The People’s Democratic Republic of Ethiopia Water Resources Development Authority
AMIBARA IRRIGATION PROJECT II PASTORALIST & FORESTRY DEVELOPMENT STUDIES
Interim Report
MacDonald Agricultural Services Limited
Demeter House. Station Road. Cambridge C0i 2 RS UM August 1990The People’s Democratic Republic of Ethiopia Water Resources Development Authority
AMIBARA IRRIGATION PROJECT II PASTORALIST & FORESTRY DEVELOPMENT STUDIES
Interim Report
MacDonald Agricultural Services Limited
Demeter House, Station Road, Cambridge CB1 2RS UK August 1990U)
Summary
The. development ofi irrigation tn the middle Awash. project area has resulted Ln
the loss o{ A^an dry season grazing bordering the riven, and the displacement
Afar. settlements. The most noticeable consequences o^ this change in land use ane:-
(a)   a loss oi 23 000 ha o{ dry season grazing representing about 69 000 tonnes on 37$ o{j the total annual dry matter production (livestock faed) in the
project area
(b)    the concentnation A^an herds, and an resultant increase in grazing pressure
on faagile range lands
(c)    the destruction o^ fanest and Moodland, nesulting faom the increased demand fan faeluiood, estimated at 80 000 tons pen yean, faom the large numbers ofi Monkens on the irrigated state fanms
(d)   environmental degradation and Soil enoSion consequent on the above change
Various attempts have been made to deal Mith the pnoblems, including settlement, farming on small holdings and innigated pasture development.
This report addresses the above pnoblems and the attempted Solutions, and in the light o^ this analysis presents proposals fan an Afian programme,(^c)
The 54/ 377 ha project aAea it a bAanch the Rift Valley,
u)tth extentlve aAeat alluvial ioilt, iuppoAtlng declduout Acacia buthland.
The climate it tAoplcat ieml-aAld with an aveAage annual Aalnfall of 500mm. The pAetent land ute comptiltet a gAoii iAAlgated cotton aAea of 22 84/ ha, plot tevcAal tmall
A($cw. faArrt, the Aett of the valley being uted by the AfaA foA gAazlng.
The aAea it occupied by tome 30 AfaA ctan-i of the Veblne and Walma tnlbet, thelA
total ettlmated population being 68/25. In addition thete aAe tome 5/255 non AfaA
State Taom woAheai, plat taall numbeAt of KeAeyu and Itta.
The patt attemptt to tolve the AfaA pattoAal pAoblem one analysed. Thete Include settlement, foaming on small holdings, and iAAlgated pattuAet. None have been
veay successful. With. Aegaad to the foAettny paobtem the Aecently staAted planting
will only go a tmall way to pAovldlng the fuelwood and building tlmbeAt AequlAed.
The Aeatont foA the United, success of attemptt to help the AfaA aAe alto analyted.
Thete Include a faltuAe to Aetpecl AfaA clan tQAAltoAy; and schemes which weae
(a)    too big, with complex, management pAecludlng AfaA paAtlclpallon, and
(b)    eitheA economically oa Institutionally un sustainable. The cuAAent iAAlgated gAazlng tcheme will suffeA fAom thete pAoblemt, In addition to sevete technical pAoblemt 0^ management.
Against this gAound, pAopotalt aAe made foA a five past pAogAamme:
-    An Advance PtogAamme:
-    A Livestock and AgAlcultuAal PAogAamme
-    A foAettny and tAee planting pAogAamme
-    A training pAogAamme
-    Institutional aAAangementt
The AfaA aae suspicious of any Initiative because of thelA feaA that it will lead
to a fuAtheA lots of grazing land. The Advance pAogAamme, aimed at building up AfaA confidence, would IhcAefoAe pAovlde foA needs expAetted by the AfaA: human and animat health, and education, plut IntCAlm tuppoAt foA the Aun down RRC settlements.The. other initiatives . would mostly" be based on JOO ha blocks 0($ land
either within, or on the edge o^ the existing irrigation schemes, each block being
located within a specific Ilan Territory. These block* would be used far irrigated pasture, Small scale {arming and tree growing amongst other thing*. To date 5000ha have been dedicated to A{ar uselihe Amibara settlement and the irrigated pasture scheme)
The concept underlying the 100 ha block programme would be a distribution o{ this land: some o{ the 100 ha block* being on State Farm land while the latter would gain access to an equal area o{ dedicated Afar land.
Proposals are also made far possible crop diversifacation on the State Farm* to pAovide crop by-product* far animal faed; an Alledeghi.'Plain* and an animal requisites programme. The farmer would probably call far the supply o{ capital equipment.
In addition to tree planting on 100ha block*, proposals are also made far planting on roadside* within, and on the eastern boundary o{ the irrigation schemes, a* well a* on abandoned Saline land.
Proposals are also made regarding training and institution*.
The proposed programme would last ten years, a minimum time in which to achieve a degree o{ progress with the Afar pastoral problem. This would also enable the proposed approach o{ more closely associating the Afar with the State Farm* to be
{ully tested, white at the same time the efficiency of the State Farm irrigation
could be improved. It has been recommended elsewhere that the latter should be accorded high priority, and should precede any farther irrigation development. It is envisaged that these two ob jeetives which would be reached about the Same >: time, would then enable the Boihamo Angelele extension to be implemented, the lessons from the above programme having been incorporated in the scheme.Amibara Irrigation Project II Pastoral and Forestry Development Studies
Contents
SUMMARY
I         INTRODUCTION
II       THE ENVIRONMENT
2.1 The Project Area
2 .2 Geology, Land Forms and Soils
2 . 3 Lakes and Springs
2 . 4 Ground Water
2 . 5 Climate
2 . 5 Vegetat ion
2 . 7 Present Land Use
III      THE PEOPLE OF THE PROJECT AREA
3.1 Afar Clans and Territories
3.2 The I ssa
3.3 The Kereya
3.4 The Afar Population
3.5 The Non Afar Population 
3.6 Population Increase
IV      LIVESTOCK
4.1           The Pastoral Problem
4.1.1 The Livestock Population
4.2           Attempts to Solve the Pastoral Problem 4.2.1 Settlement
4.2.2 Farming on Small Holdings
4.2.3 Irrigated Pasture
4.3           Conclusions
V        FORESTRY AND TREE GROWING
5.1           The Forestry Problem
5.2           Energy
5.2.1 Fuelwood
*
3
4
4
4
8
9
9
11
13
17
17
17
18
21
23
25
26
26
26
28
28
31
33
38
39
39
39
405.2  . 2 Cotton Stalks and briquetting 5.2.3 Charcoal
5.2.4 Dung
5.2.5 Electricity
5.2.6 Kerosene
5.2.7 Domestic Fuel Consumption
5.3   Construct ion Timber
5.4   Other Wood Uses
5.5   Exi st ig Wood Resources
41;
/***
*.• « / '
43/
U-
45.
VI AN AFAR PROGRAMME
6.1   Introduction
6. 2 The Amibara II Irrigation Scheme and the Afar Programme 6.3   An Afar programme
6.4   The Advance Programme
6.5   The Livestock and Agriculture Programme
6.5.1 Angelele Irrigated Pasture
6.5.2 Small Irrigated Pastures and Units within State Farms 6.5.3 Small Irrigated Pastures and Units outside State Farms 6.5.4 Improved Management of the RRC Settlement Scheme
6.5.5 Crop diversification on State Farms
6.5.6 Development of Cotton Oil Extract ion at Melka Werer 6.5.7 Integrated Water Supply, Fodder and Range Conservation 6.5.8 Alledeghi Plains Programme
6.5.9 Animal requisites programme
6.5.1OL1veatock Marketing
6.6 Forestry and Tree Planting Programme
6.6.1 Development of Afar Owned Tree Plots for Firewood 6.6.2 Planting of Roadside Verges within Irrigated Areas 6.6.3 Tree Planting on the Eastern Boundary
6.6.4 Tree Planting on Abandoned Saline Areas
6.6.5 Tree Planting around State Farm Worker's Villages 6.6.6 Planting of Fruit Trees and Browse
6.6.7 Establishment of Shelter Belts 6.7  Training Programme
6.8   Institutional Aspects 6.9   Implementation
47
47;: bo. 5.1. S3- 53
54
54:
54
55
56
56
57
57
58
58
58  • 59. 59
59."?. 60 V wu
61  ■' 61
627; 63.'.,
9Amibara Irrigation Project II
Pastoral and Forestry Development Studies
I INTRODUCTION
The Water Resources Development Authority,  which  is responsible for the development of IrrigatedAgriculture,  i nc ludi ng  t he R i ver Awash,  hopes to  implement the extension  of the Amibara - Bolhamo Irr igat ion  Scheme .  Th  i s extension  would-di sp  lace a number of local Afar pastoralists.  In  the past such  displacements have resulted  in a serious disruption  of Afar society,  due to  the loss of critically important areas of dry  season  grazing.  Because past attempts to compensate the Afar for this loss of grazing  lands have not been very  successful,  the EEC provided  finance for a study  by  the
British         firm         of         consultants         :         MacDonald         ’s         Agricultural         Services, 
with the following terms of reference:
(a)    to provide sound proposals to alleviate prevai ling and future
problems of pastoralists in the region
(b)     to devise an appropriate plan to restore the ecological
balance, following the expansion of irrigation in the region.
The        team        comprised        -        Mr        M.A.Brunt:        team        leader/land        use        planner;
Mr         M.A.Rhi         nd:         livestock;         Mr         R.M.Bennett:         forestry;         Mr         P.Devitt:
social          anthropology;          Mr          R.Goodyear:          irrigation          agronomy,          plus          Mr
Kidane         Wolde         Yohannes         :         1         ivestock/range         management;         Mr         Tedesse
Wolde            Michael:            forestry;           Mr            Tibebu            Teshome:            sociology;           Mr Alemayehu       Geleta;       environmentalist;       Mr       Mahamud       Mohamed       and       Mr       A1       i Hurduno:       Afar       technical       assistants;       Mr       Abdi       Hassan       and       Mr       Mohamed Dawed: Afar translators; Mr Ali Hassan: guard.
This Interim Report is financed  by  th$,  European  Communities from credits of the European  Development J?und,  and  is presented  by MacDonald  Agricultural Services Limited  for consideration  by  the
Ethiopian             Administration 
.             It             doesnot             necessarilyreflect             either
the opinion  of the latter or of the European  Communities.  The report covers the work  of the Consultants in  the field,  which  was undertaken between June and August 1 990.1 I THE ENVIRONMENT 2 . 1 The Project Area
The approximate area was defined in the terms of reference by:
-  Lat itude 9 degrees N to 10 degrees N and
-   Longitude 39 degrees 45’ E to 40 degrees 30' E.
Following 
initial
fieldwork 
revised 
project
boundaries              were
confirmed  at a meeting  with  WRDA,  the EEC and  the Consultants on July  3  1990.  The revised-boundaries,  compr ising  the 1000  m contour to  the east and  west of the Awash  River,  were further revised  after subsequent work  revealed  the presence of appreciable numbers of Afar livestock  above the 1000  m contour.  The boundary  finally adopted  was the 1  305  m (4000  ft) contour.  The project area thus comprises the alluvial floor of the Middle Awash  Valley,  and  the flankinggravel fans; it extends from Awash  Station  in  the southto Lake Hertale in  the north.  The greater project area covers 925  450 ha,  of which  the land  below the 1000  m contour comprises 541  377  ha. The ILCA survey covered a slightly larger area namely 745 000 ha.
2 . 2 Geology, land forms and soils
The     middle     Awash     is     defined     by     Tertiary     R     if     t
;         Valley     faulting     to     the
east and  west,  the valley  floor having  been  block  faulted downwards.  Within  this area,  a great thickness of silts,  gravels and  boulder beds were deposited  on  the valley  flanks ,  while in  the area close to  the present Awash  ,  true flood  plain  deposition occurred,  giving  rise to  alluvial and  heavy  clay  plains: the Bolhamo-Amibara-Maro  Galo  plains.  Previously  the course of the Awash  crossed  the Alledeghi Plain,  at which  time this great expanse of black clay was deposited.
Subs eq  uent ly.  volcanic /act ivity  in  the Awash  valley  result ed  i n; t he intrusion  of basaltic and.  cinder cones through  the alluvial deposits .  From north  to  south  the main  basalt cones and  assoc latd lava f lows are :I I 'rill*', KNV 1 HONMKNT
2 . I The Projec t Area
Th e app r ox i m a t ♦* a r e a w a ■» < I •• t I n e <1 in I hn l t • i • m t» < • t co I tirnuut by hat ttudr '» ilriiroi'" N Io IO dagrwnm N .mil
Longitude »‘) degrees 4 !> ' E In 40 fleginrm 30' E ,
Following 
initial
fieldwork 
revised 
project
boundnrlwrt               were
con  f i r med  at <i meet Ing  with  Wl<I)A ,  the E!'.(.1  mud  I lit*  IIoiimu 1  tmit n  on July  3  1990.  The rev  1  sed'boundin'i ms , comp  r I n  I,  nq  the 1000  in  contour to  the east and  west of tho  Awnnli River .  were further revised  utter subsequent work  revenied  the presence of appreciable numbers of Afar livestock  above the loot) m contour.  The boundary  finally adopted  was the 1305  in  (4000  ft) contour.  The project area thus comprises the alluvial t loor of the Middle Aw as h  V a 1  .ley  ,  and  the flanking  gravel fans; 11  ex  tend  s from Awash  Stat 1  o  n  .1  ri t h® south  t o LakeHertale in  the north.  The greater project area covers 925  450 ha,  of which  the land  below the 1000  in  contour comprises 541  377  ha. The ILCA survey covered a slightly larger area namely 745 000 ha.
2.2 Geology, land forms and soils
The mi ddle Awash  is defined  by  Tert iary  Ki ft,  Vai ley  f aul t .1  ng  t o  t he east and  west,  the valley  floor having  been  block  faulted downwards.  Within  this area,  a great thickness of silts,  gravels
and       b      ou      Ider      beds      were      deposited      on      the      valley      flanks      ,      while      In      the
area           close          to          the          present          Awash          ,          true          flood          plain          deposition occurred,           giving           rise           to           alluvial           and           heavy           clay           plains:           the Bo         1hamo-Am         1bara-Maro         Galo         plains.         Previously         the         course         of         the
Awash           crossed          the          Alledeghi          Plain,          at          which          time          this          great
expanse of black clay was deposited.
Subsequent ly.Volcanlc/actlvity lnthe Awash valley resulted i q t he intrusion of basaltic and cinder cones through the alluvial
deposits, from north to south t he main basalt cones and assoclatd lava t 1 ows are :Mount ai n                  He i ght(m)
Aside Ehaile Derabale
Dof an Gurmile Filweha
1 099
112 1
1 15 1
8 26
1 5004-
The peak  of Filweha lies just south  of the project area.  These mountains are a conspicuous feature of the Middle Awash,  and contrast sharply  with  the valley  floor,  the average height of which  falls from about 850  m in  the Awara Melka area,  to  550  m at the southern end of the Maro Gala plain.
The Middle Awash was divided into seven major land form units by Currey ( 1972,1973) . These are listed and described below, 
together with the area of each unit within the project area. 
The undifferentiated Hill Margin is land outside Currey’s area, which wi 11 largely be Gravel Fans , and Outwash Plains . These land form units are shown on Map 1 .
Land form unit
Alluvial Black Clay 
Alluvial Brown Soil
Alluvial River Levees
Gravel Fan
Gravel Outwash Plain
Acid Lava
Basalt
Undifferentiated               Hill Margin 61 Lakes/Springs etc. 
Total Project Area
Hectares
100 010
25 127
2 1 159
111 912
64 469 95 878 56 700
824
4 298 54 1 377
These        units        were        adopted        by        the        project        as        the        basis        for        the development proposals.10-00’
Map I : TMg LAND FotfvMS OF THfi
BC ’. AHuviU
S/<6< Chy
•  QlluviU. Rrouin Soil
• A IIuviaX. XiStr l*'v«r
! travel
pAf\
: GraveX Ouh*tk P/ain^y
•  AciW L-ava
*  Baaa/6
: <A»«6X Hill Q*
K*rr»*t Sni*»
IP.GiMTuO PASTu^g
1
1
s
ur ? X v V (
1L
LIJ
L1J
LIJ LIJ
•l J LIJ
•l J
•IJ ..J
Alluvial         Black         Clays         (Vertisols)         -         these         probably         formed         in         situ over           the           newer           basalt           lavas,           and           these           soils           are           relatively shallow.         The         black         clays         were         also         transported         ond         deposited         in basins         of         internal         drainage,where         the         soils         are         deeper.         They         have been        dated        as        less        than        4000        years        old.        The        soil,        which        develops pronounced      surface      cracks,      has      a      high      moisture      retent     ionc      ap      ac      i      t     y      , which          supports          vegetative          growth          through          the          dry          season.          Currey 
(1973)        noted        "        the        natural        vegetation        of        these        soils        has        been        the saviour       of       tha       Afar's       cattle,       for       as       other       areas       .       .       .       are       desolate with          numerous          animal          carcasses          strewn          around,          it          is          here          in Angelele           that           the           animals           are           still           able           to           survive           without difficulty."
Alluvial Brown Soils - these are mainly derived from the Gravel
Outwash Plains - from which the silt and clay has been washed. They 
usua1 ly occur at higher elevations than the A1luvial Black Clays, 
and are the main soils of the irrigation schemes. In the Angelele
area they have been carbon dated as less than 12 000 years old 
(Currey 1973).
Alluvial River Levees - levees at a higher level than  the flood plain  of alluvium have formed  along  the Awash,  as a result of alluvial deposit ion  along  the banks during  times of flooding.  Such Levees also  mark  former abandonded  courses of the river .  In  places alluvial fans abut the levees where the river breached  its banks in the past (e.g.Bolhamo  area) .  The Levees and  adjacent alluvial land which  were subject to  deep  flooding  before the Koka dam was constructed,  are character i sed  by  dense Acaci a ni lot ica forest and woodland.
