  report classified  "pastures" into  three main  types: dry  season  grazing  (seasonally flooded  pastures),  wet season  grazing  of the higher plain,  and  pastures of the foothills and mountain  slopes.  These were presented  according  to  their estimated  yields and  periods when grazing  w ould  be desirable in  terms of grass palaiability.  However,  this appears to  have been arrived  st from a perspective of an  ideal situation,  where grazing  systems would  be imposed under some kind of (unspecified) overall authority.
What does not however emerge from the Russian  study  is the actual relative use made by livestock  of these potential grazing  lands.  This is particularly  relevant under pastoral nomadic systems,  in  which  seasonal migration  of livestock  takes place in  response to  changing  status of pastures.  Further comment on  this is made under the Section  on  Wildlife data needs (Section 1.6.83).
For the Upper Basin,  livestock  rely  on  a diversity  of food  resources,  depending  on  season,  and on  the intensity  of associated  cropping.  The option  to  use crop  residues to  supplement natural pastures appears to be restricted largely to the Upper Basin.
5.4.3     Forestry
One aspect of forestry  as a form of "land  use" that is not covered  in  that submission  is the use of indigenous plants as a primary  source of drugs for traditional health  practices.  Although dependence by  local peoples on  forest resources has not been  quantified  to  any  large extent, Abebe & Hagos (1991) have outlined  the traditional healthcare system in  Ethiopia,  and  list some 50 species. Over 95% of the traditional medical "recipes" arc of plant origin.
There is a paucity  of data on  biodiversity  of both  forest fauna and  flora,  as well as on  the use of wild  forest plant species for medicinal purposes.  Certainly,  the apiculture report of the Inception Report states quite clearly  that the "exploitation  of bees may  be the major,  sustainable,  non-wood income source for people living in or near to tropical forest and woodland".
5.4.4     Fisheries
Little is known  of this form of landuse in  the Upper Basin.  For the Lower Basin,  fishing  is mainly  at subsistence or artisanal levels,  reduced  al times of flooding.  Some of the catch  rates for co-operative fishing  concerns in  the Lower Basin  are reviewed  elsewhere in  this report,  but the paucity'  of data prohibited  a clear economic analysis.  Nevertheless,  but there appears potential to increase production fourfold.
5.4.5      Relocation of people
I  he  relocation  of  people  in  new  areas  can  take  different  forms:  for  the  project  areas  we  are concerned    with    resettlement    programmes,    refugee    programmes,    and    nomadic    movements    in response  to  pastoral  requirements.  With  regard  to  the  first  category,  the  Derg  regime  relocated 
□ver  600  000  people  from  the  famine  prone  parts  of  the  northern  Highlands  of  Ethiopia  to  the better-watered  areas  of  the  south  west  (Butcher  &  Wood,  1995),  with  a  resultant  sudden  increase
TAMS'ULG BAKO-AKOBO HI vest basin integrated development MASTER plan5.          ENVIRONMENT
in population pressures.
As to  the number of people living  in  the "Park",  only  crude estimates are available (it is suspected  that UNHCR,  will 'play  down" the estimates,  whilst local authorities,  wishing  to  see some action  taken,  will tend  to  inflate figures).  Even  if extremely  broad  confidence limits are imposed  on  the estimates,  the results are staggering.  Mission  estimates are shown  in  the Table below.
Table  5.2:  Mission  estimates  for  populations  within  (he  boundaries  of  the  Gambella  "National Park".
YEAR
1992
1995
Slate Farms
70 000
40 000
Resettlement
1 500
3 500
Refugees
No Data
43 600
TOTAL
87 100
Local government estimates give 70  000  people in  the park,  compared  to  nearly  90  COO in  Table 5.2.  Both  these estimates exclude permanent Gambella plain  residents,  pastoralists and  nomadics. The temporary  placement of refugees on  specific sites within  the project area can  justifiably  be regarded as a form of land use, in that it forecloses options for other forms of land use, especially in  the long  term.  UNHCR estimates for October 1995  gave figures of 33273  refugees for the Fugnido camp, of 10267 for Dimma I. and 15614 for the Bonga camp.
5-5 Environmental concerns
What art the broad  implications for these current forms of land  use for natural ecosystems,  and what environmental concerns emerge? This section  lists both  those environmental concerns currently  evident,  as well as those which  may  be of concern  in  the future.  As it is not expected that all these concerns listed  here should  be addressed  in  subsequent phases of the project,  those considered of major concern are then summarised in Section 5.7.
