Amibara Project Design Report Summary

IMPERIAL ETHIOPIAN GOVERNMENT
AWASH VALLEY AUTHORITY
• C'- X
AMIBARA PROJECT
DESIGN REPORT
SEPTEMBER, 1973
SIR WILLIAM HALCROW  & PARTNERS
P. O. Box 2306
ADDIS ABABAGENERAL                         SECTION 1 INTRODUCTION
SECTION     2       REVIEW OF AVAILABLE DATA
DESIGN
SECTION 3 DESCRIPTION OF PROPOSED PROJECT SECTION 4 DESIGN ASSUMPTIONS AND CRITERIA SECTION 5 DESIGN PROGRAMME
IMPLEMENTATION SECTION     6      USE OF LOCAL LABOUR
SECTION
7
METHOD OF IMPLEMENTATION
SECTION
8
IMPLEMENTATION PROGRAMME
ZiCTION
SECTION
9
DECISIONS REQUIRED FROM AVA
APPENDICES
APPENDIX
A
EXISTING DEVELOPMENTS IN THE AMIBARA PROJECT AREA BY ABSTRACTION OF WATER
APPENDIX
APPENDIX
B
C
ONE OR TWO DIVERSION STRUCTURES
EFFECTS ON AMIBARA DESIGN OF DEVELOPMENTS
AT ANGELELE AND BOLHAMO
APPENDIX
D
DESIGN OF WESTERN DYKE
APPENDIX
E
POSSIBLE INTERPRETATION OF GROUND WATER
SITUATION
APPENDIX
F
TYPES OF CONTRACT
APPENDIX
G
CONDITIONS OF CONTRACT FOR MINOR WORKS
TZJBLES
TZxBLE
1
CONSUMPTIVE USE REQUIREMENTS
TABLE
2
SECONDARY CZJIaL DUHTES AND DISCHARGES
TABLE
3
SCHEDULE OF PRIMARY CANAL OUTLETS AND
DISCHARGES
TABLE          4 FLOOD MAGNITUDES ZJJD RETURN PERIODSFIGURES
FIGURE       1      SCHEMATIC LAYOUT OF AMIBaRA PROJECT FIGURE       2       FLOOD FREQUENCY ANALYSIS
FIGURE       2A    PEAK RAINFALL INTENSITIES AT MELKA WARER FIGURE       3      ECONOMIC ANALYSIS OF DYKE HEIGHT
FIGURE       4       DESIGN PROGRAMME
FIGURE 5 IMPLEMENTATION PROGRAMME FIGURE 6 EXPATRIATE STAFF PROGRAMME
4, 5
•
DRAWINGS
(in envelope)
Figures Drawings
DR.l
and 6 are in envelope containing Design Report
SCHEMATIC PROJECT LAYOUT
DR.2                  PROPOSED REGIONAL LAYOUT
DR.3                  PROPOSED PROJECT LAYOUT
DR.4                  TYPICAL CONCESSION FARM LAYOUT
DR.5                  MELKA SADI BARRAGE AND HEADWORKS
DR. 6                  MELKA WARER BARRAGE AND HEADWORKS
DR.7                   DYKES, DRAINS, ROADS AND BRIDGES
DR.6                  MAIN AND PRIMARY CANAL SECTIONS AND STRUCTURES DR.9                  SECONDARY CANAL SECTIONS AND STRUCTURES
DR.10                SETTLEMENT HOUSING.SECTION 1 INTRODUCTION
1.1          PURPOSE AND SCOPE OF DESIGN REPORT
1.2          ROLE OF AMIBARA IN AWASH VALLEY
1.3  EFFECT OF EXISING AND FUTURE DEVELOPMENTS1. INTRODUCTION
1.1 Purpose, and scope, cf Design Report
Halcrows are called upon, in their consultancy agreement, to submit a Design Report before starting detailed design. Though the purpose cf this is to 
reach early agreement on design assumptions and criteria, it has been thought wise to widen the scope
of the Report to include many aspects of the implementa tion since these have a bearing on the design.
Thus the Report starts, after the Introduction, with a section on Design wherein the proposed project
is described and the design assumptions outlined. This
is followed by a section on various aspects of the Implementation, wherein is described the proposed method of implementation and the types of contract foreseen. The Report i6 accompanied by layout drawings and preliminary sketch-designs.
It is hoped that, by early consideration cf these proposals, agreement can be reached both on the approach and on the methods, sc that implementation of the project can be started immediately.
It should however be appreciated that the design aspects have not yet been studied in detail. In some
cases, further investigations and/or data are required to confirm assumptions. While it is envisaged, that discussions with AVA will continue as the design is developed, approval and comments on the proposals outlined below will enable Halcrows to proceed without delay.- 2 -
1*2 Role of Amibara in Awash Valley
1*21 General description:
Irrigation development in the Awash Valley extends from the sugar estates at Wonji/Shoa, some 15 km. downstream of Kaka Dam, to the cotton estates in the Lower Awash, 500 km. down- stream* In the Upper Awash Valley, between Koka Dara and Awash Station there are about 19,000 ha* under irrigation*
In the Middle Awash Valley, between Awash Station and the Gewani Swamp, irrigation development commences at Melka Sadi on the right bank and at Kessem/Kebena and Aware Melka on the left bank* At Melka Sadi, where some 5>000 ha* arc being developed by MAESCO, the Amibara Project begins, (see Drawing DR* 1)* 
Its distance from Koka Dam is about 220 km*
At Bolhamo, on the left bank of the Awash River opposite the 10,000 ha* of the Amibara Project, there are 9^000 ha. of potentially irrigable land on which a feasibility study will
be started in November 1973* This feasibility study will also include 4,750 ha* at Angelele, situated on the right bank irrme- diately downstream of Amibara* Development of these two areas is planned to follow as soon as the viable area for irrigation 
has been confirmed*
On the left bank, opposite Angelole, there is a further
area of some 3,000 ha* whose lands bolong to the same morpholo gical group as those at Angelele, and to which the feasibility study may be extended if the soils are shown to be suitable*- 3 -
About 50 km. downstream of Angclele, immediately upstream of the Gewani Swamp on the left bank, is the Maro Gala area, where 23,000 ha. 
of potentially irrigable land have been identified.