Gravel Fans - these deposits are derived from the breakdown products from the lava/tuff-clay sequences of the higher ground bordering the Awash River. They are contemporaneous with the outpourings of the lavas with which they are interbedded. Beds of clay also occur, some of which are saline.II" II
IW ’ I i If I 'I I 'll r
"
The soils,  which  are brown  to  yellow,  are shallow and  are prone to wind  erosion  after the vegetation  has been  overgrazed,  e.g.  on  the land  bordering  the access road  to  Melka Werer.  Erosion  surfaces reveal salt accumulations.
Gravel Outwash  Plains - these are similar to  the Gravel Fans,  but have gentler slopes.  Gravel ridges occur,  e.g.  to  the north  and south  of Lake Bilen,  above the general plains level,  and  in  the Bolhamo  area gravel terraces occur which  are above flood  level, where concentrations of Afar huts are found.  The soils are similar to those of the gravel fans.
Acid  Lava - these Eocene lavas,  interbedded  with  gravel beds,  and tuff layers,  comprise the hill masses flanking  the Middle Awash. They  are associated  with  the major faulting  which  gave rise to  the Awash  and  the Alleghedi valleys.  The latter was subsequently filled  inwith  Alluvial Black  Clay  to  formthepresentplain.  These near horizontally  bedded  lavas form the eastern  margin  of the present Awash  Valley,  giving  rise to  the low north-south escarpment,  on  which  the Haler ow construction  camp  is sited  to  the west of Melka Werer.  The tuff clays readily  erode to  form the Gravel Fans and  the Outwash  Plains,  with  local basalt gravel deposits.  These lavas also  form a topographic barrier at the southern  end  of the Maro  Gala Plains.  Springs are associated  with faults in  the lavas.  South  of Kessem-Kebena the lava plain  is covered by a black basaltic clay, with numerous rock outcrops.
Basalt - these Plio-Pleistocene to  Recent lavas are associated with  two  of the major Rift Valley  faults 7  km apart.  They  give rise to  the major mountains,  the associated  lava flows and  cinder deposits noted  above,  as well as to  hot springs (e.g.  Filoa north of Fantale),  fumerolic activity  and  sulphur deposition.  Pockets of soil occur, but much of the basalt lava surface is bare rock.
Undifferentiated Hill Margins - these will mainly comprise Outwash Gravels, Gravel Terraces and Fans described above.
7B#«||LWI J 1 • |> J VI il lT II W 1 H w...................................
b s <r        in
WI.JJB
2.3 Lakes and Springs
There are four major lakes in  the area: Hertale (salty) in  the nort h  ; Lehadu  and  Asbahari near Ehaile mountain,  and  Deb  hi le Bahir on  the flanks of Dofan  Mountain.  There are also  several crater lakes.  In  addition  there are important springs which  are listed below from north to south:
Springs                Water Temperature
(degrees C)
Discharge                    Water Quality
( l/'sec )
(total dissolved solids mg/1)
L. Hertale
50
1
-
(E.shore) Metaka
65 and 84
350
-
Little Bi len Bi len 
40 to 43
- 100(est)
- 65 0
Filoa (Rahena) 44                                          1000                         1000
The Metaka Springs are just outside the northeastern  boundary  of the project area.  The Bilen  Springs,  which  discharge into  the Awash  River via a series of channels are a key  watering  site for the Afar inthe dry  season.  The Filoa spring  is similarly  important .  It discharges at a rate of 1  cumec into  the Bulga creek.  In  1973  when both  the Kessem and  the Kebena Rivers dried  up,  this creek continued  to  flow.  Currey  (1973) has suggested  that the Little Bilen  Springs,  to  the north  of Bilen,  could  be developed  by  the construction  of acont inuous interceptor channel along  the toe of the escarpment.  A water balance study  would  be needed  however
first
to          ensure          against          the          over-development          of          the          Alledeghi
ground water resource.
There are a number of water holding  surface depressions in  the AlluvialBlack  Clays.  The largest,  south  east of HaleDebii called Kortume,  is an  important grazing  area of seasonally  flooded Sporobolus grassland.  Smaller Acacia lined  depressions occur on the Alledeghi plain,  which  are used  by  the Afar as watering  points in the wet season.4 Ground Wat er
The       around       water       resources       of       the       Alledeghi       plain       are       of       interest because      the      use      of      this      area      for      grazing      is      limited      in      part      by      the lack      of      water      The      FAO      t°am      from      the      Australian      State      Rivers      and Water       Supply       Commission       of       Victoria       dri       lied       eight       product       ionbore hole*   in   1   Q   7   1     foil   ow   i   ng   an   exploratory   programme.   Cur   r   ey   (   1973) record*       ♦       he       following       data,       the       bore       hole       locations       being       shown on *1ap 2 .
Fore
Hole
Depth Water Y ip J d         Draw down Depth          ( L / m i n)         ( 1 2 hours)
(«)
(m)
1
1 50
-
-
-
2
-
-
-
-
3
-
-
-
-
4
101
86
400
1
5
1 1 2
abandoned
6
90
85
-
-
7
-
89
-
0.6
8
70
abandoned
Data from bores 1-3  are not available.  It is believed  however that the current Alledeghi Holding  Centre and  Quarantine Station  draws its water from bore hole 3.
2 5 Climate
The climate is tropical semi-arid,  the annual rainfall,  which  is weakly  bi modal,  varying  from 390  to  790  mm.  The 1  in  5  year maximum twenty  tour hour rainfall has been  estimated  to  be 57  mm (Ha 1crow,1895) .  The variation  in  the annual rainfall at Melka
Werer for the period 1965-1989 is shown in Fig 2.1.
The considerable variat ion  between  years will be noted; 11  years being  at or above the annual average rainfall of 566  mm,  compared with 14 years below,W < W II » •
Temperatures vary  from mean  monthly  minima of 1  1  .  6  C and  2  4  .  9C to mean  monthly  maxima of 3  2.9C and  40C in  December and  June
respectively.  Mean  relative humidities are lowest in  June (36.6%) and  highest in  August (58.0%) .  Mean  daily  sunshine reported  on  an annual basis is 8.5  hours.  The highest wind  speeds occur in  June (monthly  average 9.09  km/hr ) ,  in  which  month  and  July  the wind  run 
may  exceed  200  km/day.  During  thewinter,  the air flow is from the NE,  while the stronger summer winds come from the SW .  Very  strong 
windsareassociatedwiththe
intense
summer
thunderstorms. 
Annual
potential
evapotranspiration,             estimated             by 
the             Penman             method, 
approximates to  2  400  mm,  ranging  from a monthly  mean  of 170  mm in August to  25  2  mm in  June.  A summary  of monthly  meteorological data from I AR at Melka Werer is shown in Table 2.1.
MelkaWerer for 1965-84.
Rainfall MeanTemp Rel.Humidity Hours of Windspeed Radiation
(mm)
(Deg C)
(%)
Sunshine
(km/h)
(ca/cm2/d)
Period (years) 20
20
20
16
16
15
J
15.89
23.53
50.19
8.45
5 . 24
49 2.7
F
47.62
2 4.62
52 .40
8 . 54
5.40
529.7
M
59.10
26.47
48.60
8 . 23
5 . 76
5 6 1.3
A
53.13
27.63
49.40
8.49
5.46
582 . 3
M
28.66
28.92
42.10
8.96
6 . 72
584 . 1
J
2 2.95
30.69
36.60
8.59
9.09
577.5
J
108.09
28.57
49 . 30
7.20
8.83
545.5
A
119.50
27.00
58.00
7.32
7.48
562.4
S
38.68
27.62
52.10
7 . 84
6.29
562.2
0
24.11
2 6.08
44.20
9.06
5.58
5 71.9
N
23.27
23.99
46.90
9.54
5.45
532.1
D
2.54
2 2.66
48.80
9.24
5.11
502.7
Y
543.59
26.48
48.22
8.46
6.37
550.32 . 6 Vegetat ion
The dominant vegetation  of the project area is deci duous bushland, there is however considerable local variation  from the Sporobolus grassland  on  the Alledeghi Plains to  the closed  canopy  Acacia nilotica forests borderimg  the Awash  river.  The major vegetation types are described  below in  relation  to  the land  form units of the project area.  More detailed  descriptions of the vegetation  in  the Kesem-Kebena area will be found  in  Macdonald  (1987) Annex  A,  and Floc'H (1981,1982) and Cory (1981,1982).
Alluvial Black  Clay: most of the Black  Clay  areas in  the Melka Sadi,  Amibara,  Angelele,  Dolfan  and  Gidian  blocks have been 
cleared           for
irrigation.           The           Alledeghi           plain           is           almost           pure
Sporobolus grassland,  with  depressions in  which  water collects, 
whose margins support 5-7  m high  Acaciatortilis.  The Black  Clay  in 
the Angelele irrigated  pasture area,  which  is being  cleared, 
supports Acacia tree and  shrub  grassland,  some trees reaching  15m
plus;            while
that
bordering            the            Kortume            seasonally            flooded 
grassland  has been  invaded  by  Acacia nilotica forest from the adjacent River Levee area .  The Black  Clay  area north  of the Kebena river is characterised  by  Salvadora persica (tooth  brush  tree) clumped thicket.
Alluvial Brown  Soils: large areas of these soils have been  cleared for irrigation.  Where they  occur adjacent to  the River Levees they carry  closed  canopy  Acacia nilotica forest up  to  30  m high. Elsewhere Acacia open  woodland  occur s compr ising  A .  nilotica,  A. tortilis and  locally  Tamar ix  aphyalla; an  under st ory  of scattered evergreen  shrubs may  also  occur ,  grass and  herbaceous cover being sparse.  South  of the Kessem river the seasonally  waterlogged saline Brown  Soils carry  dense thickets of Suaeda monoica
(saltbush).'Alluvial seasonally 
River
Levees;
Acacia
nilotica
forest
dominates
flooded             areas,             with             very 
little             herbaceous             ground 
cover.        The        height        varies        from        8-30        m.        Inplaces,        e.g.        to        the        south 
of      the      PCC      at      Melka      Werra,      this      forest      has      been      cut      out      and      replaced 
by           scattered           A.           tortilis
trees           and           Salvadora           persica           bushland, 
with       other       shrubs       and       sparse       grass       and       sedge       ground       cover       .       Close       to the river, groves of Tamar.ix aphyalla may occur.
Gravel Fans: large parts of these fans carry  grassland  which  is more or less heavily  grazed.  Where the grass is protectedby  Acacia bushes Cenchrus,  Cynodon,  Dactyloctenium and  Panicum are common. In  S.  Kessem and  N.  Kebena the grazed  areas are characterised  by 
short
and 
sparse
Cynodon-Chrysopogon-Eragrostis
grass
which 
commonly  includes t he herb  Tr ibulus terrestr is .  Si mil ar grassland areas occur to  the SW of Dera bale,  N of Dofan,  and  on  the fans on the east side of the Alledeghi plain.  Elsewhere in  the Kessem- Kebena area the fans may  carry  Acacia shrubland,  (e.g.  SE  from Dofan  bordering  the Wadi T'Unfeta,  where Acacia nilotica
shrubland 
characterises
sodic
clay 
flats), 
Salvadora
clumped 
thicket        or        Cadaba        shrubland.        In        the        area        between        Melka        Werer        and 
the           main           road           the           grass           cover           of           the           Acacia           Senegal/Acacia
mellifera           shrubland           has           been           virtually           destroyed           by           grazing           and 
trampling.        The        absence        of        Acacia        mellifera        from        the        Kessem        area        is
notable
.            C            ad            ab            ashrubland            issometimesassociatedwithwater            holes
and       it       can       be       inferred       that       concentration       of       Afars       in       the       areas       has ledto           thedestructionof           virtually           all           the           Acacias,           which           ar           eused 
' 
for stock compound fences. 
/
Gravel          Outwash           Plains:           just           north           of          the          Kessem          river          these
Outwash           Plains            carry           extensive           areas           of           open           Acacia
Acacia
nilotica                woodland, 
interspersed                with 
short
tortilis- Cynodon 
grassland and areas of Cadaba rotundifolia bushland and shrubland. East of the Awash river the vegetation on the outwash plains has been much more heavily grazed, and large areas of bare soil occur between the Acacia bushland.
/ZUH k .11 I I _’F_ »              « U
Acid  Lava: These are characterised  by  tree and  shrub  Acacia grassland.  The hollows in  the lava surface where soil has accumulated  may  carry  almost pure short Cynodon  grassland. Converslythe lava ridges are almost pure Acacia Senegal thickets 3  —  4  m high.  Occasional evergreen  shrubs occur such  as Grewia,  but these are very  heavily  browsed  by  goats ,  and  reduced  t o  low tufts. The lava on the SW flank of Aside carries open Acacia woodland.
Basalt: large areas of the Basalt flows are devoid  of vegetation, but scattered  Acacia trees and  shrubs do  occur,  and  where pockets of soil have accumulated  there is a sparse grass and  herbaceous cover.  These areas are utilised  by  the Afar,  and  Filweha crater (just out side the project area) has a permanent Afar sett 1  erne nt in it,  a local water source supporting  a stand  of Acacia woodland. Botanical details for this land form are not available.
Lakes/Spr ings : the Kortume area carries Sporobolus grass land  ,  the spring  serving  the area,  and  the important Bilen  spring  are both surrounded  by  large Typha latifolia swamps.  Another such  swamp occurs north of the Kessem and SW from Gurmile hill.
2 . 7 Present Land Use
The area was originally  used  entirely  by  the Afar for extensive grazing; their semi-nomadic system based  on  mixed  herds of camels, cows,  sheep  and  goats being  in  virtual balance with  the grazing 
resources,           except           in           drought           years.           In           the           last
thirty           years
extensive           irrigation           has           been           developed           adjacent
to           the           rivers, 
which          has          replaced          large          areas          of          critically          important          Afar          dry season grazing.
Apart          from          the          irrigation          schemes          which          are          largely          devoted          to 
cotton             production,             although             fruit, 
vegetables
and             tobacco             are
produced        at        Awar        a        Me        lka,        there        is        virtually        no        arable        agriculture in the area.The     except     ions     are     the     small     areas     farmed     by     a     group     of     Soudan     i     who 
settled       on       the       banks       of       the       Kebena       river,       and       a       number       of       small
gardening            initiatives            undertaken            by 
individual            Afar,            notably 
in 
the     Kessem-Kebena     area     near     Yalowith     NOMADEP     ass     i     s     t     anc     e     ;     inHiddi
and        Wassero;        and        around        Melka        Werer        with        the        help        of        IAR.        In 
addition,          there          are          three          RRC          Afar          Settlement          Schemes          (1)          a
nominal      80      ha      of      maize      and      cotton      at      Doho;      (2)      a      further      2000      ha      at
Amibara       subsequently       reduced       to       1127       ha,       of       which       only       152       ha       are
currently             cultivated 
to             cotton,plus             some             small             vegetable             plots;
( 3  ) HaleD eb  ii where pumping  andcultivationconstraintslimit the area cultivated  (cotton,  maize and  horticultural crops) to  200  ha of the 4 26 ha scheme.
The Alledeghi plains contain a 2000 ha area of fenced grassland 
which is operated as an Animal Holding Centre and Quarantine
Station, (based on a borehole) which was started in 1984 under the
Ministry of State Farms. An additional 4000 ha of grassland, 
outside the fenced area are also used. Somali blackheaded sheep 
are purchased, maintained on highland hay for one month,and then 
exported to Saudi Arabia from Aseb.
Lastly  some 2  2  978  ha of the 830  km2  National Park,  (the northern flanks and  adjacent Gravel Fans of Filweha mountain) are in  the project area.  Until recently  pastoralists (Afar and  the Oromo  Etu and  Kereyo  people) were excluded  from this area; but the National
Park's           Policy           is           being           changed,           and           pastoralists           are
to           be
integrated  into  National Park  planning  and  management.  This is a
significant development with  regard  to  the Afar situation  on  the
west s ide o  f the Awash  river.  The National Park  is also  flanked  by 
th 
extensive
Awash 
West
and 
Alledeghi
Wild 
Life
Reserves
where
game
are
protected, 
but
grazing 
is
permitted. 
The
latter
extends
norxh to a line through Melka Werer.
The loss (present and  projected) of Afar grazing  land  within  the project area,  as a result of the various land  use developments outlined above is summarised in Table 2.7.Table 2 . 7 Loss of Afar wet and dry season grazing land resulting from existing and proposed developments.
WET SEASON GRAZING
Alledeghi plains National Park (N part)
DRY SEASON GRAZING Irrigation Schemes
Existing (Ha)
6 000
23 000
29 000
Existing & Proposed (Ha)
Net           Gross
Net           Gross
Melka Sadi *
4  600
5 520
4  600
5  520
Melka Werer IAR
300
360
300
360
Amibara RRC Settlement
2    000
2 400
2   000
2   400
Amibara State Farm *
3  400
4 080
3  400
4  080
Amibara Extension
1   650
1 980
1   650
1   980
(Alget a,Ambash,Sub lale)
AngelelePump Irrigation
2   150          2 580              4   235        5  082
Angelele Irrigated Pasture
1   00 0
1 200
3  000
3  600
Hale Debii RRC Settlement
426
51 1
426
5 11
Bolhamo-Gidiaro-Dofan
1   400
1 6 80
6   927
8  3 12
Bolhamo Irrigated Grazing
-
700
840
Bolhamo Fuelwood Plantation -
-
2   800
3 000
Kesem-Kebena:Awara Melka
1   400
1 680
1   400
1   680
: Yalo
6 25
750
625
7 50
.RRC settlement                         80             100                   80            100