5,5.1 Crop Production and Soils
To  what extent are present fanning  systems in  the project area compatible with  accepted principles of soil conservation? The general practice is to  construct furrows (using  ox-drawn ploughs),  and  then  to  seed  manually.  This is followed  by  weeding,  again  manually.  Replacement of nutrients after harvesting  is minimal,  outside of that low input which  will lake place due to 
TAMS-ULG baroakobo rjver basin integrated development master plan
S-95,           ENVIRONMENT
breakdown  of soil minerals and  nutrients contained  in  rainwaler and  runoff- Fertilizers are seldom used,  apparently  in  response to  both  price and  lack  of availability  ♦  Under such  a system, aggravated  by  very  limited  fallow periods,  non-sustainabilily  of production  levels will be the case, particularly where erosion plays a role
This falling  productivity  is of particular concern  environmentally,  as the effect will be to  convert yet more uncultivated  land  for cropping  purposes,  land  which  is often  marginally  suited  to cropping.  For example,  vertisols (which  predominate in  certain  parts of the bottomlands of the Lower Basin),  are notoriously  prone to  waterlogging,  and  must be considered  marginal for cropping.  Consequently,  when  these lands arc used  for cropping,  farmers tend  to  plough  these soils before the onset of the rains (as they  subsequently  become unmanageable),  and  the soil lies exposed  to  the first rains,  which  are often  violent enough  to  cause erosion  of the exposed surfaces.  The fanning  strategy  is to  plant towards the end  of the rains,  to  maximise the water retained by these heavy clays.
This conversion  of land  for cropping  takes place al the expense of grazing  land,  as well as at the cost of deforestation  Where in-migration  and  population  growth  arc marked  this process will be accelerated.  The Russian  report acknowledged  these issues for the Lower Basin,  and  in  fact advocated  intensification  of existing  arable lands as opposed  to  the opening  up  of new arable areas.
A relatively  recent development in  the Upper Basin  has been  the drainage of wetland  areas in  an attempt to  intensify  land  use The extensive wetlands of the Upper Basin  fulfil important hydrological functions (flood  amelioration,  water quality  control and  groundwater recharge),  in addition  to  their contribution  to  biodiversity.  This form of land  use (drainage and  multi-cropping)
has  become  prevalent  in  northern  lllubabor  Butcher  &  Wood  (1995)  have  identified  about  240 
km2 of on-going  wetland  drainage in  highland  lllubabor to  permit multi-cropping  in  valley bottoms.  Within  the project area,  a major zone of concern  extends from Bure to  the east,  as far as Kumbabe on  the boundary  of the project area.  In  this zone,  these drainage practices can  be observ ed on cither side of the road.
These trends in  wetland  landuse have been  assisted  by  an  NGO operational in  the region (Menschen  fur Mcnschen).  If wetland  drainage continues to  be advocated,  it could  lead  to widespread  hydrological impacts,  which  in  turn  will affect the remaining  patches of tropical montane rain  forest,  as well as impairing  the capacity  of the catchment to  meet the expectations for impoundments downstream.
5.5.2 Livestock
As more land  goes under cropping,  the general pattern  is for livestock  to  be forced  onto  marginal grazing  lands,  including  steep  slopes.  Livestock  tend  to  initiate gulley  erosion,  particularly  where they  are confined  to  paths leading  directly  to  waler sources,  or along  the edges of roads where forest margins on steep slopes restrict them to repeated passage over the same area.
TAMS-ULG baro-akobo juver basin integrated development master plan
5,105. ENVIRONMENT
5.5J Forestry
The estimated rate of deforestation in Ethiopia is over 100 000 ha annually, a rale which will remove al! remaining forests in under 30 years It is the removal of indigenous forest which is critical; in the Upper Basin most of the remaining forest is now restricted to the "steps" between a series of plateaux; these steps have slopes quite unsuited to agriculture, and any attempt at cultivation here runs the risk of soil erosion, leading not only to rapid decline in crop production, hut to siltation of rivers downstream. It is in fact this relative restriction of forested areas to steeper slopes which has prevented erosion in many areas. Nevertheless, the potential for erosion in the upper catchment is enormous, given the removal of the forest cover.