Downstream of the Gewani Swamp, in the Lower Awash Vallc-y ithsce
t
arc about 30,000 ha. of irrigated cotton.
1.22 Water Availability
While it is estimated that, under the present hydrological regime
in the Awash River, sufficient water will be available for the 5,000
ha. at Melka Sadi (of which about 1,400 ha. arc now developed) and the 10,000 ha. at Amibara, a detailed analysis of available water, both 
with and without the anticipated modification in the operation of Koka
Dam to suit irrigation requirements, will be required to determine the
water available for irrigation at Bolhamo and Angclele, taking into 
account the proposed developments in tie Upper Awaeh V&lloy and fturthor developments in the Middle and Lower Awash Valley,
Should irrigation water for Bolhamo and Angelele be supplied from the Amibara supply system, as proposed by Italconsult, then the results
of this analysis of available water arc required to be known before the
design of the Amibara Project can be completed. The feasibility study 
of Bolhamo and Angclele i6 planned to commence in November 1973 and an Interim Report, containing these results, will be submitted in April
1974.
1»23 Regional Planning
Because of the extent of the full proposed development from Melka Sadi to Angelele, of which the Amibara Project is the central and key area, the layout and design must include at least outline plans for the- 4 -
infrastructure      for      this      development,      as      these      will      affect      the      layout      of
the     project.     (Drawing     DR.     2)
1.24 Existing use of land at Amibara
There are five existing development within the area to be developed for the Amibara Project, which are described in Appendix A. They are the Melka Warer Research Station, the AVA settlement farm and three large concessions. Most of the remainder of the land is occupied and used by 
the Afars as grazing area for their domestic animals.
1.3     Effect of Existing and Future Developments on the Design of the Amibara Project
1.31 Existing Developments:
In planning and designing the final project layout, the layouts of present development will be retained as far as is consistent with good engineering. However, some alterations will be necessary, mainly resulting from converting the supply of water, now pumped from the Awash River, to gravity feed and to the introduction of drains (which have not been considered necessary so far). Such alterations will be kept to a minimum and their introduction will
be kept to a minimum and their introduction will be palnned and phased to cause the minimum interference to farming operations.• 1*32 Future Developments
The potential future developments, for which allowance must
be made in the design of the Amibara Project, are those at Angelele
and Bolhamo.
Angelale is, in effect, a continuation of Amibara to the north* 
Italconsult proposed that the Amibara primarly canal be extended to 
supply water to Angelele; this 6ecms the most practical method* They 
also proposed that Bolhamo be supplied, from the Amibara primary canal, 
the water being carried via a syphon under the Awash River* It is con
sidered that a critical appraisal of t.iis proposal is necessary*
The main effeots on the Amibara design of development at Angelelc and Bolhamo are considered below,SBCTION 2             REVIEW OF AVAILABLE
DATA
2.1              GENERAL
2.2               SOGRBAH
2.3              ITALCCNSULT
2.4  UNDP/SF PROJECT
2.5              IBRD2. REVIEW OF AVAILAEtB DATA
2.1 General
The four main previous studies ano investigations
covering the project area are those carried cut by 
SCGkijAH, ITALCCNSULT, UNDP/SF PROJECT and IERD/IJA. 
These are reviewed briefly below. Mere detailed comment
is made in the sections on the respective design para-
meters.       All       data       obtained       by       AVA       have       also       been       studied and utilised,
2.2        Sogreah (1965)
The purpose of this study was to appraise ’’the resources of the (Awash) basin in land and water, and their potential development for these purposes." It therefore covered the whole Awash Valley and identified the project area a6 one with development potential. 
A schematic irrigation layout and approximate estimates of cost were prepared. The report considered that th" irrigation water for Amibara and Angelele could be suppli ed either by an extension of the Melka Sadi canal system or by a barrage at A.elka Warer. It concluded that the* latter would be the more economical but anticipated 
that 6ome interference of drainage areas upstream would occur, though the extent was not considered "undesirable."
Details of the irrigation layout were not discussed 
by SCGKBAH but the need for extensive dykes to protect the lower parts of the area against flooding was
emphasised•These have proved invaluable as a basis on which tc asses flood flows and water availability. The UNDP/SF Project have updated the data on the latter in their comprehensive Informal Technical Report No. 11.
2>3 I talconsult (1969 and 1971)
This was the technical and economic feasibility Study of the A elka Sadi - Amibara irrigation area, which 
r
concltuied that the area.was “fully feasible”. The report presented detailed proposals for project development, 
which have formed the basis for the decision to implement the Amibara Irrigation Project. All data and conclusions
in the report relevant to the layout and design of
the project have therefore been carefully considered 
in preparing our proposals.
The main canal and main urain alignments proposed by Italconsult have generally been retained but the
rigidly geometric layout of secondary canals and farm units has been relaxed anu adapted more to the natural topography. & reassessment of the channel cross—sectionr has led to revised bank slopes in some cases.