Kesem-Kebena Extension
TOTAL
-
-
10  5 35     13   270
19   03 1   22 84 1             4 2   686      51   4 15
*           Including           the           four           sites
totalling           418           ha           earmarked           for
forestry, which if developed will displace further Afar families.From     this     it     will     be     seen     that     some     2     9     0     00     ha     of     wet     season     grazing 
has      been      lost      ;      this      however      may      be      reduced      as      the      integrated      use      of
the              National
Park 
isprogressively 
introduced.              Moreserious
isthe
loss of virtually  23  OOOha of dry  season  grazing  as a direct result of irrigation  development.  It can  be argued  that the 1000  ha Angelele irrigated  pastures houldbe excluded  ; however t he success of this scheme must remain  in  some doubt and  for this reason  it has been  included.  The full achievement of the various development proposals increases this figure to  51  500  ha,  again  this includes a further irrigated pasture. The present and projected
development of irrigation has three major consequences:
-   reduced quantities of animal feed
-   degradation of forest, woodland and trees in the grassland
-   increasing sheet and gulley erosion, and wind erosion.
In  subsequent sections of this report,  the Afar situation  is described,  the various proposals made to  deal with  the above problems are reviewed,  and  in  the light of this review proposals are made for a phased  programme of assistance for Afar development.Ill THE PEOPLE OF THE PROJECT AREA
3. 1 Afar Clans and their Territories
The project area is occupied  by  some 30  Afar clans.  There well be more,  but that cannot be determined  without extensive field surveys.  Some of the smaller clans identified  may,  on  closer examination,  turn  out to  be sub  clans.  For the purposes of this study  precise knowledge of the clans and  sub-divisions is less important than  an  understanding  of the relationships of the clans
with        one        another,and        with        their        acncestral        land.        Map        2        shows        the distribution of the clans identified in the present survey.
The clans of the project area are divided between two tribes - the
Debine and the Waima. The Debine are the larger and more widely 
dispersed tribe, with territory on both sides of the Awash river . 
According to some versions of tribal history the Debine have been 
established on their land for many centuries, while the Waima
moved in from the east some 50 years ago.
The       droughts       of       the       1970s       and       1980s       seem       to       have       accelerated       these movements of the Waima,and to have put Debine land, already
depleted
pressure.
by the expansion of state farms, under increased Tribal conflict over land rights has led to disputes and
fight ing,            wit h many fatalities, and in the Kessem/Kebena Area the Waima were forced to move eastwards out of Debine territory.
3.2    The Issa
The Issa, a people with Somali affiliations, live to the east of the Project Area. They are the traditional opponents of the Afar and recent battles have left large numbers of dead and wounded. The Afar say the Issa come heavily armed at night and steal their livestock, killing any Afar who oppose them.i:
Hap 2,- A-f*r zlans in fa pryccf* am.
CAT AU*
UAMl
P*B<TAA<
&
AW**L_^
kofeiufet
AVJM’'*
fcWArSH
A\V«^a
kSSoB
fcA<W*J
I (KiEkJi »/
|z t_____ 7 jp *<0- s7S                   k /l£k|A\ V.
C. A*s*JtKHL SiHL*M.v-T
3.  AavJxm-k     1
z*.      *      2
5.       •      3
C.  A^dJt I
H
fk'7V\
I
• A It* A
1. -            a 8. *           3
1. MtllcA 0U4J-
'/
A W^A
10 .         $A<k Ckrin l(.
I2L Sl>M-|>
13.
U^. *&oIUa*^.o
kZ_These raids come as far west as Bilen, which is at the northern 
extension of the Amibara State Farm , and the Afar in that area say 
they are afraid to venture towards the road or over it into the
Alledeghi plain to graze their livestock. This account, if true, 
would explain the relatively untouched pastures between the ridge
which bounds the flood plain to the east, and the tar road at the
northern end of the Amibara block.
Afar living  further south,  m the vicinity  of Melka Werer andMelka Sadi,  have adopted  a more positive and  agressive stance against the Issa menace .  They  combine their herds into  large groups,  with up  to  15  or 20  armed  men  to  guard  them ,  and  take them to  graze in  the hills to  the east of the Alledeghi plain.  They  may  even  use Issa wells in  that area if they  feel strong  enough  to  fight off the owners.
3.3    The Kereyu
In the Kessem/Kebena area the Afar confront the Kereyu, a well armed pastoral people who live to their south. Here also, raids, blood feuds and armed encounters over grazing and water rights are frequent.
One result of these tribal animosities is that large areas of
grazing become a pastoral no-man's-land, where no group is able to bring its animals, unless it is strong and organised enough 
to win a shoot out with the enemy, or to deter attack by appearing to have military superiority.
The necessity to mobilise sufficient numbers of armed men to protect herds and grazing territory, and if possible win some animals off the enemy, makes heavy demands on the Afar labour force. The unwillingness of the Afar to take up settled agriculture or permanent employment can be better understood in this light.Rain 
has
a
stron
g
influ
ence
on 
tribal
and 
clan 
conflict
over
land 
and 
grazin
g. 
Follo
wing 
good 
rain
s
there
is
abundant
grass
and 
browse
, 
and
the
animals
grow
fat
on 
the
home
pastures;
or,
if
taken 
to 
distant
pastures, 
the
animal
s
of
all
groups
grazing 
there
find 
enough  to  eat.  Conversely,  during  drought the home pastures are grazed  out ,  and  herdsmen  move to  where they  can  find  something  for their animals to  eat,  where they  will encounter other clans or tribes with the same intention, and fighting is t he usual outcome .
In this context the expansion of irrigated agriculture in the
Awash Valley has the same effect on pastoralists as continuous
drought: the home pastures no longer produce forage, and the
origional owners of the land are forced outwards, encountering 
others in a similar condition. Heavy concentrations of livestock 
on the remaining land aggravate the depletion of natural forage, 
and increasing levels of conflict result.
Afar clan  territories are more clearly  defined  as they  approach the Awash  River,  where flood  plain  and  riverine forests provide dry  season  grazing.  Territorial boundaries become vaguer as they recede from the river towards the higher plains and  the hills. “Thus the Alledeghi plains are generally  regarded  as open  grazing for the Afar who are willing to risk encounters with the Issa.
The grazingwithin  a clan  territory  is not used  by  members of that clan  alone.  Others,  when  short of grass,  may  also  graze there if they  have obtained  the permission  of the clan  elders,  which  is
seldom            refused            unless strained.
relations            between            the            two            clans            are
No  clan  territory,  as far as we are aw ar e ,  i s i nhab  i t ed  by  members of that clan  alone,  but one clan  is nevertheless recognised  as the owner of the area: others who  come to  live there do  so,  and  remain there,  only  with  the permission  of that clan,  and  they  may  have to leave if there is trouble their clan and their hosts.The mixing of clans is most pronounced in the areas where the
Afar have lost most land to the state farms and have been 
displaced from their origional habitats. Examples are the clans
of the Melka Merer and Melka Sadi areas, who have lost virtually 
all of their dry season grazing and their settlement sites. Map 2 
shows the concentration of clans in this area.
Some clans have within their territory a resource that is freely 
used by other Afar, regardless of their clan membership. An 
example is the Bilen hot springs, a perennial water source, 
believed to contain minerals that are beneficial to cattle and 
camels. Although Bilen is within Rakoba clan territory it is used 
regularly by several neighbouring clans.
Asked  about their territorial claims Afar often  say  "We are all Af ar .  We help  eachother.  " Nevertheless,  eachclan  has a clear idea of where its boundaries lie.  There may  be exceptions to  this generalisation  in  the case of some smaller clans who  have lost most of their own  territory,  and  have been  allowed  to  settle in  the territory of larger neighbouring clans.
The cl an  boundaries have not been  obliterated  from people’s minds by  the imposition  of State Farms or settlement schemes across the
traditional
territories,           when           the           cotton           is           harvested           each           clan 
claims       prior       right       to       graze       the       stalks       in       the       fields       within       its       own clan boundaries. On the Amibara Settlement Scheme a recent
dispute between  the Harkamela,  the clan  who  originally  owned  the area,  and  the Sidihabura,  who  had  taken  up  a block  of land  on  the Scheme,  resulted  in  15  or more deaths and  the departure of the Sidihabura from the scheme.  The Harkamela were thus left in possession  of their ancestral land,along  with  a few members of other smaller clans who did not pose any threat .
An inference to be drawn from such events is that Afar territoriality is a major factor determining where and how land is to be developed for Afar benefit.There is a fairly  wide agreement among  Afar in  the vicinity  of the Third  Livestock  Project’s 3  000  ha grazing  scheme that this will become the scene of continuous conflict among  the clans.  Afar in that area see it as compensation  for all the Afar who  have lost land to  the state farms.  It is,  however,  sited  in  the clan  territories of one or perhaps two  clans,  neither of whom has been  as severely affected by loss of land as have many of the more southerly clans.
If        all        Afar        are        to        have        access        to        the        scheme,they        say,        the irrigated       pastures       willsoon       be       grazed       down       by       the       large       number       of 1      i      vest      ock,      and      there      will      be      fighting      over      the      remnants      of      the forage.       If       only       the       clan       or       clans       whose       territory       the       scheme       is, are        to        have        grazing        rights        there,        as        the        available        information        on the          scheme          suggests,          jealousy          and          rivalry          will          undermine          the scheme        and        embitter        relations        between        the        Af'ar        and        the        Authorities yet       further.       What       can       be       said       in       favour       of       the       grazing       scheme       is that       it       does       at       least       recognise       that       the       Afar       have       some       land       rights and some land requirements in the area.
The problem is further discussed in section 4.2.3
To  thewest of the project area,  the Afar adjoin  the Galla,  who  live
in           the           hills. 
It
is           understood           that
relations           with           these
highlanders          are          generally          harmonious,          and          there          is          a          mutually useful exchange or saleof cattle and grain betweent he two groups .
3.4    The Afar population
The January  1990  ILCA survey  of the project area mapped  the
distribution          of          pastoral          dwellings.          For          practical          purposes
it          has
been  assumed  that these are occupied  by  Afar ,  although  a few in  the south  west are probably  Kereyu  dwellings.  A previous survey  in this area,  by  MMP ( 1986) ,  showed  that an  average of 5  people were associated  with  each  dwelling.  Using  this information  it is therefor possible to  estimate the Afar population  from the number of dwe11i ngs.The ILCA survey data i
persons per dwel11  ng,to  show th 
were therefor analysed using the figure of 5
r
r
river. The
Table 3.4
J
--------  ..ie population along each bank of the results are shown on Map 2 and in Table 3.4
Distribution of Afar Dwellings and Population
Location
Sect ion
Owe11i ngs
Population
r
J
J
East Bank
E 1
143
715
E 2
78
390
E 3
1 023
5 115
J
E 4
1 450
7 250
Total
2 694
13 4 70
West Bank                  W 1
3   969
19  845
W 2
3   6 76
18
380
W 3
9 11
4
55 5
W 4
2   35 7
1 1
875
Total
10   931
54
655
Area
Total
13 625
68 125
3
J
1
1
1
J
»
Source ILCA, 1990
The           only           other           reasonably           comprehensiuve           source           of           Afar           population data           is           the           records           kept           by           the           administrative           offices           of           the awraja.         These         are         based         on         the         number         of         Afar         receiving         food         aid on          the          last          occasion          this          was          distributed,          in          1988.          Awraja          and party          are          themselves          doubtful          as          to          the          accuracy          of          this          source, but          do          not          have          a          better          alternative.          The          1984          census          did          not enumerate the Afar, who in any case are averse to being counted.
There are large discrepances between  t he ILCA and  the awr a j a data, and  possible explanations for this will be given  in  the final report .
On zUs
c [4jellnx^S a.nd hurr^a^ yC»e/lu^*’^® -
rv
3Uq M IQ Sus
UKWA^J.
**
3C?(» Wi 18380 |»|»
1023k
All U*4i ZjSSS
ui
8375 Qi U <?S
3820 (*f
Hr. * <utAci
t »UC A.                iw3.5    The Non-Afar Popular ion
In  addition  to  the Afar population  there are large,  but varying,numbers of people from other parts of the country  in  the Project area,  Including  temporary  an  dperaanent staff of th  eState Farm Enterprises and the service organisations.
Infornat ionon  temporary  andperman  ent employees is aval la ole from the records kept by  each  employer .  For the purposes of calculating population  numbers it has been  assumed  here that each  permanent employee has a household  of six; that six  temporary  employees fora one household  equivalent; and  that town  populations have six persons per household  .  Town  populations are based  on  est iiates by the awraja administrators.Table 3.5 The Non-Afar Population of the Project Area
EAST BANK Location Workers Households Population
Permanent workers: MAAE
361
Melka Sadi Farm
776
361
776
2 116
4 656
Melka Werer Farm
1 276
Amibara Farm
1 420
Ginnery
80
Sub-total
3 913
1 2 76
1 420
80
3 913
7  656
8  520 480
23 4 78
Temporary Workers: (Average
full time equivalent (1) )
Sub-t ot al
7 758
Inst, of Agric. Research
84
Amibara Set. Employees
100
Town Populations:
Mel ka Sadi
-
Melka Werer
-
Arba ( Est )
-
Sub-total
184
1  293
84
100
825
526
500
2  035
7 758
504
600
4 126
3 156
3 000
11386
WEST BANK
Permanent Workers
Dofan/Bolhamo Farm
624
Temporary Workers
2 10
Sub-total
834
624
2 10
834
3 744
1 257
5 001
KESEM/KEBENA
Permanent Workers
Awera Melka Yalo Farm
372
Temporary Workers
233
Sub-total
605
372
233
605
2 232
1 400
3 632
TOTAL
13 294
8 680
51 255
( 1  ) The number of temporary  workers ranges from a maximum of 12  0  00 in  December to  a minimum of 5  200  in  January.  The number of workers here is the average monthly workforce.3.6     Population increase
There are no  data on  the rate of population  increase in  the project area.  MacDonald  (1986) used  the natural average of 2.9% per year for the State Farm workersin  the Kessem/Kebena area.  However as the rate of population  increase among  the pastoralists is usually lower than  it is amongst the sedentary  farmers,  a figure of 2.0% per year is suggested.  Applying  these factors to  the tot al Afar and non  Afar populations,  the projected  popu  lat ions in  AD 2000  and  AD 2 0 2 5 wou Id be :
1990 Populat ion
Projected AD Projected AD 2000 population 2025 populati
on
Afar
68 125
Non-Afar 51 25
5
83 044
68 2 1 7
136 242
139 404
The        figures        for        the        increase        in        the        Non-Afar        population        should        be
used            with            caution.            In            the            absence            of
further
irrigation 
development          in          theProject          area,          this          population          could          well          remain more or less static, as workers come and go from the highlands.
V ■IV LIVESTOCK
4 . 1 The Pastoral Problem
The problem of the Afar pastoralists displaced by the irrigation 
development was virtual ly d i sregarded when Tendahoe , t he f irst big 
Awash Valley irrigation development was undertaken in 1961, thus
giving rise to intense Afar mistrust; an attitude that persists. 
Since then numerous proposals have been madeto the Afar but what
has been done has not been very successful. The droughts of 1973 
and 1984 exacerbated the problem, and stimulated international
hfelp for the Afar. Attempts to help have taken three forms:
-   Settlement, based mainly on cotton production;
-   Farming on small holdings;
-   The provision of irrigated pasture.
These        are        discussed        below.        First        ,        however        the        consequences        of        the loss of dry seaon grazing should be noted.
4.1.1 The Livestock Population
The ILCA 1990  estimates of livestock  numbers in  the Project area are shown  in  Table 4.1.1,  together with  the equivalent number of tropical livestock  units (TLU) .  The latter is a means of bringing the different types of livestock  and  their feed  requirements to  a common denominator.
Table          4.1.1          Livestock          numbers          and          their          Tropical          Livestock          Unit
(TLU) equivalents in the Project area.
Number of Head
Came 1s
19 950
Cattle
46 200