Although  much  of the "informal" forest clearing  is achieved  often through the use of fire (EVDSA, 1990). the establishment of both Bebeki and Tepi Stale Coffee plantations (viewed as
a more planned and thus "formal" deforestation programme) was undertaken at the expense of vast tracts of indigenous forest, apparently without due consultation with forestry authorities (EVDSA, 1990).
Although most reports have focused on the importance of the forests for containing erosion and for protection of catchment areas, little attention has been given to the loss of biodiversity resulting from deforestation. Given the reservoir of wild plants which is found in natural forests and woodlands, some of which have the potential for "domestication" as crops (refer Section
5.3.5  above), deforestation and loss of indigenous species is cause for considerable concern (Edwards, 1991).
5.5.4 Resettlement
With regard to refugees, the environmental impacts arc of grave concern. Firstly, there appears no clear policy for location of refugee camps, and siting apparently takes place on an ad hoc basis.
UNHCR operates in any country by invitation only, and allocation of areas is a government responsibility.  If time is available,  a survey  of a proposed  site is undertaken,  but often  an emergency situation is at hand, and this step is pre-empted. Environmental problems arise when an emergency holding camp (usually established initially with the intention that it should serve as a transit post only) takes on a measure of permanence. Current "survey" procedures used by UNHCR address issues like camp boundaries, layout, access roads, scheme for water delivery, agreements on where wood can be cut etc; UNHCR (1982) specify criteria for site selection, but these do not take into account potential environmental impact,
It is clear from the Gambella that inadequate attention  is given  to  longterm environmental impacts of refugee camps. Not only are refugees removing riparian woodland along the Baro River (Bonga camp is located where the Baro river emerges from the gorges of the escarpment onto the plains), but they arc using uncontrolled fires to clear land for cropping. In addition, there is evidence that a charcoal industry  has been  established  by  refugees,  in  which  charcoal is conveyed out of the Lower Basin using the very transport used to carry' refugee provisions into the Basin. During the consultant's field visit that is the Geld visit, "early burning" (that which lakes place at the beginning, as opposed to the end, of the dry season) was much in evidence.
TAMS-ULG      BARO-AKO BO RIVER BASIN INTEGRATED DEVELOPMENT MASTER PLAN
5-15.           environment
This practice leads to  permanent changes in  the herb  layer (and  hence to  soil cover),  as well as to  slirub  encroachmenL  There are however efforts to  ameliorate the situation,  in  that proposals have been  advanced  for the reafforestation,  centred  around  the ffonga and  Fugnido  refugee camps.
A further area of concern  related  to  the Bonga camp  is that the upstream area is eminently  suited to  ecotcurism ventures (refer Section  1.7},  and  any  continued  adverse environmental impact will deny this as an option for land use.
5.5.5   Impoundments currently under construction
Of those water resource development plans proposed  in  earlier reports,  only  (he dam on  the Alwero  river is being  built.  No  consideration  has been  given  to  the environmental impacts of this dam.  Al the very  least,  there will be impacts on  spawning  and  general migration  of fish; further, change in  fish  community  composition  can  be anticipated  in  response to  changes in  water flow rales,  depth  and  eutrophication  arising  from nutrient retention  Further discussion  on  these concerns follows under Section 5.5.7, which deals infer alia with proposed impoundments.
5.5.6  Projects envisaged under previous Master Plan recommendations
In  this section,  consideration  is given  to  some of the projects which  previous reports have advocated  If these are to  be carried  out,  then  there will be environmental impacts arising  from these planned  developments,  and  due consideration  must then  be given  to  the nature of these impacts as well as to the data needs in the case of mitigatory requirements.
Upper Rasin
The ARDCO-GEOSERV report employed  a system of land  classification,  based primarily  on  30  soil mapping  units,  these units were in  turn  classified  into  land  classes, which  took  into  account limitations for agricultural use (ie land  capability).  This land classification  approach  allowed  for some assessment of the suitability  of different areas for different forms of land  use by  further taking  into  account the Length  of the Growing Period,  In  this manner,  land  suitability  was arrived  at for the following  forms of land  use: annual and  perennial cropping,  livestock,  forestry,  and  wildlife/catchment protection.  For crop  suitability,  further subdivision  was made on  the basis of expected  crop  yield  as a percentage of maximum yield,  to  give force additional categories: Highly  Suitable, Marginally  Suitable,  Not Suitable.  Using  this scheme,  land  suitability  for 14  rainfed  crops were mapped.