Italconsult•s proposal to construct two barrages, one at Melka Sadi to divert water to land developed 
by MABSCO and a second at A'elka Harer co divert water to Amibara, Eolhamo and Angelele, has been critically assessed, as it is estimated that savings in cost ano improvements in flood and drainage control would result if one barrage at 5elka Sadi could supply water to all
four components of the project*
In the light of development since Italconsultsubmitted their main report in 1969, the Regional Township and the AVA Administration Centre have been relocated.
The additional hydrometeorological data obtained by Italconsult have been fully utilised in assessing 
design and catastrophic floods and in estimating crop water requirements. While substantially agreeing with their estimated floods at the K.elka Sadi barrage site their argument that the large potential floods frcm
the Kessem and Kebena catchments are absorbed by the flood plains of these rivers and do not increase the flow in the Awash River, is not accepted. Similarly 
it is not considered that the pondage behind the Melka Warer barrage site will have any significant effect
on the flood flows in the Awash River. Therefore the flood flows at the A.elka Warer barrage site are estimated to be about double these estimated by Italconsult.
2 *4 UNDP/SF Project (1970 - date)
This project, managed by the State Rivers and Water Supply Commission of Victoria, Australia , was established to assist AVA in the technical and administrative development of the Awash Valley. The series of Informal Technical Reports prepared by the project have provided valuable additional data and further assessment of some of the technical problems
at Amibara. Useful discussions with members of the Project team have been held.2 . 5 IBRD Appraisal (.19 71)
The IBiJJ Appraisal Mission prepared a report cn all aspects of the Amibara development. Extracts from
this report were issued with the Termsof Reference for consultancy services fcr Amibara. The recommended design criteria contained therein hive been carefully studied and certain changes made in the light of further information and recent project concept development.
For instance, the advantages, both in operation and in saving in costs, of diverting the water by rreans
of only one structure at A elka Sadi have been studied. In order that flood waters can be passed without attenuation and the silt flow be unimpeded, barrage - type structure is proposed.SECTION 3 DESCRIPTION 0? PROPOSED PROJECT
3•1 GENERAL
3.2                 RIVER DIVERSION STRUCTURE 3-3                 FLOOD PROTECTION WORKS 3.4                     IRRIGATION SYSTK
3.5                    DRAINAGE NETHOT
3.6                     ROAD SYSTEM
3.7                     ON FARM LEVEL'/MELT 3.8                     SETTLEMENTS
3.9                     REGIONAL TOWNS IIP
3-10                  ADMINISTRATION CENTRE- 2 -
3. IESCRIPTIOA' OF PROPOSED PROJECT
3.1 General
3.11 Abstraction of Water from river,
Warer, with a second upstream weir at Melka Sadi to supply the Melka Sadi area,
Howeverj the headpond formed by the downstream weir would be above the level of the adjacent MAESCO concession and thus give rise to drainage problems• Also the foundation conditions to the downstream weir are not good. Thus, the
possibility was investigated of supplying both the Amibara
and Melka Cadi areas from the one upstream weir at Melka Sadi» The considerations are outlined in Appendix 3 hence
it can be f>oon that this alternative results in a saving of
cost and has certain other advantages and few disadvantages.
Thus it is proposed that the Amibara area (and possibly later ZngoToln and Bolhamo) are supplied with water abstracted from the River Awash at Melka Sadi. The water would be convey, to Amibara through a main canal running along the eastern
side of the MAESCO concession.
3.12 Irrigation Layout (.^rawing 2) Melka Sadi:boundary, with a drain alnng 'the central depression in the area.
The engineering works in Melka Sadi are being constructed by MA3SCO, with the exception of tho western flood protection dyke along the /.wash River* The main canal to Amibara will run along the eastern boundary and include the construction of the eastern dyke.
Amibara:
The nmibara irrigation layout utilises a central ridge to carry 
the primary supply canal and the primary drains follow lower land along the eastern and western boundaries* Both drains will be protected by flood dykes and discharge into Angelcle* If Angclele is developed these drains will continue as the primary drains for Angelele and discharge into Lake Cadda Bilen. Secondary supply canals following the contours take the water from the primary canal to the concessions
in which tertiary and quaternary canals serve individual fields. Quaternary and tertiary drains carry rainwater runoff and occasional surplus irrigation water to secondary drains discharging into the
primary drains. Apart from the AVA farms and the existing developments, the area of concessions is between 200 and 300 ha. Further details concession layout are given in section 4’
3*13 Communications
The proposed road system into and within the Melka Sadi
? -bd     Amibara     areas     is     described DR. 2.
It is proposed that
the j'.VA farm and at Melka >/arcr deration bo giver, cither to upg
in section 3-6 and shown on Drawing
the aircraft landing strips now at
be retrained and maintained and conai-
' ip nt Melka Itfarcr to take the strip
g an airport near the Regional town
aircraft of DC 3 size or to providing.
G hip to cater for Ethiopian
internal flights.J
-4-
•                   3*14 Regional Layout:
It is proposed to locate the regional township outside the irrigated area, and to locate the AVA administration centre at Mclka
Warer which is a focal point and a meeting place for all main roads feeding the project areas including tho potential extensions into 
Bolhamo and Angolele. The village at Melka Waror would remain to serve the Agricultural Research Station and the administration centre. 
Settlement villages for Afars are planned within, or adjacent to, the
AVA farms but no other communal villages are planned within the irrigated area. Labour camps or villages have been or will be established within 
the AVA farms to serve the farms and a similar arrangement is anticipated within each concession. Further details of town, villages, offices and 
housing are given in sections 3.8, 3«9 
3»10.