Sheep and goats
1 19 950
Equines
580
Number of TLU
19 950
32 340
1 1 995
290
64 575It i s genera lly  accepted  that one TLU requires 2  .  5  Tonnes of fodder dry  matter annually  to  maintain  condition.  The Middle Awash  herd therefore requires some 161  437  tonnes of dry  matter annually. Calculat ions were therefor made based  on  the estimated  dry  matter yield  from the vegetation  of each  of the major land  form units, this is summarised in Table 4.1.2.
Table 4.1.2 Overal1 consumable dry matt er est imat e by land form zone
Zone
Area                 Yield/ha
Y ield/Zone
(ha)
(kgs)
(tonnes)
1. Black Clay (Awash)
7 935
2 000
15 870
Black Clay (Alledeghi)
23 143
300
6 924
2 . Brown Soil (Irr ig. )
20 000
2 000 by
-prod 40 000
Brown Soil (Non-Irrig)
5 127
3 000
15 381
3 . Grave 1 Fan
111 912
2 13
23 501
4 . Acid Lava
95 878
17 1
16 395
5 . Basalt
56 700
200
11340
6. Gravel Outwash plains
64 469
250
16 117
7 . River Levees
21 159
1 000 o 1
grazed21 159
8. Undifferentiated
Hill Margins
61 824
200
1 2 364
9. Lakes/Swamps
4 298
4 000
8 596
( 50% swamp)
(2149 )
TOTAL
187 625
From     this     it     will     be     seen     that     the     supp     1     y     of     187     625     tonnes     compared 
to         the         demand         of         161         437         tonnes         is         only         slightly         positive:         a
difference         of         16.2%.         In         years         of         poor         rainfall         this         position         would 
certainly         be         negative.         Furthermore         as         some         parts         of         the         area         are
inaccessible,          the          present          situation          appears          to          be          in          a          delicate
balance.          There          is          thus          serious          local          overgrazing,          and          many          Afar
reported             deteriorating decade.
livestock 
feed             conditions             over
the
lastThe 23  000  ha of dry  season  grazing  lost to  irrigation  devel op  me nt (Alluvial Black  Clay  and  Brown  Soils) would  yield  about 3  000  kgs ha of consumable dry  matter anually.  The total annual loss r e su  It i ng  from the irrigat ion  development is therefor of the order of 69  000  tonnes or about 37% of the total annual dry  matter production of the project area.
The development of 3  000  ha of irrigated  pasture at Angelele is seen  as providing  compensation  for this lost production.  While theoretically  this would  go  some way  to  replacing  the dry  matter lost ,  fora number of reasons discussed  in  section  4.2.3  ,  this is not believed to be a satisfactory solution.
4 . 2 Attempts to Solve the Livestock Problem
The three initiatives to date are discussed below.
4.2.1 Settlement
There are 3  settlement schemes in  the Project area: Amibara ( 2  000 ha ,  reduced  to  1  127  ha),  HaleDebii (426  ha) on  the East bank,  and Awara Melka (80 ha) on the West bank.
Amibara was started  in  1967  under the AVA asa 1  000  ha pilot scheme which  with  World  Bank  assistance was later expanded  to  2  000  ha. Two  hundred  and  fourteen  families were initially  involved,  and  FAO (1971) reported  an  average net return  per settler in  the 1971/72 season,  based  oncotton  cultivation,  of Eth$  2  000.  A trust fund  to ensure regular incomes,  a housing  fund,  and  a livestock  fund  to finance a 60  ha pasture development were established.  Average yields were 3  tonnes/ha of seed  cotton,  and  20% of the area was double cropped to maize. All cultivation was mechanised.What was not recorded  was that most of the work  was done by  workers from the Highlands,  and  that Afar settler participation  largely consisted  in  collecting  their share of the net profits.  It was a good  example of a top  down  solut ion  to  a problem,  and  advice by  the sociologist Voelkner ( 1  974  ) ,  that future operations should  be on a communual clan basis, was disregarded.
The       Amibara       Settlement       was       taken       over       by       the       Relief       and       Rehabi       1 
itation  Commission  (RRC) In  1980.  Cultivation  by  Highlanders continued,  each  participating  Afar family  being  paid  60  Birr per month  from the profits of the cotton  cultivation.  In  1984  960  ha were transferred  to  the State Farm,  and  Government decreed  that the Afars themselves must cultivate the remaining 1 127 ha.
A         254         ha         dairy         was         alosstarted         with         263         exotic         Holstein         cows. However        serious        skin        disease        resulted        in        the        cattle        being        sold        off fcr          2-300          Birr          each,          the          dairy          has          now          been          closed,          the alfalfa/rhodes          grass          pastures          have          been          invaded          by          Acacia          bushes, and only 60 ha of grazing remain.
The         settlers         were         organised         as         a         Farmers         Association,         and         under
this           collective           arrangement
it           was           difficult
to           recruit           enough 
settlers,  since 1984  therefore the annual area cultivated  has not exceeded  400  ha.  A more serious problem arose in  October 1989  when a quarrel developed  between  two  clans: the Harkamela and  the Sidihabura.  The settlement is on  Harkamela clan  land,  and following  the death  of a Sidihabura man,  a fight ensued,  resulting in  15  deaths,  the withdrawa 1  of all the Sidihabura from the Scheme, and  in  consequence the disputed  180  ha are no  longer cultivated. This underlines Voelkner '  s advice that settlement ,  and  indeed  any form of land  use development must respect existing  clan  rights to land .
In        1990        the        RRC        had        its        recurrent        budget        withdrawn.        The        Amibara
Settlement         only         has         funds         for         its         100         staff,         which         include         12 
tractor drivers, 5 of them Afars.What was not recorded  was that most of the work  was done by  workers from the Highlands,  and  that Afar settler participation  largely consisted  in  collecting  their share of the net profits.  It was a good  example of a top  down  solution  to  a problem,  and  advice by  the sociologist Voelkner ( 1974) ,  that future operations should  be on a communual clan basis, was disregarded.
The Amibara Settlement was taken  over by  the Relief and  Rehabil itation  Commission  (RRC) In  1980.  Cultivation  by  Highlanders continued,  each  participating  Afar family  being  paid  60  Birr per month  from the profits of the cotton  cultivation.  In  1984  960  ha were transferred  to  the State Farm,  and  Government decreed  that the Afars themselves must cultivate the remaining 1 127 ha.
A        254        ha        dairy        was        alosstarted        with        263        exotic        Holstein        cows. However        serious        skin        disease        resulted        in        the        cattle        being        sold        off for         2-300         Birr         each,         the         dairy         has         now         been         closed,         the al      f      alf      a/rhodes      grass      pastures      have      been      invaded      by      Acac      ia      bushes      , and only 60 ha of grazing remain.
The settlers were organised  as a Farmers Association,  and  under
this           collective           arrangement
it           was           difficult           to           recruit           enough 
settlers,  since 1984  therefore the annual area cultivated  has not exceeded  400  ha.  A more serious problem arose in  October 1989  when a quarrel developed  between  two  clans: the Harkamela and  the Sidihabura.  The settlement is on  Harkamela clan  land,  and following  the death  of a Sidihaburaman,  a fight ensued,  resulting in  15  deaths,  the withdrawal of all the Sidihabura from the Scheme, and  in  consequence the disputed  180  ha are no  longer cultivated. This underlines Voelkner ’ s advice that settlement ,  and  indeed  any form of land  use development must respect existing  clan  rights to land.
In 1990 the RRC had its recurrent budget withdrawn. The Amibara Settlement only has funds for its 100 staff, which include 12 tractor drivers, 5 of them Afars.It has also  been  decided  to  abandon  col lect i vei sm and  to  allocate
0.75  ha blocks of land  to  individual Afar settlers (0.5  ha
maize/vegetables and  0.25  ha cotton) .  However because of funding 
problems ,  the settlers have now to  pay  all the runnung  costs of the
mechanised  cultivation; however they  weed  and  harvest the crops
themselves. The Farmers Association has 80 
Commercial Bank of Ethiopia to support this new development.
000 Birrin the
There must however be some doubt about the long  term viability  of the present scheme.  In  add  i t i on  to  t he probl ems o  f 1  and  allocation between  clans,  and  the provision  of current machinery  costs,  a large capital sum is required  to  replace th  fleet of worn  out machinery  ( a reason  for late cultivation  and  therefor low yields).
This year only  130  ha out of 1  127ha have been  cultivated  because the machinery  is delapidated,  and  there are not enough  settlers. In  addition  there are now 450  ha of Acacia bushland,  the rest of the area being rough grazing.
Turning  to  the smaller Hale Debii Scheme,  the problems are similar.  Only  200  of the 426  ha are cultivated  due to  pump  capacity being  too  small,  and  to  an  antiquated  fleet of tractors.  The team did not collect details of the small settlement.
It has not been  possible to  conduct an  analysis of the Amibara Settlement Scheme accounts.  There is little doubt however that in addition  to  a very  high  capital investment,  both  per ha and  per
settler, 
the            scheme            has            benefitted            from            considerable            annual
subsidies.  While this expenditure can  be presented  as the social compensatory  costs to  the Afar for the loss of grazing  lands,  the inefficiency  and  lack  of equity  between  c lansassociat ed  with  the schemes would  seem to  rule out this type of solution  inthe future. Thus the proposed  2  000  ha Bolhamo  Settlement scheme must be reviewed and revised.4.2.2 Farming on Smallholdings
There have been  a number of such  initiatives,  which  for the West bank  have been  well described  in  the Kessem Irrigation  Study 
(MacDonald,  1987).  The most notable initiative was that supported by  NOMADEP,  a Freeh  NGO ,  at Saboret,  and  their assistance to  groups at Doho  and  Yalo.  Other more or less spontaneous initiatives were started  at Carre,  Riffodda,  Hiddi,  all in  the West bank.  For various reasons these initiatives were not sustained.  They  were also  undertaken  more or less in  isolation  from government organisations.
Assistance was also  given  by  Cooperation  Francaise to  the Afar at Gewane in  response to  the drought in  the early  80  ’ s.  Although  this is just outside the Project area it is mentioned  here because of its significance to the present study.
The programme was a small irrigation  development which  included pl oughingwith  camels,  and  theconstruct ion  of an  earth  ant i-f lood bund.  Tractors were not used.  Each  part ic ipat ing  f ami ly  was given a 0.5 ha plot, and progress was as fol lows :
Year                Area cultivated
Number of clans
(ha)
1984/85 
5
1985/86 
30
1986/87 
80
participat ing
2
5
12
Total expenditure over this period  was 80  000  Birr .  However in  view of the magnitude of the problem,  other donors felt the number of part ic ipant swas too  small.  A donation  of 5  million  Birr was made through  the Ethiopian  Red  Cross ,  Buildozers were used  and  not long afterwards the scheme collapsed.  ( Denis Gerard  1990,  personal communication) .4.2.3 Irrigated pasture
This was anticipated  in  the FAO/SOGREAH 196  5  report,  which  noted that it is "usually  a tough  job  to  insert a fodder crop  in  the
irrigated 
rotational
pattern, is
the
. 
unless
highly 
organised 
andparticularlydairyhusbandry business".              However agreementmightbereached. 
integrated 
inthe
report
. 
continued  "some
.withnomadicpastoralists, 
raising, agricultural
association allowing 
their        cattle        on        pastures        sown        with        forage        crops        and        irrigated        by the plantation".
Italconsult (1969) noted on theoretical grounds that irrigated 
stock breedi ng would not be profitable in the Amibara Project and 
that of the options milk, milk and meat, and meat product ion, milk 
alone was the most attractive. Several problems were recognised, 
including the lack of suitable experience, animal disease, high 
capital investment required, and the inflexibility in the face of
changing markets.
The provision  of irrigated  pasture to  compensate the Afar for lost grazing  land  was investigated  by  Lines (1972) for a short period 
(June 1972—Jan 1974) on a small 17 ha plot at Amibara.
This work  was elaborated  by  Lines and  Pattison  (1974) as part of the FAO Development of the Awash  Valley  Project .  Notwithstanding the many  problems that were identified  (discussed  below),  the
absence            of            any            other            significant
trials
information            or            of
consultation         with         the         Afar,         the         Interim         Report         on         this         Project
(FAO 1974  ) recommended,  inter alia that the 17  ha trial be run  as
a joint venture dairy  farm with  the Afars,  and  that a further 60  ha
irrigated 
pasture
be
run 
as
a              breeding/fattening/haymaking 
enterprise also in conjunction with the Afar.
The report cone luded  " the current projections for the development of irrigated  pasture in  the Valley  indicate that by  1980  a total of 6000  ha could  be established,  rising  ultimately  to  some 13  000  ha to be used by nomadic people" .However        "management        of        this        new        and        sensit        ive        technology        requires
skills           and           knowledge           not           available           at           present
to           Ethiopian 
nationals".         It         should         be         noted         that         no         steps         were         subsequently taken to provide suitable training.
Subsequently  the FAO Senior Livestock  Officer,  Knoess (1976) reported  critically  on  the lack  of progress and  on  the technical
problems
associated 
with 
irrigated 
pasture
development. 
Thereafter,          although          little          experimental          work          was          carried          out          -
trials      at      the      I      AR      werewith      exot      ics,      and      did      not      involve      the      Afar_-
each 
subsequent
consultant
report
concerned              with 
irrigation 
development in  the Middle Awash  contains proposals for irrigated pasture development.  They  are all based  on  the very  limited invest igat ion  by  Lines (op  cit) and  his suggestion  that 5  000  haof irrigated  pasture were needed  to  compensate the Afar for lost dry season grazing.
Thus Halerow’s ( 1975) report on Angalele and Bolhamo proposed a 1 300 ha irrigated pasture unit, using cross bred improved Zebu, for meat and milk product ion. The subsequent Halerow (1986) report retained this concept, but proposed the use of Boran rather than Zebu crossbred cattle.
The 1982 NEDECO report rejected the Halcrow proposals on the grounds that they had underestimated the problems inherent in the settlement of pastoralists. Instead they proposed the provision 
of 612 ha of irrigated pasture, plus a further 700 ha of tail water irrigated pasture . Dairy and beef product ion and sheep fattening were considered and discarded for goodwell argued economic and/or technical reasons. Unfortunately no mention was made of the way in which these pastures were to be utilised by the Afar.
Subsequently the Amibara Irrigation Project prepared a report on Alternative grazing for the Afar (1983) , while the RRC produced another report on Irrigated Pasture for the Afar Livestock at Amibara (1985).At the time of writing  neither of these reports have been  seen  by 
the consultants.  In  1985  WRDA reported  on  the development of 30  00 
ha of irrigated  pasture for the Afar livestock  at Angalele,  which 
i s pre sum a b  ly  the basis for the World  Bank  (1986) Project out 1  i ne
Report
for
irrigated 
pasture
development
(Third               Livestock 
Project) , Phase I of this is being implemented.
One curious aspect about the Bank  Proposal is the recognition  of someprobl ems associatedwith  such  development s ,  whilethe examp  le of similar failed  projects has been  ignored.  Thus the potentiasl clash  between  Afar clans was recognised  - the scheme is to  be the preserve of twoclans ( althoughtheywerenot identified  by  name ) ; on  the other hand  it is proposed  that a substantial percentage of the cows are to  be mated  to  Friesian  bulls by  Al ,  despite the well known  fact that such  animals are subject to  severe heat stress ,  and the example of the failed  exotic herd  at both  the IAR and  the Amibara Dairy.
The most serious omission  however is the absence of any  reference to  a system of management ,  beyond  such  unreal istic statements that the "pastoralists would  be advised  to  adopt a rotational grazing 
system on their allocated land". This problem had 
anticipated by Tafesse in 19^0 in an internal paper.
been
He noted  that the use of irrigated  pasture requires great care, the main  alternatives being  grazing  and  haymaking.  He concluded with a well reasoned case why the latter should be preferred.
The background  to  this irrigated  pasture development has been detailed  because this is the most important attempt that has been made to  compensate the Afar for their lost grazing.  But of greater importance,  and  in  spite of the lapse of 25  years since such compensation  was proposed  by  FAO,  the social and  technical problems in  using  such  pastures have still not been  adequately investigated. These problems are summarised below.Social Problem*