Under this system,  land  with  severe erosion  hazard  was allocated  to  forestry,  and  land with  yet further limiting  factors (constrained  by  inadequate soil depth/rockiness,  as well as erosion hazard) was deemed suitable only for wildlife or catchment protection.
Although  this is a fairly  standard  (if dated) approach  to  land  use planning,  there are some environmental concerns were it to  be adopted  as specified.  Firstly,  there are sound grounds for preserving tracts of indigenous forest between arable lands, irrespective of
TAMS-ULG BAXO-AKOBO FUVER basin INTEGRATED DEVELOPMENT MASTER plan
5*12environment
the land suitability cl asst Heal ion for that area presently under forest. The justification is as follows:
- conservation of biodiversity
-  maintenance  of  a  "network"  of  indigenous  tracts  of  vegetation  is  preferable  to  isolated islands; not only would this allow for "stepping stones" for bird migration (both at a local as  well  as  an  international  level),  but  would  also  serve  the  interests  of  the  economically important beekeeping sector, as Ethiopian bees are migratory.
-  dependence  of  people  on  forest  products;  in  this  case,  little  is  known  of  the  degree  of dependence  on  such  products,  although  the  importance  of  beekeeping  has  been  stressed (refer Section 5 5.3) 1
ARDCO-GEOSERV  also  gave  eight  site  options  for  dams  in  the  Upper  Basin,  and  any dam constructed will require its own environmental appraisal.
Lower Basin
The Russian  report maintained  that any  development of irrigation  on  the Gambela plain would  require regulation  of river flow,  including  flood  control and  drainage.  That study proposed  205  000  ha for irrigation.  For this the construction  of the hang  reservoir was proposed  (between  Gambela and  Itang  towns),  to  be supplemented  by  the Abobo reservoir (tinder construction  on  the Alwero  river),  and  by  water from the Chiru  reservoir. Two  alternative sites were proposed  for reservoirs on  the Gilo  river (again  requiring regulation).
The Russian report also proposed dam sites solely for hydropower generation, from
which diversion of water for irrigation was not considered, and favoured a site 50 km upstream of Gambela town (referred to as the TAMS project).
In the event of large scale cultivation in the Gam be Ila basin (as advanced by the Russian report), (he following impacts will arise;
-   alteration   of   the   current   pattern   of   flooding   on   the   Gambella   plain   (although consideration  in  the  Russian  report  was  given  to  the  obligatory  release  of  water  from upstream  reservoirs,  this  was  seen  as  being  for  the  purpose  of  providing  for  the  Abobo reservoir   rated   yield,   and   not   in   any   way   to   "mimic"   natural   flooding   patterns   on ecological   grounds).   The   Fisheries   submission   of   the   Inception   Report   points   to   a potential to increase fish yields by a factor of lour, il’this estimate is valid, then it is clear
1  the clear distinction between "honey hunting" and ‘bee keeping" is made in the apiculture submission: there is a
definite need for data on the relative importance of these two activities, in different localities within'the project area, and for different forest types.
TAN1S -U LG aaro-akobo river b asin integrated development master nENVIRONMENT
that any  disruption  of the natural flooding  regime should  not foreclose the option  to realise this potential Further comment on  some of the potential impacts on  fisheries arising  front impoundments is found  in  the Fisheries submission; al this stage,  any recommendation  for the introduction  of exotic fish  species (as advanced  in  (he Russian report,  where elaborates calculations arc made on  the potential for "fish-rearing") should be viewed with the utmost caution;
-   run-off  of  fertilizers,  herbicides  and  pesticides  from  lands  leads  to  the  dual  concerns  of eutrophication   (itself   impacting   on   fish   community   composition),   as   well   as   the   potential for concentration of pesticides in higher trophic levels,
- change in the status of bird, insect and rodent communities.