Three 100 m. wide reserves for livestock access to the Awash River are provided - at Mclka Warer, the AVA farm and between the
Amibars and Angclele areas. The Italconsult report recommended further reserves but it is considered that two ponds, supplied from secondary canals, each with 500,000 litres per day, situated in the class 6 soils
on the castorn boundary would be more practical for the following reasons a# Greater utilization of class 1 and 2 soils.
b.     Minimum trespass of livestock into the irrigated areas. c.     Reserves tend to become dust bowls.
d. 
Fewer channel crossings, hence less expenditure on construction and maintenance.
3*2 
River Diversion Structure
Tho purpose of the river diversion structure at Mclka Sadi
is to divert such flow from the River Awash as is required to irrigate5
the areas developed. To do this effectively, the structure should 
be such as to:
a* minimise accumulation of silt in the river,
b.     discourage entry of river bed load into the canal system,
c.     minimise the rise in river level during floods,
d.     amid erosion undermining the structures themselves,
c« avoid the creation of permanent still water which might encourage schistosomiasis.
To fulfil these functions, the structure must be a ’’removable dam”, 
i.e# having a cill level below that of the canal intake with the minimum amount of fixed superstructure above that level. This results in a barrage composed of piers sitting on an epron at river bed level;
between the piers must be gates which can be lifted or lowered tc form no obstruction to the river in flood# The canal headworks, just up stream of the barrage, must be capable of controlling hho u^cunt of
water abstracted and of providing protection from high river levels. 3»3 Flood Protection Works (Drawing DR. 3)
The Western flood dyke6 arc designed to protect the Kclka Sadi and Amibara project areas from flooding by the Awash, Kcssem and Kcbena rivers. Italconsult’s recommendation that the dyke alignment should not be closer than 100m. to the Awash river bank 
has been adopted. Outside this limit, the alignment has been chosen to optimise the use of cultivable land and minimise disturbance to existing development. Development within tho smeller meanders of the river is not anticipated. The economics of limited tree and 
bush clearance on tho banks of tho river will be studied as it is considered that a significant reduction (about 1 metro) on the height
of the dyke required oould rosult.- 6 -
Investigations for the Bolhamo development will cover the offect of restricting the flood plain on the western bank of the Awash 
at Bolhamo and the results of these investigations will decide the
fixc.1 height of the dyke in this area. Consideration will be given 
to too protection using brushwood, particularly in areas where the
dyke is close to the existing river channel.- Tree planting about
10 to 20 metres to the river side of the dyke whero necessary would provide additional protection though it may prove difficult to protect
the young trees against livestocks
The Eastern Dyke is designed to protect the Melka Sadi and Amibara areas from floods from the Alledeghi plain and the intervening escarpment. It follows the line of the main canal through Melka Sadi
and the Eastern drain in Amibara. Where possible the natural drainage depression will be used to encourage the flow of water away from the
dyke. It is not considered that this dyke will be expensive to 
construct because most of the material will came from the excavation 
for the adjacent drain.
f
The two dykes will converge towards the Angelele construction at the north end of the project area and will be taken into Angclcle
for a distance sufficient to prevent the flood waters flowing back into the project area. When the feasibility report on the Angelolc
area i6 completed more detailed proposals for the outfall arrangements for the main drain and the adjacent dyke alignments can be formulated.
3‘ 4 Irrigation System.
Drawing DR.3 shows the proposed project layout for the Amibara Irrigation project. The primary canal will follow the
central ridge of the project area with secondary canals supplying- 7 -
the settlement and concessionaire farms on boxh sides. The conces sionaire farm6 range in size from 210 hectares to 660 hectares, 
jlpart from the existing concessions at Algeta and Ambash, the area of concessions has been kept between the recommended 200 hectaros and 300 hectares, as far as the topography will allow*
For ease of construction and future access, tho primary layout to concessionaire level has been designed on a geometrical basis with secondary canals approximately following the contours; on-farm reticulation layout will primarily depend on the 10 cm. contour configur^tion and will be designed when this information is available.
All supply canals will be excavated to the design section 
with lining if required in some areas of high permeability. Measure-
mef; structures will be placed at suitable points along the primary 
canal and each structure supplying a concessionaire’s area will incorporate a volumetric measuring device. Section 4.7 gives more
details#
Provision will be made for surplus or rejection flow in 
the primary and secondary canals by means of outlet spillway struc
tures discharging from the primary canal into a secondary drain and 
by tail escape structures on secondary canals discharging into secon
dary or main drains#
Along  the  east  side  of  the  scheme,  outlets  from  secondary canals  will  also  supply  watering  ponds  and  rough  pasture  irrigation for livestock in areas of class 6 land*
Measures to curtail schistosomiasis will bo considered for ell channels and ''ends#- 8 -
The future supply for Bolhamo if required will be taken 
from the primary canal adjacent to the Melka Warer Research Station 
and cross the Awa6h River by means of an inverted siphon west of the
existing Amibara Settlement Farm* Angelcle will be supplied by 
continuing the Amibara main canal into that area#
Further details are included in Section 4.6 and 4.7.
3• 5 ^Drainage Network
The drainage network is based on two primary drains, one
draining the eastern half of the project and the other the western 
half. The eastern drain is a continuation of the Melka Sadi primary 
drain which passes under the main canal and the link road at Melka
Warer, then runs along the eastern boundary cf Amibara to Angclele. 
The western drain carries drainage water from the Melka
Warer Research Station and AVA Farm No. 1 then moves to the western boundary along which it runs into Angelelc^
The primary drains will be fed by a network of secondary 
and tertiary drains interlocking with the supply canal system. 