1            The        Afar        have        a        very        strong        territorial        sense,        as        is        evidenced by          the          fiqhtina          between          clans          at          th**          Amibara          Settlement.          The dominant           clans           with           riahte           over           the           Anqe1e1p                  irrigated           pasture area        are        probably        the        Abl**k        Haisnmale        and        the        Fakoba        clans,        who would       strongly       resent       the       use       of       the       land       by       other       clans.       On       the other       hand       th?       NiMion       was       told       by       many       Afar       that       theyall       see       this sc    he    ire    as    being    com    pen    sat    ion    for    al    1    the    Afar    who    have    lost    grazi    ng land      They      will      all      wish      to      graze      there      ,      and,      they      ruefully      a      da      it this           will           give           rise           to           endless           and           perhaps           mortal           quarrels, uncontrolled        grazinq        and        abuse        of        the        scheme.        This        is        borne        out        by the       fa]       lure       of       the       irrigated       Afar       pasture       at       Galifagie       in       t       he       Lower Awash (another Third Livestock Project initiative) .
2            ^he        Afars        are        not        accustomed        to        the        type        of        grazing        discipline that       would       be       required       to       make       such       a       scheme       a       success,       and       no attempt       has       been       made       eit       her       to       consult       them       about       this       aspect,       or *c         initiate         a         training         scheme         to         prepare         them         for         the         problems invo1ved.
The
scheme
is
within 
the
zone
that
is
not
infrequently 
invaded 
by 
the
Issa
The
concentr
ation 
of
cattle
at
full
development,
at
least     15     000     and     possibly     25     000,     head     would     be     a     great     t     empt     at     i     on to t he latter tribe.
technical Problems
The scheme is sited on black clays which are notoriously difficult to manage, and would render the pasture liable to severe poac hi ng 
(damage to the grass by the cattles* hooves). The management required would be very suphut icated, and a similar Lonrho scheme in Zambia, with very experienced expatriate management, has recently failed for that reason.5.       There       are       no       Ethiopian       3taff       with       experience       of       the       kind       of management that would be required.
6      .     The     animals     would     be     liable     to     foot     rot     ,     and     to     infestations     of
internal
parasites,            which            require            a
regular            and            expensive
prophylatic           counter-regime.           It
is           significant
that           Lines           (1972)
noted  this as a serious problem on  the 17  has of irrigated  pasture at Amibara,  and  that Afar using  the seasonally  flooded  grassland at Kortume Swamp report the same serious problem.
7       .       11       would       be       technically       feasible       to       introduce       exotic       stock       into such          a          scheme,          although          this          would          increase          the          management problem.         However,         the         economics         virtually         rule         this         out         (NEDECO
1982,  and  1986) .  The probability  that the Afar would  be able to manage these del ic at e animals is also  very  doubt ful,  and  it should also  be remembered  that the exotic stock  introduced  to  Amibara suffered heavy mortalities due to disease.
8. Hay making, the obvious alternative to maximise the use use of
available fodder , suggested by Tadesse (1982), isalsonot without
problems . Such product io non 3 000 hawould have to be mechanised 
in order to cope with quantity of dry matter production.
The amounts produced  would  be considerable,  and  the problems of
distribution 
to 
the            various            Afar            clans            would            require            the
construction       of       a       number       of       new       roads       to       enable       delivery       by       tractor and trailer of the hay bales.
The capital investment required,  and  the recurrent costs,  would  be considerable,  and  it is very  doubtful if the scheme could  become economical ly  viable,  even  if Afar participation  in  such  a complex operation  could  be secured.  Lastly,  a system of distribution  of hay within a clan, or between clans, has yet to be devised.The            conclusion            of
this            brief            review            of
irrigated            pasture
production  is that it has been  inadequately  researched; it would therefore be unwise to  proceed  with  the planned  3  000  ha of irrigated  pasture to  full development ,  unt i 1  much  more experience has been  acumulated.  Even  the Phase I 1  000  ha development will present practically insurmountable obstacles.
In  view of the recurrent funding  problem already  experienced  with this scheme,  and  the above conclusions,  it is recommended  that development of this pasture scheme be scaled  down  to  a more manageable pilot scheme of 100  ha,  and  that the rest of the 1  000 ha be sown  to  rainfed  pasture,  and  left in  a ’ land  bank’ until the social and  technical problems associated  with  this type of development have been  resolved,  through  a sustained  programme of of adaptive trials involving the Afar.
4.3      Conclusions
It
is          a          sobering          observation          that          so          many          well-intentioned          and 
expensive initiatives have resulted  in  so  little actual benf it to 
the Afar ; that thepratical trials and  investigations that should 
have been  carried  out to  provide the basis for a development
programme are conspicuous by  their absence.  In  consequence any 
Afar programme in  the Middle Awash  will be starting  virtually  from
scratch, 
apart
from              an              extremely              valuable
catalogue
of
unfortunate experiences.V FORESTRY AND TREE GROWING
5.1. THE FORESTRY PROBLEM
Before the Middle Awash  Irrigation  Schemes were developed,  the area was sparsely  populated  with  Afars who  lived  pastoral lives in harmony  with  the natural vegetation.  The Irrigation  Schemes have upset this harmony  by  substituting  Afar grazing  with  irrigated cotton.  This has concentrated  the Afar grazing,  and  created  a huge demand for fuelwood.
The natural vegetation  which  supplies fuelwood  to  both  the Afars and  the cotton  scheme labour force,  consists of Acacia forest and woodland  bordering  the river,  and  dry  bushland  on  the rising ground  out side the irrigated  area.  In  both  areas the demand exceeds the supply,  leading  to  increased  erosion  on  the dry bushland  and  the conversion  of productive forest near the river to impenetrable thorn  thicket.  In  both  areas the fodder resources are being steadily reduced.
Wood        is        used        for        both        fuel        and        construction        but        fuelwood        accounts for an estimated 97 % of the total wood used.
5.2 Energy
Sixteen households were interviewed to assess the fuelwood 
situation, (5 Afar and 11 non-Afar) , using a specially compi1ed 
and tested questionnaire. Households were selected in order to 
sample as much of the area and perceived variation as possible. 
Given the small number of households interviewed, the results are
indicative rather than being statistically valid.
All households light more than one fire every day, but household 
fires can be divided into three main categories:
“Small       fires"       are       lit       to       heat       water       to       make       coffee       in       the       morning, and at other times during the day;"Main  fires" are used  to  cook  food  such  as "njera " and  "wat". Lastly,in  the three cold  months October,  November and  December, many  households keep  fires burning  in  the evening  for "space heat ing".
5.2.1 Fuelwood
Wood  is the preferred  fuel amongst all except those few senior officials who  are supp  lied  with  electric cookers.  Most households collect fuelwood  from the nearest available source,  which  may  be up  to  2  hours walk  away,  or buy  it from merchants,  usually  Afars. It is sold  in  varying  units according  to  the buyer’s needs and pocket.  Bulk  users,  such  as bakers and  restaurants,  may  buy  a donkey  load,  90  kg,  for 6-7  Birr.  Other s wi 11  buy  a headload,  25kg, for 2  .50  Birr or still smaller bundles of split logs or branchwood at roughly 10 cents a kilo.
There is considerable variation  in  the quality  of fuelwood available.  Dry  Acacia wood,  especially  A.nilotica is considered the best firewood,  apart from rare pieces of Olea,  a hard  wood washed  down  the river from the highlands.  Of the introduced  trees on  The State Farms Eucalyptus is thought the best fuelwood  and Poplar the worst .
Although  no  live wood  is supposed  to  be cut for firewood,  little of the wood  offered  for sale is well seasoned  and  air dry.  However trees are ring  barked  to  prepare them for felling,  and  the Afar use pieces of bark  to  tan  their water carrying  skins.  As the Afar consider the natural vegetation  to  be their property  anyone who tries to  collect fuelwood  without their permission  does so  at their peri 1.
For cooking the main meal of "watt" or "njera" firewood is always preferred. For small fires the most important alternative fuel is cotton st alks , f ol lowed by char coal , electricity and occasi onal ly animal dung.
/r /X5. 2 . 2 Cotton Stalks and briquetting
Cotton  stalks are a satisfactory  substitute
for            firewood            for
making  small fires,  as they  light easily  and  give sufficient heat toboil water.  Intheory  the 10  800  ha ofc otto  n  crops could  provi de over 20  000  tons of fuel.  This would  imply  that it could  all be transported  to  the labour camps before it had  rotted,  and  burnt in the 4  months between  harvesting  and  the next growing  season  to minimise the risk of its harbouring cotton pests.
At present all State Farm Managers discourage labour from st oc kp  iling  cotton  stalks in  labour camps ,  because of the risks of
pest
infections;            and            some            Managers            forbid            its            use.            Although 
unsatisfactory          in          many          ways,          cotton          stalks          remain          widely          used          as they are free and seasonally abundant.
Much  of this resource would  be lost to  the local fuelwood  balance if plans to  process the stalks into  briquettes go  ahead.  These plans are described  as part of a feasibility  study  in  the UNDP/WorId Bank Energy Sector Management Assistanceprogram , 19 8 6.
In  this study  a sample of about 100  kgs was shipped  to  Europe and a test batchof briquettes produced  .  During  the testing  proceedure only  38% of the origional cotton  stalk  material ended  up  as briquettes,  9% being  lost as moisture and  53% as dust produced  in preparing  the stalks for briquetting  (27%) and  in  the briquetting process it sei f ( 2  6%) .  Once manu  f ac t ur ed  t he br i quet t es ar e f a i r ly fragile and  in  a test which  simulated  handling  conditions only half the briquettes survived  intact,  30% being  broken  into  small pieces and 20% being lost as dust .
At 1985  prices the project requires a capital investment of US$ 1.  15  million  anda working  capital of US$  530  000.  If the operation goes according  to  plan  all costs will be covered  by  selling  5  000 tonnes of briquettes at 114 Birr a tonne , delivered to Addis Ababa.
///However no  account is taken  of losses in  transportation  or shortfalls in  production.  Producing  5  000  tonnes of briquettes at the factory  gate would  involve removing  13  000  tonnes of cotton stalks, or about half the crop, from the State Farm fuel supplies.
Given         the         severe         shortage         of         fuelwood         resources         it         seems         unlikely that the benefits will exceed the costs.
5.2.3        Charcoal
Charcoal is not free.  It is mostly  bought from non-Afar charcoal burners who  pay  Afars to  allow them to  cut and  burn  large trees in the riverine forest.  Charcoal is then  bagged  into  sacks weighing approximately  25  kgs which  are sold  for 12-16  Birr ,  half the Addis Ababa price.  Charcoal is also  sold  in  small heaps of 300-400  gms for 15-20 cents, just sufficient to make one hot drink of coffee.
It is widely  appreciated  as the most convenient fuel for making small fires but thought extravagant for main  fires.  For those who can  afford  it,  it is the preferred  fuel for space heating  in  the winter,  as it can  be burnt in  a portable brazier to  give clean, smoke free heat.  However it is wasteful in  terms of the overall fuel wood  equat ion  ,  because in  convert i ng  wood  to  charcoal over 7  0% of the initial calorific value is lost.
5.2.4        Dung
Dung      is      regarded      as      the      fuel      of      last      resort      ,      and      is      not      widely      used at present.
5.2.5       Electricity
Electricity  is distributed  to  some labour camps and  parts of the townships.  Many  people regard  it as the ideal energy  source, provided  someone else,  such  as the State Farm,  pays for it .  This is the case in some permanent worker camps.It      may      be      a      viable      alternative      to      woodfuel      in      the      long      term,      given the surplus of hydro-electricity available.
5.2.6              Kerosene       ,       being       expensive       at       one       Birr       per       litre,       is       only used for lighting.
5. 2 . 7 Domestic Fuel Consumpt ion
The daily  consumpt ion  o  f fuel wood  dep  ends onwhat is being  cooked  . The staple diet is "njera" pancakes made of teff,  cooked  every other day,  and  "watt",  a stew of vegetables and/or meat,  cooked daily.  Wood  consumption  on  "njera" days is about 18  kgs per household,  and  8  kgs on  non  "njera" days,  making  an  average of 13 kgs.  In  addition  to  cooking,  some households heat their houses for the three cold  months October to  December increasing  the average household consumption to 15 kgs per day.
Pressure on  natural fuelwood  is greatest in  the area bounded  by  the northern  limits of the Angelele irrigated  pasture scheme in  the north,  the main  asphalt road  in  the east ,  Arba Town  in  the south  and the Western  limits of the Middle Awash  Agricultural Enterprise (not including  Gewane).  The approximate number of households in this area is 14 500 made up as follows:
4 500 permanent MAAE employees, most of whom are householders; 1 500 temporary MAAE employee "household equivalents" (being 6 
men who cook together) ;
300 households in Melka Sadi Township not employed by MAAE; 520 households in Melka Werer Township not employed by MAAE;
80 householders employed by IAR;
100 households on the Amibara Resettlement Area;
500 households in Arba Township, making a total of
8 000 permanent households. In addition there are an estimated
6 500 semi-nomadic Afar households on the East bank of the Awash, ------------ thus the total number of households in the described area isThe      number      of      Afar      households      are      derived      from      Awraj      a      information; Non-Afar households are assumed to have an average of 6 people.
At 15  kgs per household  per day,  the total demand  is equivalent to over 80  000  tons of fuelwood  per year.  In  1990  about 10  800  ha of cotton  have been  planted,  which  will produce about 20  000  tonnes of cotton  stalks.  However MAAE  dicourages the stockpiling  of cotton  stalks in  labour camps,  as they  may  harbour serious cotton pests.  Most of the balance,  the remaining  60  000  tons of fuel will therefore be removed  from the existing  natural vegetation  in  the form of wood .
5.3 Construction timber
Housing  the Afar and  non-Af ar populat ion  makes constant demands on the area’s wood  resources but they  are small by  comparison  with fuel wood.  Even  so  posts and  poles of the required  dimensions may be hard  to  find  and  command  high  prices.  While the irrigation schemes were being  developed,  bush  clearing  provided  much  of the construction  timber needed.  In  the areas proposed  for development fuelwood,  and  other demands,  have already  removed  most of this material.
Housing  takes many  forms.  Managerial houses are all built of imported  materials.  Some permanent labour and  most temporary labour are housed by the State Farms.
They  live in  dwellings made of imported  Eucalypt poles frames, roofed  with  corrugated  iron,  whose walls are made of clay plastered  over a mesh  of fine branches,  such  as cotton  stalks.  Seme permanent labour live in  terraces of two  or three roomed  units with separate kitchens,  and  improved  earth  stoves.  Temporary  labour live inb ar racks, 80 to a room, and cook outside in groups of six..1  Other       State       Farm       employees       and       residents       of       the       townships       of       Melka
.1 ] ] ] ] ] ]
1
- 1
1
Sadi,  Melka Werer and  elsewhere,  and  many  of the Afars who  have settled  in  these towns,  live in  pole framed  rectangular houses with  both  walls and  roofs made of clay.  These are made of local materials and  expected  to  last 7  to  12  years depending  on  the quality of the origional materials.