The Russian  report acknowledged  some of these concerns,  and  indeed  proposed  measures to  counter them (refer Annex  17  of that report).  Their work  involved  meticulous estimates of potential impacts for a range of agro-chemicals,  which  included  retention times.  They  estimated  that (depending  on  development option),  up  to  116  000  ha would be subject to  chemical pollution.  Measures to  counter adverse impacts were recommended: selection  of proper crops and  crop  rotation  patterns,  choice of optimum time and  methods of fertilizing,  and  the use of short-term pesticides.  However,  all these conservation  measures were proposed  from a perspective of a "top-down" intervention, as would  take place under a State farming  system.  Il is unlikely  that any  of these "agrotechnics! measures" proposed  by  (he Russians could  be implemented  under revised approaches to agriculture,
With  regard  to  floodplain  ecology,  the Russian  report recommended  research  on  the fish potential "of the inundated  flood-plain  and  its dependence on  the river hydrological regime".  This need  is reinforced  here,  but with  the additional requirement to  address a range of biotic and abiotic concerns, in addition to that of fish.
5.6       Principal Environmental Issues
Of  the  environmental  concerns  noted  above,  the  intention  is  that  only  the  principal  concerns  will be addressed in subsequent phases of the project. The principal concerns are listed below.
5.6.1      Human Population Pressures
I  he  rapid  (and  largely  uncontrolled)  change  in  numbers  and  location  of  human  populations  in the  project  area  is  the  major  cause  of  current  environmental  impact,  and  the  major  constraint  to effective land use planning for the Master Plan.
5*6.2 Floodplain Dynamics
From   an   environmental   perspective,   die   Master   Plan   will   require   an   integrated   model   of floodplain ecology, including;
TAMS-ULG
RfVER flASFN INTEGRATED DEVELOPMENT MASTER PLAN5.          ENVIRONMENT
-  knowledge of the extent to which various wild species and co nun unities are dependent upon the timing and intensity of water flooding;
’  the  role  of  nutrient  deposition  from  annual  flooding  in  the  maintenance  of  floodplain ecosystems and ecological processes;
- pheno logical changes of the herb layer in response to floodwater advance and retreat;
- associated patterns of movement of domestic herbivores;
- associated patterns (albeit historical in some cases!) of wild herbivores;
-    patterns   of   cropping   (including   planting,   weeding   and   harvesting)   in   response   to flood waters.
Some of these data are already  available from the Russian  report,  particularly  those data dealing with  pasture phenology,  livestock  and  cropping.  Nevertheless,  with  radical changes in demography  experienced  by  the area (including  the presence of people not traditionally  from the region), these patterns now need confirmation.
Data on the hydraulics of proposed impoundments will be made available in Phase IL In some
cases, it can be expected that for hydropower purposes, the flow into an impoundment will
approximate the flow* out (in which case there would no major effect on volume or timing of
Hooding). However, where water is released to match demand, or where the intention is to use impoundments conjunctively, then the timing and volume of release could be altered.
5.6.3      Wildlife and Forestry Issues
These are covered  tn  detail under the Natural Resources submission  (Sections 1.6  and  1.5). Briefly,  in  the absence of any  land  use system which  allows for the demarcation  and  enforcement of clear land-use zones,  virtually  all the above systems of land  use will impact negatively  on  both the wildlife resource and on indigenous forests.
As the situation  with  regard  to  forests is covered  elsewhere,  suffice it to  say  here that recognition of the importance of forests both  in  catchment conservation  and  as reserv  oirs of biodiversity  have been  longstanding,  the russian  report /the role of the forests as a soil and  water protection  factor is inestimable,  since they  keep  in  themselves the sources for rivers flowing  in  the gambella plain and  carrying  life to  the people./,  that report recommended  sanctuary  status for forests,  and stated: ’preservation  of the tropical forest biocenosis has now become an  issue of national importance’1.
5.7       Towards a Master Plan: data requirements
For the successful implementation  of the environmental component (and,  indeed  for most components) of a Master Plan  over such  a vast area as that of the BaroAkobo  Basin,  there are four levels of “operation'*,  each  to  some extent a precursor of the next,  and  each  requiring  some
TAMS-ULG BAROaKOBO RIVER BASIN INTEGRATED DEVELOPMENT MASTER FLAN
MS5.           ENVIRONMENT
degree of data supplementation:
*  the  first  level  concerns  data  on  the  nature  of  the  resource  base  and  on  current  patterns of land use;
* the second concerns the legislation necessary to meet Master Plan needs;
* the third level relates to the capacity to enforce the necessary legislation, and
*   the  last  level  relates  to  the  institutional  framework  for  effective  administration  of  the Master Plan.
for the first level,  it is a relatively  simple exercise to  identify'  outstanding  data needs,  and  even to  collect such  data within  an  acceptable time period.  The "shortfall" with  regard  to  necessary data for effective preparation  of the Master Plan  is of two  types.  Firstly,  there is basic information,  largely  of a scientific nature,  regarding  the status of natural resources.  Secondly, there are policy  documents,  which  although  not data per se, will impede Master Plan  preparation until their finalisation has taken place and their direction known.