Whenever possible underdrain structures have been avoided.
However,   a  few  arc   required   in   order  to   provide  irrigation   and   drain age to some concessions.
The drain outfalls have not yet been established and further site investigation will bo carried out. Should Angelelc be developed the Amibara primary drains will continue as the Angelelc primary drains.- $ -
3•6      Road Systom
3.61 Main Roads:
The road system on Drawing DR*2 shows how access to the
project is obtained from the Awash - Hille Highway, Within the
project the road system has throe purposes:-
1.     Transport of porsonncl throughout the project for
administration, maintenance and farm work.
2.     Transport of goods and produce to and from farms.
3.     Movement of plante
For these purposes a main road system linking centres of administra tion, labour, accommodation and industry to all sections of the
project is required which will consist of:-
a.      The link road between the Awash Mille Highway, the Regional Township and the Central Administration Offices and Plant Yar \
r
b.      The eastern road running from the Melka Sadi barrage, north along the eastern boundary of the project to Angelele.
c.      The western road running along the western boundary of the Melka Sadi area.
d.    The central road running through the centre of the Amibara area, with four spur roads.
e.    The Bolhamo road..
Theeo roads all intcreoct at Melka Warer which is therefore, a natural selection for the site of the project administration centre* They are themselves linked at salient points.
3.62 Secondary Roads:
A network cf all-weather secondary roads is required for roaintainenco access along major canals and dykes and tc ether major- 10
engineering works. Thoso roads arc net expected to carry any farm traffic except for the movement of major items of plant. Should any section dovelop a flow of traffic it can later be uprated and incorporated in the main road network.
3.63      Farm Road6:
Those roads will be fair woather roads linking the fields and farm buildings to the main road network; their upkeep will be the responsibility of the concessionaires. Structures for passing 
all roads over canals will be constructed under the project. Their design is dealt with under Sections 4.7 and 4*8.
3-7 On Farm Development
Field ^tertiary and quarternary) canals will supply irrigation units of approximately 50 hectares. The furrow lengths, nrill be of the order of 250m. to 300 m.,) but irrigation units will not be laid 
out on a strictly geometric basi6. The final configuration of tho field unit6 will depend on the 10 cm. contour to surveys. Tertiary and quaternary drains will collect storm water runoff and any excess irrigation water from each field and discharge into secondary drains at the bottom of each concession area*
Throughout the project thoro aro a number of high areas unsuitable for cultivation. Most of them contain sources of road surfacing material and could serve as areas for concessionaire housing, offices, workshops etc. In addition, there are substantial areas of
Class 6 land unsuitable for irrigated cotton development. These
could also sorve as housing and management centres fcr the concession aires thus saving valuable irrigablo land within the scheme. Some secondary canals will bo extended to irrignto pasture for /.farlivestock as well as supplying watering ponds in these areas. This
will lessen the need for livestock to cross the project area for
watering at the Awash River.
It is suggested that the largo area of Class 6 at the north 
of Amibara be used for pasture grazing by the Afar livestock. .Floods
from the Awash would irrigate this area if the protection dyke ib
brought into the project area to follows the linG of the western 
drain. Additional irrigation of the pasture areas in the dry season 
could be provided by a small pump on an outlet from a nearby secondary canal.
3.8 Settlements
3«8l AVA Farms:- General:
The existing Amibara Farm No. 1 includes 1,050 hectares of developed land (535 hectares settlement area and 515 hectares training farm), and the nucleus of a farm complex and village comp rising office, workshop, fuelling point, manager’s house, two ’A’ type houses, eight B’ type houses, guest house, generator house and a rudimentary water supply system. A large number of traditional houses have been constructed over portions of the area designated 
in I971 for tho development of mere pormanent buildings and houses.
A eocond Amibara Farm (No.2) controlling the development of a further 1,000 hectares east of the primary oanal is to be located approximately 4 km- to the south-east of Farm No. 1T
f
It is envisaged  that Farm No.  1  will operate as the parent to  Farm No.  2  and  to  a future farm cn  Bolhamo  in  respect of overall farm management,  maintenance and  miner repairs of farm vehicles and implements, and of general social sorvicos.- 12
The following factors arc to be investigated or substantiated 
with regard to the layouts and developments of the Farms:
Farm No. 1:
a.           The projected numbers of permanent houses cf various
types required to be built during the next three to 
four years. These will accommodate staff required by 
the Farm.
b.    Operational buildings required for the Farm i.e. addi
tional office accommodation, stores, etc.
c.    Buildings etc. to bo provided for housing casual labour.
d.    Buildings etc., required for social services. E.g. 
clinic, shops and market place, bank, hotel, junior schol, 
comruhity hall, church, mosque,. reareation facilities.
e* Upgrading of public services, water, electricity and 
sewage treatment, to adequately serve the total development. 
Farm No. 2
a.   The projected numbers of permanent houses cf various
types required to be built during next four years. 
These to accommodate staff required by the Farm.
b.   Operational buildings for the Farm. i.e. offices, stores fuelling point and workshop for minor everyday maintenance, etot
Ct Buildings etc. to be provided for housing casual labour. d.   Buildings etc. required for social services, i.e. shops
and market placo, community hall or area, recreation facilities.
e.    Provision of public services, water, eloctricity and sewage treatments.3,02 Settlement Villagc-s:
The general plan is to settle 800 Afar families in villages situated around the 2,000 hectares of settlement land  each family 
r
being allocated a 2.5 hectares plot for cultivation. Whenever possible
the settlement villages are to bo located on Class 6 land or in irre
gular shaped pockets of land within the farming area; they will not
be located in areas subject to flooding and they will be positioned 
so that no settler will have to walk further than 3 km. to his plot. 