A      typical      house      might      cover      a      surface      area      of      20      m2      and      contain about       1.2       m3       of       poles,       the       most       important       and       expensive       of       which are      10      forked      posts      (costing      up      to      10      Birr      each)      which      support      the whole          structure.          Replacement          every          7          years          implies          an          annual consumption        of        0.171        m3,        a        negligable        amount        compared        with        the 11.3 m3 the household consumes in fuelwood.
Other permanent labour house themselves in  round  huts made of light poles,  both  roof and  walls being  thatched  with  grass.  These require little timber but must be replaced every 3 years.
Finally there are the traditional Afar dwellings, made of a
circular framework of light Cordia poles bent into a central
point, and covered with woven grass mat s. The whole st ructure can 
be di smant ledandtransportedby camel, when c ir cumst ances demand, 
but they only last about a year.
Many         of         the         buildings         require         long         straight         poles,         especially         the State Farm quarters, but these aredifficult to find in the natural
vegetation,  where branches tend  to  be short and  thick.  Renewal of the State Farm quarters means buying  Eucalypt poles from several hundred kilometers away.
5.4 Other wood uses
West of the Awash, at Awara Melka on the Kessem River, the State
Farm       buys       3340       m3       of       firewood       annually,       to       cure       its       tobacco       crop. This wood comes from several hundred kilometers away, at a cost ofA       small       amount       of       wood       is       used       to       build       fences       near       houses       to contain    dome    st    ic    stock,    and    the    Afar    trad    it    i    ona    1    ly    make    dry    t    horn bomas       to       contain       their       animals.       There       are       some       irrigated       hedges       of Pitheco1obiurn dulce and Poplar, on State Farm compounds.
5.5 Existing wood resources
The final report will contain  an  analysis of the satellite imagery,  the 1961  photography,  local measurements and  other relevant data from elsewhere.  This will attempt to  quantify  the existing  resource,  show how the natural resources have been depleted,  and  highlight the need  to  plant trees to  meet future needs.
The MAAE  has made considerable efforts in  the past to  grow trees, mainly  around  labour camps and  other installations.  However shade and  amenity  appear to  be the most important objectives,  little thought being  given  to  fuelwood  supply.  In  the last year or so however Eucalypts have been planted to supply building poles.
The IAR has conducted  a series of trials on  Leucaena,  Sesbania and other fodder trees.  These show that Leucaena is extremely  well
adapted         to         local         irrigated         soil         conditions,         producing         up         to         70 
tonnes         of         dry         matter         fodder         per         ha         and         carpets         of         natural regenerat ion.
On  Malka Sadi State Farm wind  breaks of Poplar and  Casuarina were planted  about 20  years ago,  and  Eucalyptus sargentii about 9  years ago.  The latter show the potential of producing  at least 20  m3  of fuelwood  per ha per year on  soils with  t he worst salinity  problems . They  would  need  irrigation  for the first two  years but thereafter should  continue to  grow,  with  their roots in  the shallow water
table,           without
irrigation.           Once           established           they           should           produce
a     firewood     crop     of     100     m3/ha     at     age     5     and     similar     amounts     as     coppice crops every 3 years thereafter.VI AN AFAR PROGRAMME
6.1 Introduction
The Government of Ethiopia has implemented  a number of programmes for the benefit of the Afar,  including  the irrigated  grazing scheme at Angelele which  is currently  being  implemented.  Despite the scale and  cost of these efforts,none have been  very successful.
Some of the more popular projects,  from the Afar point of view are clearly  unsustainable.  The grazing  scheme at Galifage,  on  the lower Awash,  for example provides free irrigated  pasture to  all who  wish  to  use it.  The Afar participate in  the scheme,  but bear none of the responsibility for its upkeep.
Of the government-supported  Afar projects studied,  the Afar were only  nominally  concerned  with  the operation,  management and maintenance of the project.  It seems unlikely  that any  would survive withdrawal of government support.
With a few notable except ions, such as the Afar farmers assisted by 
the I . A.R. , government sponsored projects for the Afar seem to 
have certain flaws.
a)        they        are        too        big:        the        scale        and        complexity        of        the        management
required 
to            keep            them            working            precludes            most            Afar            from
participation in their management.
b) They are too intractable: settlement or grazing schemes usually occupy a large block of land, which is expected to provide Afar of different clans with opportunities for farming or grazing. This arrangement has given rise to conflict of various kinds.It     is     clear     what     should     be     avo     i     ded,     and     what     should     be     incorporated in the future Afar programme. To be avoided are
-Large single blocks of land  developed  for use by  several clans, unless they  have an  established  tradition  of co-operation,  and expressly wish to work together on the designated land.
-Taking  Afar clan  land  into  government control,  and  after "improving" it ,  of f er i ng  it back  to  the Afar as a project or scheme intended for their benefit.
-Projects which  will depend  on  continual external support and assist ance,  because they  ar e not ec onom ically  or institutionally sustainable.
-Starting  any  project or programme without either the corporate agreement (on  a clan  or sub-clan  basis) or the active participation  of the Afar,  even  if this means that the initial actions are very modest and tentative.
To be incorporated into an Afar programme are the concepts:
-That Afar need  convincing  that this programme is actually  an opportunity  for them.  This cannot be done bywords but ,  initially, only  by  direct actions which  are to  the immediate benefit of the Afar ,  such  as the provision  of medical and  animal health  services. This would  be technically  and  administratively  fairly  simple.  In this way  trust could  be established  for joint act ion  wit h  the Afar on the more complex issues of land and livestock.
-That the action  should  be seen  by  the Afar as their action,  taken by  them,  on  their own  behalf,  albeit with  some assistance from the programme.  A central objective of the programme will therefore be to  enable Afar to  gain  confidence in  their ability  to  adapt to changing  circumstances and  opportunities,  and  thus to  undertake increasingly ambitious activities.Ll
r
L 1     That      the      programme      should      be      flexible      and      open      ended,      designed      to 
r
evolve       at       a       rate       and       in       a       direction       determined       primarily       by       the Afar themselves.
That      the      traditional      forms      of      Afar      social      organi      sat      ion,      the      clan, sub-clan         and         lineage,         provide         the         context         within         which         decisions are taken and implemented.
-That the physical layout of the programme - irrigated  pasture blocks,  gardens,  orchards,  settlements,  etc - conform to  the Afar’s own  conception  of clan  territories and  their boundaries, and  the rights that the various groups have within  those territories.  This will require a radically  different approach  to land  use and  physical planning  to  that usually  adopted  on  an irrigation or settlement scheme.
Afar are suspicious of the government’s motives,  yet for any programme to  succeed  it is essential to  gain  the Afars'  trust. This process could  be initiated  by  providing  them with  a range of basic services which they often asked for but seldom received.
However ,  if they  came to  believe ,  or were told,  that the programme was designed  to  secure their acceptance of further State Farm development on  their remaining  dry  season  grazing  land,  it is
highly           unlikley           that           their forthcoming.
trust           and           cooperation           would           be
It     is     a     fact     ,     however     ,     that     the     proposed     Afar     programme     is     designed 
to            achieve
this
objective,            even            though            further
irrigation 
development may  not be commissioned  for many  years.  In  this respect there is an  element of duplicity  in  the proposals: the Afar are being  offered  with  one hand  a programme to  enlist their part icipat ion  and  trust,  while t he other hand  i s prepar ing  to  take yet more of their land.
There are, nevertheless, strong arguments that Ethiopia shouldensure that its natural resources are used  to  the optimum,  in  the national interest.  Tn  the Middle Awash  this entails convert i ng the natural grazing,  which  is the basis of Afar existence,  into irrigated  crop  land.  It wi 11  be far from easy  to  persuade the Afar to  accept the necessity  for this conversion  of their land  to alternative uses,  and  to  ensure that they  are not disadvantaged when it occurs
If Afar are to  be convinced  that justice is being  done,  a period  is needed  during  which  their own  programme is established  and  begins
tc            bear
practical
results, 
so 
that
alternative            opportunities            are
opened        to        them        before        more        of        their        land        is        taken.        It        is        estimated 
*hat         a         programme         of         at         least         ten         years1              duration         is         needed         to 
establish            the            Afar
in            new            agricultural            and            animal            husbandry 
practices
proceeds.
before
further
irrigation              development
on 
their
land 
In  the following  pages a draft Afar development programme is outlined.  The ideas embodied  inthiswill be elaborated  and  costed in  the final report .  Some of the proposals may  need  to  be modified in  the course of this further work  ,  though  substantial alterations are unlikely. Indicative costs of the programme will follow.
6    2 The Amibara II Irrigation Scheme and the Afar Programme
It nas been  argued  that many  years will be required  to  establish  a sustainable Afar development programme.  Apart from needing  to estab  1  ish mutual trust with  the Afar ,  the lessons of previous Afar settlement and pasture schemes must be assimilated and applied.
The recent Awash  Valley  Master Plan  (Halerow,  1990),  has reviewed the issues connected  with  irrigation  in  the valley.  It concludes that there is an  urgent need  to  improve the efficiency  of the existing  irrigation  schemes,  and  that new schemes or extensions of
existing             schemes accompli shed.
should             be             delayed             until
this             has             been It follows that the best Interests of the State Farms and  of the
Afar will be served  by  a careful ly  phased,  long  —  term programme of
improvement and development. 
This will allow for the
rehabilitation 
of
existing 
schemes, 
leading 
to 
greater
efficiency,  and  it will accomodate the proposed  Afar programme .  In both  cases ,  each  step  forward  must be consolidated  before the next step is taken .
6.3    An Afar Programme
This programme draws heavily  on  the positive and  on  the negative
experience
of
previous                 Afar
development
and 
settlement
programmes,            endeavouring            to            learn            the consists of five parts:
1 .     The Advance Programme
2.      Livestock and Agricultural Programme
3.      Forestry and Tree Planting Programme
4.      Training Programme
5.     Institutional Programme
lessons            of            each. 
It
It
should            be            noted            that            there
is            no            separate            conservation 
programme, 
since
conservation 
has
been 
incorporated,                where
appropriate, into each of the programmes.
The programme components have a number of features in common. They 
are all conceived from the 1
 bottom up   . In each case, that is, the
1
programme would start with full discussions with the Afar and with other institutions invlolved (e.g. State Farm Enterprises, 
R.R.C.) in order to establish the needs of those concerned , the problems to be resolved, and the mutually agreed course of action and allocation of responsibility.hi iir- in i "un 'ih m
, I I rt X I. -I >
• J ••
In          practice,          several          aspects          of          the          programme          might          be          discussed simultaneously.
The clan  is a central Afar institution,  whose active support is required  before any  development is initiated.  The basic economic unit is,  however the household,  or extended  family,  and  it is to this level that development actions must be directed.
One of the most fundamental problems to  be addressed  is the shortage of fodder cause by  the conversion  of grazing  into  coton fields.  Proposals to  assist Afar to  redress this deficiency  are a feature of several of the programme components.
The entire programme is designed  to  enable the Afar to  diversify their economy  and  to  help  their integration,  at their own  pace, into  the national economy.  This process will assist them to  adapt
to           the           impact           of           future           irrigation           development
in           the           Middle
Awash,           and           to           develop           a           more           robust           means           of           supporting 
themselves, 
pastoral ism.
no 
longer
dependent
solely 
on 
transhumant
6.4     The Advance Programme
This           comprises           five           elements,           which           Afar area told the team are high priorities:
throughout the project
-         Provision        of        a        travelling        rural        health        unit,        on        each        bank        of the Awash River, to serve remote Afar communities;
-   Establishment of 4 to 6 Drimarv communit ies ;
o       pr       imary       schools
+-
to       serve       remote       Afar
-         Development         of         domestic         water is distant or of low quality;
supplies i n ar eas where wat er
-           Assistance           to           Afar schemes, whereR.R.c. s
farmers
on 
government
settlement
upport has been withdrawn.‘ V - Ni • M III '• 
!H>      .1 I U IW 1 ir / I •
VJ
This programme should  be started  as soon  as possible.  It is relatively  simple to  set up  and  implement; it meets immediate and urgent needs of the Afar; it is non-controversia1; and  its benefits will be felt immediately.  It should  help  to  secure the conf i dance of the Afar in  the larger programme,  and  will provide a context for the growth of trust between programme staff and Afar.
The advance programme could be implemented by an NGO. 6.5     The Livestock and Agricultural Programme
This           includes           ten           specific blocks of land within the territory of each clan.
6.5.1 Angelele Irrigated Pasture:
initiatives, 
to           be           implemented           on 
For reasons already  explained  (para 4.2.3) it is believed  that this initiative should  be limited  to  a trial block  of 100  ha in  the first instance and  that the balance of the cleared  land  should,  for the time being,  be sown  to  dry  land  pasture,  preferably  including a legume.
This         would         enable         critically         important         management         problems         to         be
investigated 
-
including              appropriate              pasture/legume              mixes
for
the      Alluvial      Black      Clay      soils,      the      method      of      harvesting      i      nc      lud      i      ng 
cut          and          carry          green          fodder          versus          haymaking,          and          of          critical
importance, part icipat ion.
the             most
appropriate             way             of
ensuring             Afar
The results of this trial work, which would need to continue for several years, would enable a rational decision about the future of t he scheme .6.5.2          Small          (max          100          ha)          irrigated          pastures          and          individual          farm units on the edge of but within State Farms :
The idea would  be to  develop  a series of clan  based  enterprises, each  unit of which  (say  1  ha) being  vested  in  an  individual or preferably a family of the clan.
The excision  of small blocks of developed  irrigated  land,  from existing  State Farms,  would  enable this operation  to  be started without delay.  The schemes would  draw on  the experience of IAR support to  individual Afars,  and  in  the earlier NORMADEP work. This initiative would  include a training  element,  linked  to  the proposed training institute.
6.           5.3        Small        irrigated        pasture/farm        units        out        side        t        he        State        Farm schemes.
These would  be very  similar to  the units described  above ,  but they would  be outside the State Farm Boundaries.  It is envisaged  that they would draw their irrigation water from the State Farm system.
6. 5.4 Improved management of the RRC settlement schemes:
This       is       a       big       and       important       issue,       and       at       the       time       of       writing       it       is uncertain what is to be done with the RRC irrigated schemes.
As a matter o  f pr i nc ip  le t hi s land  which  has been  dedicated  to  Afar use should  be retained  for this purpose.  However as noted  above there are severe management problems with  such  schemes.  In 
addition 