5.7.1      Policy Refinement
There  are  several  areas  in  which  final  policy  decision  is  required  before  some  parts  of  the  Master Plan can be prepared.
The   most   important   of   these   is   the   National   Conservation   Strategy   (NCS),   which   is   under consideration for approval by the Council of Ministers.
The relevant point here is the intention  for Regional Strategies to  play  an  important role.  Indeed. Gambella has submitted  its own  Regional Proposals (not made available to  the consultant,  as the NCS secretariat was moving  offices al the time of writing),  but according  to  Sutcliffe (pers comm.) was one of the more impressive of the regional submissions countrywide.  Due cognisance in  any  Master Plan  will have to  be made to  these policy  statements,  and  project proposals will have to  be in  line with  these regional statements of intent with  regard  to  natural resource utilisation.
Further considerations with  regard  to  policy  "refinement” concern  the future status of key organisations related  to  natural resource management.  The Dept of Forestry  is a case in  point: previously  in  its own  Ministry,  it has been  relegated  io  the Ministry  of Agriculture with  a status that awaits clear definition.  Thus,  key  plans for natural resource management,  including  the Master Plan  of the Ethiopian  Forestry  Action  Plan  EFAP (with  which  the Baro-Akobo  Master plan  would  need  to  link) cannot be implemented  until such  time as institutional links have been finally determined.
5.7.2      Demography
From  an  environmental  perspective,  some  areas  of  demography  need  emphasis  which  might  not receive the attention under a standard census approach. Among these specific needs arc
TAMS-ULG flARO-AKOBO river basin integrated DEVELOPMENT master plan
5-1*5.          ENVIRONMENT
How many  people have moved  into  which  areas? What has been  the rate of increase in population,  for different areas? How does this compare with  projections made by  the Russian report? What arc the projections for population  levels al which  rates of increase will begin  to decline? Importantly,  what have been  the relative contributions to  population  growth  rates from various migrant sources (resettlement, refugees, nomads etc)?
Whilst the concern  of the development economist within  the resettlement sector is to  highlight grievances of settlers (and  thus to  cater for these in  any  subsequent resettlement),  the environmental sector needs settler-related  demographic data to  allow comment on  environmental impacts, and to meet the needs of effective zonation and land-use planning in the Master Plan.
5.73 Floodplain Ecology
To  meet any  mitigatory  requirements in  response to  river impoundments which  might disrupt natural patterns of flooding,  a basic understanding  of the ecology  of the Gambcla floodplain  will be needed.
Whilst the broad  principles of the floodplain  ecosystem have been  summarised  by  the Russian report,  more specific data arc needed  on  how various impoundments will affect the Liming  and volume of water for flooding. The following are among the data needs.
*  role  of  flooding  in  fish  breeding,  and  in  the  migration  of  fish  (ie  basic  understanding  of  fish biology as it relates to the specific type of flooding experienced on the Gambcla plains);
• role of flooding in nutrient deposition on pastures. What are the present values for annual
sedimentation rates'^ How reliable are estimates that have been made? How much will be retained 
under different impoundment scenarios, and of what significance will this be for plant and animal communities?
* role of flooding in the provision of habitats for both resident and migratory bird species;
*    role  of  flooding  in  the  maintenance  of  lakes  (both  permanent  and  seasonal),  and  in  the Limnology of those systems.
W ith rapid changes in human population pressures, there is a subsidiary need to monitor water
quality, particularly as fish community composition can respond to changes in quality. The
Russian report presented data on water chemistry including turbidity. Ph, mineralization, no supplementary data have been collected since.
A  clear  model  is  needed  to  indicate  how  various  combinations  of  impoundments  will  affect timing  and  volume  of  water  release:.  What  will  be  the  relative  contributions  of  upland  and lowland  impoundments?  To  what  extent  does  current  flooding  of  the  Gambell  a  plains  result  from backwater   movement   from   the   Sudan   (if   this   is   a   major   contribution,   then   impoundments upstream will have relatively little impact on flooding patterns).