It has been intimated that the settlers will be allowed 
limited herds which will be communally grazed on fallow • plots.
It is also intended to allocate each village sufficient land or
allotments for the growing of vegetables.
Before determining the optimum size cf a village, the number
or locations of villages, it will be necessary tG carry out some
additional research into the structure cf the Afar’s society with 
particular reference to their future way of life in the settlement
villages..
Points cf special interest are as follows
a. The average sizo and composition of a family unit. Whothor or not the Afar will retain more than one wife and hor children in the village?
b*. Whether or not individually owned herds will be retained and grazed outside of the area of the Settlement Farm?
c< What will be the basic composition of the village? Will it still be divided into housing compounds - if so what size? (4 houses?)
d« Have the Afars any prcfcrance for one of the four house designs already prepared by AVA? Or is there a need for furthor research and design?- 14 -
N.B. (The scope of this consultancy service docs not include
a detailed sociological study of tho ^f.irs, he we ver, a basic investi
gation is necessary to ascertain tho main factors which will influence
planning and building designs).
Construction of a trial portion of a Settlement Village is
to be arranged as scon as possible so that the afar’s reactions to 
their new houses can be observed, and fixture houses codified os
necessary. A settlement of Afaro has already been indicated 2.5 to. 
west of the Amibara Farm No. 1, and this would appear to bo the appro
priate location to commence the trial scheme.
2.9 Regional Township
It is envisaged that tho Regional Town will serve as the main community and government administration centre for the Middle Awash Valley and the surrounding countryside. It will net only contain essential local government offices, medical, educational and social services, but will also generate its own commercial and industrial elements. It is ccn6idorcd likely that its population could rise to between 30,000 and 40,000 persons*
Various sites have been considered for the establishment of the Regional Town. A previous proposal that it should be sited cn the "plateau” at Melka l.’arcr is not favoured as the area available
is considered too restricted for such a development, and the "plateau" is considered raero suited for the location cf tho AV;i Administration 
Centre (refer Section 3*^0)
It is proposed, that the Regional Town should bo sited cn the escarpment to tho north cf the now link read which w*±l run 
between tho awaoh-Millc Highway and Molka Harer. Thin location will allow good r.coooc to all 4crtc cf tho Project Area using the prv;x?tax.- 15 -
new road network;- it will have a relatively short road connection to the Awash-Millo Highway,, and will allow for unrestricted expansion without encroachment on to agricultural land* Also the slope of the escarpment would enhance the visual appearance and environment of the town#.
It has been stressed that it is not intended to locate
the town on or close to the Awash-Mille Highway, ncr on the new Link Read, as this would too easily encourage undosired developments to appear close on both sides cf these road6. As with other developments
in the Valley it is recommended that AVA and ether appropriate Govern ment Departments be given every encouragement to create a Regional Town Authority with powers to control building and prevent unauthorised squatting in the area cf the tewn, and along the highway and the link 
read.
More detailed consideration will be given to the precise location and extent cf the Regional Town, and a master town plan 
will be prepared indicating the principal network of reads, zoning: of elements, and various infrastructures.
3«10 Administration Centre
Three possible locations for the AVA Project Administration Centre have been considered: i.c. at Amibara, or on the escarpment
(in the Regional Town), or at Mclka Warcr.
The location of the project administration together with the farm administration at Amibara is con6idored tc be cf doubtful benefit as the two administrations will have different functions which should not be ccnfusod nor merged together.The location cf the project administration at ^mibara would produce the following definite disadvantages--
i. ii. 
It would net be centrally placed in the project aroa.
It would not be conveniently placod in relation to the
first stage link read, and satisfactory permanent read access would not be available until late in the Major
Works Contract.
iii.          Its construction and expansion will tend tc encroach onto agricultural land?
The location cf the project administration on the escarpment would have the advantage of being close to the Awash-Mille Highway to .'Ielka Warcr Link Reads, and cf being close to the centre of the Project area. However, it is considered prudent not to confuse the Administration Centre with the future planning, development and construction of the Regional Town?
The location cf the Administration Centre cn the nplato?x,: at Melka Warer is the most favoured solution, and has already been agreed in principle by the AVA. This location has the following advantages :•
ie of being close to the intersection of the main roads serving the project area.
ii.        of being on the line cf the first stage link road.
iii.      of being within a few kilomctors cf the future Regional Town.
iv.    of being adjacent to the existing research fern.
v.    of being in an attractive area suitable for the placing of offices and housing, above any flood level and nri encroaching upon any agricultural IaucVSBCTION 4
DESIGN ASSUMPTIONS AND CRITERIA
4.1
WATER AVAILABILITY
4.2
CROP WATBR REQUIREMENTS AND CANAL CAPACITIES
4.3
RIVER DIVERSION STRUCTURE
4.4
FLOODS AND FLOOD PROTECTION
4.5
DRAINAGE REQUIREMENTS
4.6
CANALS
4.7
IRRIGATION STRUCTURES
4.8
ROADS AND BRIDGES
4.9
ON-FARM WORKS
4.10
SETTLEMENT VILLAGES AND HOUSING4. DESIGN ASSUMPTIONS aND CRITERIA
4.1 Water Availability
Water availability for developments in the Awash Valley has been carefully studied by AVA and their advisers, and it is clear that that for Amibara is
largely dependent on releases from Koka Dam. The present releases will be sufficient to meet current irrigation requirements. However aVA have foreseen 
that the pattern of these releases will need to be
modified during the next three years in order to meet anticipated irrigation requirements after 1976. AVA
and EELPA have therefore established a technical committee to agree and plan operational techniques to satisfy 
both irrigation and power requirements. The feasibility study for Angelele and Bolhamo includes a requirement to determine"the best method of utilization of Awash 
water for the project area including recommendations on operation of Koka Dam to provide adequate security of minimum Awash River flows” for these arena.