the            management            required            tor            such            large            blocks            of
landprobably precludes the whole 1 individual holdings.
127 ha being turned IntoIn      vi      ?w      of      the      WRDA      proposal      to      redesign      the      irrigation      system      for the    scheme,    which    w    i    I    1    require    a    cess    to    and    agreement    on    a    land    use plan       (which       as       far       as       is       known       has       not       been       produced),       it       is envisaged        that        the        first        stage        of        this        initiative        would        be        the production       of       such       a       land       use       plan.       Tills       might       well       conclude       that the      1      127      ha      should      be      treated      as      a      land      bank,      to      be      eventually handed       over       to       the       State       Farms       for       cotton       production,       against       the provisionof     1     00     ha     blocks     within     the     St     ate     Farms     .     There     would     t     hus be an exchange of land on an equitable basis.
It is very  strongly  felt that Afar "rights" to  the 2  000  ha of irrigated  land  that were set aside for their use should  be respected.  Thus in  due course,  as a result of the settlement study,  new ways of using  this land,  probably  as a series of small discrete blocks,  would  be determined.  This work  will be of the greatest importance in  connection  with  the proposed  Bolhamo- Angelele development.
6.      5.      5      Crop      diversification      on      the      State      Farms      to      produce      crop      by products for animal feed:
Double cropping  has been  included  in  most proposals for irrigation development i n  t he Mi ddle Awash  ( e .  g  .  MacDonald  ,  1986).  There are
however             severe             organisational
and             equipment
constraints
that
prevent         this         being         realised.         To         resolve         this         considerable         problem a consultancy is proposed.
It can  be anticipated  that the solutions that will be proposed, which  will probably  include the introduction  of considerable quantities of new equipment,  will be expensive.  Such  a study  is seen  as a necessary  par t of t he Hale row ( 1990  ) proposal to  improve
the             efficiency             of             existing 
irrigation,             which             could             have
considerable benefit for the Afar livestock.  The introduction  of double cropping,  it this should  prove feasible,  would  follow a carefully designed series of trials.6.5.6               T      he      development      of      a      cotton      oil      extract      ion      plant      at      Melka Werer:
At present all the cotton  seed  (except that kept for next year s sowing) is processed  in  Addis Ababa,  and  the resultantant seed cotton  cake is sold  to  animal enterprises in  the highlands .  Thus a source of animal feed  produced  in  the Middle Awash  is not avai lable to the Afar cattle in the area.
It        is        therefor        proposed        that        a        cotton        seed        cake        feeding        trial        be
initiated             with            selected            Afar. 
If
this
trial
establishes
the
acceptability  of this food  source the next step  would  be to commission  a study  into  the feasibility  of establishing  an appropriate plant at Melka Werer.  A plant would  be constructed only if the study were positive.
6.5.7                   An         integrated         water         supply,         fodder         and         range         conservation programme for livestock:
In  many  cases these initiatives would  be linked,  and  would  i nc lude the provision  of water points linked  to  the irrigation  system,s peripheral canals and  drains,  aimed  at dispersing  the present concentration  of animals at certain  points.  This would  be a first step in reducing the local pressure on the range.
Various
soil
conservation              measures, 
including 
the
construction 
of stone lines,  are also  envisaged.  These have been  successfully introduced  in  Burkina Faso  to  limit sheet erosion  in  an environment similar to  the Awash  one.  In  addition  the accumulation of soil moisture associated  with  the stonelines should  enable grass or browse to  be reestablished  although  the problem of controlling  exp  lo  i t at i on  woul d  remain.  This technique would  seem to  be appropriate for the land  bordering  the access road  to  Melka Werer .nr -v i lu'^n^pr ib r i in .m a , j t.
It is also  envisaged  that it might prove possible to  develop  simple diversion  structures in  some of the wadis west of the river,  to enable spate flooding to water local areas of pasture or browse.
6.5.8        Ailedeghi Plains Programme:
The improved  use of these plains was included  in  early  plans for the development of the Middle Awash,  as one means of compensating the Afar for lost dry  season  grazing,  and  production  bores were drilled  to  improve the water production  in  this dry  area.  This has not been  successful ,  in  part because of the Afar - I ssa feuding.  In addition  a large block  of land  has become a sheep  holding  ground. A programme of work,  including  an  initiative to  try  to  secure some degree of cooperation  between  the Afar and  the Issa is therefor proposed.  It is reported  that improvements were achieved  in  the
1970 ' s as a result of discussions between the senior Afar and Issa.
If greater use could  be made of these plains,  and  if,  as envisaged in  the programme,  the quality  of the grazing  could  be improved, this would be a considerable step in relieving the Afar problem.
6.5.9         Animal Requisites Programme
There is a widespread  requirement among  most Afar for simple help with  basic animal problems ,  of which  liver fluke may  be the most serious.  There is a small government initiative,  and  French Technical Cooperation  provides help  on  a repayment basis with rinderpest vaccinations.
As many  Afar are outside these schemes,  it is hoped  to  establish  a corps of unpaid  Afar field  auxiliaries who  would  do  the work  and supply  drugs.  These drugs would  be paid  for,  the proceeds being used  to  establish  a revolving  fund.  Such  initiatives have been very successful in the Sahel region in West Africa.- qij: im ui. if u» . 11; u i m f >' in, i. i :                                                                               >■
6.5.10          Livestock Marketing
Although  most Afar keep  the minimum number of livestock  to  cater for their needs,  and  few animals animals are sold,  Afar do  trade animals from time to  time,  e.g.  in  order to  purchase grain.  The size of this operation  is not known.  While it is unlikely  that marketing  of livestock  will be a constraint to  development in  the near future,  it can  be ant i c i pat ed  t hat i mproved  market ing  wi 11  be required  in  years to  come.  It is proposed  therefore to  i nvest i gat e the present situation,  and  to  seek  ways to  improve existing 
private            sector            marketing            arrangements
if
this            seems            appropriate. 
Intervention in the mar ke t i ng c ha in by the state is not envi saged .
6.6      The Forestry and Tree Planting Programme
Seven initiatives under this part of the programme are proposed. 
To date the land earmarked for forestry within the irrigation 
schemes is Afar land, which is currently heavily settled. It
would not make an auspicious start to an afforestation programme
if these Afar were removed from this land, particularly as it is
hoped to gain the cooperation of these Afar. It is therefore
proposed that they be left where they are.
6.6.1            Development            of            Afar            owned            tree            plots            for            firewood pr oduc t i on.
Afars supply  a good  deal of the firewood  which  is used  within  the State Farm Settlements.  In  doing  so  they  are depleting  their own 
riverain              Acacia
nilotica
forests,              with              damaging 
environmental
results. 
It
is
therefore            proposed            by            means            of            appropriate
consultation  and  use of demonstrations,  to  encourage a number of Afar to  establish  individually  owned  tree plots which  would  be associated with the lOOha pasture—farming initiatives.Wanting of Roadside Verges within Irrigated Areas with eucaena for Browse and Firewood Production.
of
when 
the
al
roadways
within 
been 
the made
irrigation 
schemes
are
verywide, 
lowan
ce
has
for
a
ccess
by 
dragline
to 
ma
intain 
o 
cana
l, 
and 
for
land 
on 
which 
to 
dump 
the
sediment, 
there
is
still
suffi
cient
lan
d 
to 
accomo
date
severa
l
row
s
of
trees
. 
It
is
therefore proposed  that these road  verges be converted  to  Leucaena plantations,  which  would  be used  to  provide browse for Afar,  and firewood  for Afar and  the State Farm employees.  Leucaena grows exceptionally  well on  such  sites.  The Leucaena belts would  also offer some protection  fromwi nd  erosion,  aswell as amenity  shade . The plots would  have to  be fenced  initially  to  prevent livestock damage, the fences however could be reused.
6.6.3 Tree Planting on the Eastern Boundry of the Irrigation 
Schemes.
It is envisaged  that a tree belt could  be established  along  this boundary  to  supp  ly  firewood  and/or building  poles.  The belt would be irrigated  from the State Farm peripheral canals and  drains. Eucalypt species would  probably  be grown.  Fencing  would  probably be needed  in  order to  achieve establishment.  The question  of management and  the distribution  of the benefits would  be a matter to  be negotiated  with  the State Farms and  the appropriate Afar
clan chiefs.
6 6 4 Tree Planting on Abandoned Saline Areas within State Farm
Schemes.
It is understood  that some of the blocks may  be completely abandoned; in  addition  there are small irregular areas within  the schemes that it is not really  economic to  cultivate because of the difficulty of manoevering farm machinery on them.It is proposed  that such  areas be identified  in  col laboration  with 
the State Farm Management and  the Amibara Drainage Project,  and 
subject to  their agreement,  these areas would  be planted  up  with 
salt
tolerant           species,e.g.           Eucalyptus           sargentii
to           provide           fuel
wood        and        poles.        Harvesting        and        sales        proceedures        would        also        be agreed with the State Farm before work commenced.
6.6     .5 Tree Planting around State Farm Worker’s Villages.
Si mi lar proceedures to  those outlined  above regardingthe release of State Farm land,  as well as the management and  utilization  of the trees would  be held  with  the State Farm management.  The idea
underlying          this          initiative          is
to          provide          each          worker’s          village
with  its own  supply  of firewood  near the village itself,  and thereby  to  reduce the deforestation  of the country  around  the schemes that is proceeding apace at the moment.
6.6.6            Planting of Fruit Trees and Browse around State Farm Villages and Afar Compounds.
The absence of both fruit trees and browse is a notable feature of the area, although a start has been made recently with some planting. It is therefore proposed that a dialogue be opened with State Farm managers and clan leaders about this possibility. It is envisaged that a series
of demonstration plots would be established at the same time to enable the acceptance of such a plan to be gained by discussion and demonstration of the benefits. Date palmes would be included in this programme, if trials showed them to be suitable to this area; there is in fact a fear that it may be too humid.6.6.7        Establishment of Shelter Belts.
At      certain      times      of      the      year      wind      erosion      is      a      serious      problem,      and visibility      m      ayb      e      reduced      to      less      than      50      meters,      because      of      wind blown        soil.        The        material        transported        is        the        finest        part        of        the soil:        particles        of        clay        aand        silt        which        are        the        "heart"        of        the soil.        Where        wind        erosion        is        a        constant        problem,        all        the        fine material         is         eventually         blown         away,         and         all         that         remains         is unfertile sand .
This problem can  be dramitically  reduced  by  planting  shelter belts of trees at right angles to  the prevailing  wind,  which  protect the soil down  wind  from the belt .  It is therefore proposed  that atrial shelter belt programme be initiated  to  see whether this technique could  be fairly  widely  introduced  in  the area.  Such  programmes have successful1ly  been  established  in  Northern  Nigeria and  in Mali.  The establishment of these belts is expensive because fencing  and/or guards are needed  in  order to  get the trees established.  Once this is done however,  and  the trees have grown out of reach  of livestock,  the fencing  can  be removed.  It would  be important to  ensure that all partiesconcerned  were fully  involved in  the enterprise from the begining,  and  that theybenefitted  from the outcome; browse and/or firewood  can  be harvested  from such belts.
6. 7 Training Programme.
There        is        a        training        element        in        many        of        the        initiaves        outlined above,        and        it        is        felt        that        the        whole        training        exercise        needs        a tangible         focus.         There         is         already         an         Afar         and         pastoralist         training centre        at        Giwane,        but        its        future        is        uncertain        because        of        flooding problems      at      the      site.      It      is      understood      that      the      centre      may      be      moved, and        that        it        could        be        relocated        in        the        Melka        Werer        area.It        is therefore           proposed           that           these           questions           be           the           subject           of discussions        with        the        Ministry        of        Agriculture        and        the        other        parties concerned.Wherever the centre or centres are finally  established,  a series of training  courses are envisaged  to  provide general support to the Afar programme.  The courses would  not however be limited  to Afar ,  and  those for example concerned  with  tree planting  would  be open  to  all: Afar,  State Farm workers,  managers,  anyone in  fact who wished to participate in the tree planting programme.
It would  also  be important to  keep  the training  closely  related  to 
problemsat
fieldlevel. 
Monitoring 
andevaluationwouldtherefore
be done by  the teachi ng  staff as part of their normal duties. Provision  to  enable this to  be done would  therefore have to  be made .
It is also  envisaged  that the training  would  include visits to  see how problems similar to  those in  the Middle Awash  had  been  dealt with elsewhere.
6.8 Institional Aspects.
ADart from the need  to  set up  some form of management for the envisaged  programme,  there a number of institutional'quest ions that need  to  be adressed.  These include the need  to  strengthen 
coordinationbetweentheAfar
clans, 
and 
the
above              mentionedAfar-
Issa problem.  There wi11  also  be inst itut ionalproblems associated 
with  the various initiatives,  about which  it will be important to 
retain  an  open  mind.  As Ethiopian  socoety  moves from collective to 
individual
responsibi1ty 
for
production, 
so              will
institutions
have       to       change.       Provision       for       such       change       will       have       to       be       built
into           any           institutional programme.
initiatives
that
result           from           the           AfarThe way  in  which  this broad  .1.0  year programme could  be implemented will require very  careful consideration.  At this stage it is envisaged  that there would  be a need  for a Programme Coordinator, who  would  work  in  tandem with  the Afar Programme Director.  These two  appointments would  be of critical importance,  as the wrong person  in  either of these key  posts could  seriously  jeopardise the whole programme.
The programme would  also  require strong  technical and  logistic support.  It is envisaged  that this could  come from a major aid agency, or from a reputable firm of consultants.
A           considerable           role           is           also           envisaged           for           non           governmental organisations.     Such     or     gani     sat     i     oris     have     f     lexi     bi     1     i     ty     ,     sympathyfor, and         an         understanding         of         rural         people         that         is         often         lacking         in 
larger
organisations.             These             qualities             will
be             important
in 
securing  Afar cooperation  in  the early  stages of the programme .  It is hoped  that volunteers would  live in  selected  centres in  the Afar areas, in order to run the Advance Programme.
It will be essential that the majority of the staff concerned with 
the programme are able to speak a reasonable amount of Afar. 
Language courses will therefore be included in the mobilisation 
phase of project operations.
Finally  the Afar programme is seen  as a joint venture which  will involve many  agencies.  It will therefore be important that everyone concerned  with  it understands what the objectives are, and  why  par t i cul ar init iat ives are being  undertaken  .  Only  in  this way will it be possible to prepare for the sustainable developmentDate Due
IIILE
~ CALL NO.
O
EDITION VOLUME CAcc. NO