Without this information, it will not be possible to predict the consequences of impoundments
TAMS-ULG baro-akobo rivfx basin integrated development master plan
$475.           environment
for downstream ecological processes, and no satisfactory mitigatory procedures can be proposed.
5.7.4      Wildlife
Detailed  data collection  needs for wildlife arc covered  in  the Natural Resources submission. Nevertheless,  there is one wildlife-related  issue which  cuts across several other forms of land  use: the need  for a simplified  system of vegetation  classification  which  is standardised  for Ethiopia. or at least for the WRDA basin studies.
5.7.5      Land Use Change
In  order to  more effectively  model the impacts of proposed  project interventions,  it is necessary to  know the rates of change of various forms of land  use,  together with  the loss of natural resources or vegetation  types that this implies.  The report of the Land  Use Planner also  draws attention to the need for definition of changes in land use over time.
These  rates  of  change  should  ideally  be  linked  with  human  population  density  classes.  Firstly, some common system of vegetation classification needs to be adopted.
Using aerial photographs, it should be possible to measure:
- rates of loss of indigenous forests;
- rates of change in grazing areas;
-  rates  of  change  for  arable  lands  (changing  ratios  of  grazing  land  to  arable  provide  can provide indices of the nature or environmental impact trends);
- nature and extent of wetland drainage in the Upper Basin.
Other data needs related to changes in the patterns of land use include:
•   to  minimise  impacts  due  to  deforestation,  effective  management  of  forest  resources  is  needed at   the   level   of   peasant   associations:   to   devise   management   strategies,   a   great   deal   more information   is   needed   on   regeneration   rates   of   indigenous   species,   against   which   to   devise models for sustainable harvesting;
•  quantification of developing charcoal industries;
• present distribution of tsetse fly;
5.7.6      Soils
Although the Russian report acknowledged that their proposed agricultural interventions would impact negatively on environmental parameters like groundwater, soil fertility, and water quality, 
their proposals for environmental conservation were consistent with the concept of a large
TAMS-UI.G          BAJtO-AKOBO RJVER basin integrated development mas ter pi ans.      ENVIRONMENT
intensive ly-cropped  State Farm.  For example,  the use of "Forest Shelterbelts*  was advocated,  to be planted  along  the boundaries of irrigated  lands.  Thus,  these proposals now have reduced relevance, and a new framework must be defined for environmental protection.
Data   are   needed   on   appropriate   soil   conservation   measures   ie   those   which   arc   acceptable culturally, and which are practical.
5.7.7    Institutional Structures
Some information  will be needed  on  the institutional structures necessary  to  implement the regional strategies of the NCS,  and  to  assist in  management of protected  areas.  The Wildlife submission  (Section  J .6) contains further comment on  the potential role of committees for community-based  protected  area management programmes.  What structures for example,  will assist UNHCR in  providing  environmental safeguards on  the siting  of refugee camps? To  some extent,  the structure and  linkages of relevant bodies will depend  on  the outcome of policy documents (eg Wildlife Policy).
TAMS-ULG baro-akobo river basin integrated dfvelopment .master plan5.           ENVIRONMENT
REFERENCES CITED
Abcbc.D & i Iagos,  E(1991) Plants as a primary  source of drugs in  the traditional health  practices of Ethiopia.  IN: Engels,  J M M,  Hawkes,  J G ,  & Melaku  Worcded  (eds ).  Plant Genetic Resources of Ethiopia. Cambridge Univ. Press, pp 101-113..
ChafTey, D R (1980). Forest Inventory of south west Ethiopia.
Edwards,  S.  (1991).  Crops with  wild  relatives found  in  Ethiopia.  fN: Engels.  J M M,  Hawkes, J G ,  & Melaku  Worcded  (eds).  Plant Genetic resources of Ethiopia,  Cambridge Univ.  Press,  pp 42-74.
Ethiopian  Valleys Development Studies Authority  EVDSA (1990).  Baro-Akobo  Upper catchment comprehensive development Master Plan,  reconnaissance Study  report on  socio economics and environment.
Unpubl. report, 45 pp.
Friis  I,  Rasmussen,  FN  &  Voilesen,  K  (1982).  Studies  in  the  flora  and  vegetation  of  southwest Ethiopia. Opera Botanica 63: 1-70.
Hedberg. I (ed,), research on the Ethiopian flora. Symbolae Botanicae Uppsalicnsis 26 f21:31 -47.