4.2 Crop water Requirements and Canal Capacities
4.21     Crop water requirements for cotton: Meteorological data observed at the Melka Warer
Research Station for the years 1965 - 72 were used to estimate the likely maximum consumptive use of cotton at Amibara in order to obtain the maximum irrigation duty for canals. Four methods of estimating crop 
water consumption were used. These were the twoempirical methods of Blaney-Criddle and Thornthwaite, the ’rational’ method of Penman to calculate open-water evaporation and a third empirical approach due to Hargreaves and Christiansen which computes a climate factor equal to Class Zi pan evaporation. Crop factors, which allow for the physiological age of the crop, 
then applied to the computed potential evapotranspiration to give the crop consumptive use during the plant growth period.
For cotton at Amibara, the crop is generally in the ground for 150 days; however during the last 50 days
when the second and third picks are taken, no benefit is obtained from a further irrigation application. 
For consumptive use calculations a crop season of 120 days has therefore been assumed.
Research carried out at the Kelka Warer Station indicates that the optimum planting period is from
15th Kay to 15th June. However, later planting does
not seem to effect yields adversely, although the requirement of a three month pest-free ’’dead" season necessitates final planting by mid-September. For
simplicity of calculation a three month planting season from 15th Kay to 15th August has been used. This plant ing period is currently being adopted at the AVA farm.
Table 1 shows the comparative seasonal consumptive use requirements using each formula. The maximum
half-monthly consumptive use requirement is 105 mm. in the second half of July. However, a pre-irrigation of approximately 150 mm. is applied prior to and just after planting - a period of two weeks. This require ment determines the critical duty for design purposes.For design of canal duties, assuming continuous
flow to each concession during the pre-irrigation period 
the duty required at the plant root zone would be
1.16        1/sec/ha.        Assuming        an        on-farm        irrigation        efficiency of 85% the canal duty at the concession inlet is
1,36 1/sec/ha.
Canal seepage losses are probably fairly high. 
Previous experience at the AVA Farm at Amibara indicates that losses from secondary canals are cf the order cf
1000 m /km/day and Tables 2 and 3 estimated the gross design flews for secondary and primary canals assuming these seepage losses. The resultant overall design 
duty from root-zone to headworks is 1.68 1/sec/ha. equivalent to an overall irrigation efficiency of 69%. 
However, in view of the most recent work at the AVA farm, thi6 efficiency appears rather high, and further assessments of on-farm efficiencies and canal losses
will be made to ensure no underestimation of design 
duty.
4,22  Crop water requirements for other crops:
In order that the canal capacities will be able to cater for any variation in crop during the life cf the project, the water requirements of other crops
e.g sugar, alfalfa and vegetables will be estimated. Preliminary calculations indicate that the pre-irriga- tion requirement of cotton will meet the maximum requirement of other crops.
4,23      Canal capacities:
Figure 1 8hows schematically the layout of theproject and gives gross areas of concessions together
with the maximum outlet flow to each concession, based on the above design duty of 1.36 1/sec/ha.
Table 2 gives the gross areas commanded by each secondary canal and the gross design flow at the head 
of the canal. These canals will be designed for a
constant discharge throughout their length so as to provide for rejection flows (see 4.24)
Table 3 presents a schedule of the primary canal offtakes and the primary canal design flow to supply 
zi/nibara. In addition, sufficient capacity to supply 
the existing development at Melka Sadi, will be included 
and allowance made for future developments nt Bolhamc
and Angelele as required.
The required maximum flows are estimated as follows:-
< Project
Gross
irrigable
area
(ha)
Cultivable irrigable
area
(ha )
Maximum flow on 
duty basis (cumecs )
Design maximum
1 flow* (cumecs )
Melka Sadi Ami ba ra
Bolhamo
«ngelele
—- w
5,100
10,100
8,900
4,750
*- — »• — .
4,330
8,600
7,560
4,040
6.5
14.4
12.7
6.8
5.5
12,2
10.8
S-.8
J.
|
1J
The design canal flows are calculated for commanding the culturable irrigable areas i.e. 85% of the gross irrigable area.
Secondary canal cutlet to concessions will be designed to supply the full maximum design flow..However, as it is unlikely that the whole area will
be pre-irrigated at the same time, it is considered 
that the primary canal will be required to supply only 
85% of the maximum design flew. This takes account, 
of variations in individual concession requirements, 
possible rotation of application, water management
fluctuations and limitations in design data.
4,24 Operational considerations affecting canal oapaciti e**
With the period of pre-irrigation requirements
lasting from 15th May to 15th August, it is probable
that rainstorms during July and August will reduce
concession irrigation requirements. Because of this, 
a situation could arise where concession outlets are
closed due to excess rainfall and there will be a
surplus discharge in the secondary canal until suitable
adjustment to the secondary head regulator is made. 
This discharge is known as secondary canal rejection
flow.
In any section of secondary canal the design dis charges r-HjJUl dxe Xhec               of the
discharges required by concessions downstream of the section, plus the possible surplus discharges. Thus secondary canals will be designed using a uniform canal section from head regulator to tail escape where surplus or rejection flows discharge int-o prinary secondary ^ain.