< <

Summery

Summary of Amibara Irrigation Project II Report

Amibara Irrigation Project II - Pastoralist & Forestry Development Studies

Overview

Interim report by MacDonald Agricultural Services (August 1990) examining the impacts of irrigation development on Afar pastoralists in Ethiopia's Middle Awash Valley.

Key Issues

Land Use Changes: Irrigation development has displaced critical Afar grazing lands (23,000 ha lost)
Environmental Impacts: Deforestation (80,000 tons/year fuelwood demand), soil erosion, and ecological degradation
Social Consequences: Displacement of Afar settlements, increased clan conflicts, and disruption of traditional pastoral systems
Population Pressure: 68,125 Afar and 51,255 non-Afar (mostly state farm workers) in project area

Failed Solutions

Settlement schemes (Amibara, Hale Debii) - plagued by management issues and clan conflicts
Smallholder farming initiatives - lacked sustainability and government support
Irrigated pasture projects - technical and social implementation challenges

Proposed 10-Year Programme

Component	Key Elements
Advance Programme	Health services, education, water supplies, interim settlement support
Livestock & Agriculture	Small irrigated pasture units (100ha), crop diversification, water supply improvements
Forestry Programme	Afar-owned tree plots, roadside planting, shelter belts, saline land rehabilitation
Training	Pastoralist training center, field-based courses, cross-visits
Institutional	Clan-based coordination, conflict resolution mechanisms

Critical Considerations

Need to respect Afar clan territories and traditional land rights
Importance of building Afar trust through immediate benefits
Requirement for small-scale, clan-specific interventions rather than large schemes
Integration with planned irrigation efficiency improvements

Conclusion

The report advocates for a phased, participatory approach to address the complex socio-environmental challenges arising from irrigation development, emphasizing the need to balance national development goals with Afar pastoralist livelihoods.