EWCO   (1989).   agroecological   Zonations   study,   natural   resources   conservation   &   Development Dept.. Ministry of Agriculture.
World Conservation Monitoring Centre (1991). biodiversity guide to Ethiopia. 25 pp
National Conservation Strategy Secretariat. National Conservation Strategy.
Agriconsulting    (1993).    Final    Report.    Feasibility    study    for    wildlife    conservation    project    in southern Ethiopia. 211 pp.
UNESCO/UNEP   (1990).   The   impact   of   large   water   projects   on   the   environment.   Imprimeric Duculot, Belgium
Woodford,  J  D  (1990)  The  status  of  wildlife  in  Ethiopiarits  conservation  and  utilisation.  EWCO. Hillman, (1993). Ethiopia: Wildlife Conservation Compendium. Vols I & II.
White,   F   (1979).   The   Guinea-Congolian   region   and   its   relationships   to   other   phytochoria.   Bull, Jardin.J3.ot- Nat, de Belgique 49 11 -55
Aiemneh  Dcjcne,  1990.  Peasants,  environment,  resettlement In: Siegfried  Pa use wan  g.  Fan  ru Cheru.  Stefan  Brune and  Eshetu  Choie.  Ethiopia: Options for Rural Development. Zed  Books Ltd,, London. Pages 174-186.
T.AMS-L LG baro-akobo rjver basin Integra fed devfi opment master pi an
5-205.          ENVIRONMENT
Ethiopian  Valleys Development Studies Authority  (EVDSA).  July  1990.  Reconnaissance Study Report on  Socio-Economics and  Environment. Baro-Akobo  Upper Catchment Comprehensive Development Master Plan. Addis Ababn.
Lo,  C  T,  Kloos.  H  &  Hailu  Birrie.  1988.  Schistosomiasis.  In:  Zein  Ahmed  Zein  and  Helmut Kloos, The Ecology of Health and Disease in Ethiopia Ministry of Health, Addis Ababa.
Mekonnen Abebe, Wondatir Negatu and Senait Asseffa. 1988. Trypanosomiasis. In: Zein 
Ahmed Zein and Helmut Kloos. The Ecology of Health and Disease in Ethiopia. Ministry of
Health. Addis Ababa.
Mend  Mekonnen.  1988.  Yellow  Fever.  Id:  Zein  Ahmed  Zein  and  Helmut  Kloos.  The  Ecology of Health and Disease in Ethiopia. Ministry of Health, Addis Ababa.
Ncgussie Gebre-Mariam,  Yahya Abdulahi and  Assefa Mebrate.  1988.  Malaria.  Jjj; Zein Ahmed  Zein  and  Helmut Kloos.  The Ecology of Health  and  Disease in  Ethiopia. Ministry  of Health, Addis Ababa.
SELKHOZPROMEXPORT  Al! Union  Foreign  Economic Corporation,  Soyuzgiprovodkhoz Institute.  1990.  Baro-Akobo  Basin  Master Plan  Study of Wafer & Land  Resources of the Gambela  Plain, Final Report (Draft).  Volume If - Main  Report.  Prepared  for the People's Democratic Republic of Ethiopia,  Ethiopian  Valleys Development Studies Authority,  Addis Ababa-
United   Nations   Development   Programme   (UNDP).   August   1993.   Gambella   Needs   Assessment Survey {Region 12). Group Report, Draft Report. UNDP, Addis Ababa.
UNHCR (1982). Handbook for Emergencies. Part One; Field Operations.
World     Bank.     October     1991.     Operational     Directive     4.01,     Environmental     Assessment. Washington, DC.
T AMS-ULG baro.akobq river basin integrated development master plan5.           ENVIRONMENT
Lisi of Acronyms
ARDCO-GEOSERV
CHA Controlled Hunting Area
EFAP Ethiopian Forestry Action Plan
EVDSA Ethiopian
EWCO Ethiopian Wildlife Conservation Organisation
FAO Food & Agriculture Organisation of the United Nations
ICDP Integrated Conservation Development Planning
IR Inception Report
NP National Park
NRDEP Natural Resources Development and Environmental Protection NCS National Conservation Strategy
UNHCR United Nations High Commission for Refugees WRDA Water Resources Development Authority WWF World Wide fund for Nature
TAMS-ULG BARO-AKOBO RIVER basin integrated DEVELOPMENT MASTER plan■J J 
i
I J
I *.
II

Summery