Similar surplus discharge rejection measures will be installed along the primary canal in case a situation arises where secondary head regulators are shut down without suitable adjustment being made to either the
main canal or the primary canal head regulators •
iIntermediate escapes will be installed at the cross drainage structure abc-ve Nelka Harer and at three locations along the primary canal within the Amibara project. Each escape will discharge into a secondary 
drain and will be designed to accept two-thirds of the maximum likely rejection flow between it and the preceding escape structure,.
By these means, the design capacity of the primary canal can be reduced along its length, and possible
damage or disruption due to surplus flows, resulting from cperational control errors, obviated.
4.3 Iciver Diversion Structures
4,31 General:
The site proposed by Italconsult for the Lelka Sadi weir appears suitable for the barrage, and geologically acceptable, though it would seem preferable to locate the structure a little further upstream, to 
be further from the bend in the river and to utilize
the sounder foundation material downstream of the apron.
(The geology of the site at Kelka Warer is un
attractive, but in view of the extensive investigation already carried out, it would be adopted as the best available, and a suitable design adopted for the
conditions if this structure were to be built,)
The barrage (see Drawing DR. 6) will be designed to pass flood flows without causing appreciable change
to the natural water surface of the river. Flood flows are discussed in section 4.41, These consider!tic nslead to a barrage needing^clear opening of about 30m.. (cf. 50m. at Melka Warer). Consideration has been given to the desirability of hydraulic model tests,. Such 
tests would confirm or amend the hydraulic calculations and might yield more reliable estimates of the upstream water levels under different flows, and of the distribu tion of water velocities downstream of the barrages,
i.e. an indication of areas subject to local erosion. However, since nearly as manyae.&fr*ptiaiw? TWultfl iuHte to '’built into" the models as into the calculations
and since there are innumerable examples of similar barrages, it is considered that the expenditure of neither the time nor the money on such tests is justified in this case.
4,32 Construction materials:
To maximise the use of labour, the IBRD have suggested that the structures be constructed of masonry. We have concluded however that, though this type of construction might well be suitable for weirs as previously envisaged, it would add considerably to the time and cost of the barrages now considered desirable., Experience with international contractors indicates
that if they were asked to bid for such masonry structures the majority would propose an alternative, less costly, design usixg, concrete.. The foundation conditions and possibility of earthquakes alro indicate the desirability 
of using a construction material capable of taking tensile forces. It is thus proposed that the design of the structure near the gates should be of reinforced concrete, and that other parts of the structure should be designed to encourage the use of labour where it is more economi cal, e. . gabion proteccion of slopes and downstream bed..
e4.33 Proposed form of structures:
The barrage gates will be required to maintain 
a constant pool level with continually varying flow
and to pass floating trash, (though a log boom upstream
of the barrage may/be desirable to catch large trees). 
Overshot, rather than undershot, gates would appear
to be more suitable in these circumstances. The
simplest form of overshot gate is the hinged-leaf type
and the use of this type is proposed. An odd number
of gates is desirable to maintain uniform flow and it
is proposed that there should be three 10m. wide gates
at Melka Sadi. (Five 10m. would be needed at Melka Warer
where the flood flows are considerably higher due to 
the influence of the K. Kessem and Kebena). The exact
form of the gates, and their operating mechanism, will
be settled with the manufacturer selected from tenders
obtained on performance specifications. It is proposed 
that operation of the gates should be by means of hand-
controlled prime movers, by a resident gate-keeper.
Provision will be made for later automatic operation.
In view of the foundation conditions at the barrage, site, and the possibility of earthquakes, it is proposed 
to mount each gate in a separate reinforced concrete "box" (made up of the cill, piers and bridge slab).
Such construction will minimize the risk of gates jamming due to structural movements resulting from temperature changes or differential settlement of the foundations.
Downstream cf the gate "boxes" will be an apron of sufficient length to ensure that the discharge at the toe is uniform stream flow. The river beds andbanks immediately downstream of the apron will b^proteot against erosion.. The canal headworks will'.be placed
» r-as near a6.jpossible to the barrage in order that opera tion of the adjacent gate can flush away any accumulated bed load. The design of these headworks will be depen dent on the type of central gate chosen, the choice
lying between Taintor gates or vertical lift sluices. 
Time has not yet allowed a reliable economic comparison tbv establish the better choice. Measurement of the
flow would be downstream cf the headworks.
4-, 4 Floods and Flood Protection
4-.41 Estimated Flood Flows:
Frequency analyses of the available flood data
for the Awash Kessem and Kebena Rivers have been carried 
ooi t using the following methods;^
(a ) Gumbel Type I
(b)     Log - Pearson Type III
(c)      Boldakov’s Bmpirical Method
Table 4 summarises the results, from which it
can be seen that there is close agreement between the
Gumbel and log - Ibarson results, both cf which give
higher, but similar floods than Boldakov, The results
of either of the two former methods could be adopted 
for design. The former has been chosen and the frequency 
analysis plotted on Figure 2.
From    this    graph,    it    can    be    concluded    that    design floods    of    1000    and    2000    cumecs    will    have    the    following return periods:1000 cumecs
2000 cumecs
Melka Sadi 10,000 years
frielka Warer
5-10 year9 50 years
The floods at Melka Sadi are from the Awash River only. The floods at Melka Warer assume that a flood 
in the Kebena River will occur at the same time as a
flood in the Kessem River and will be magnitude. Under these circumstances which are
of the same
considered      rare,      the      flow      in      the      /.wash      River      is      assumed to be base flow only.
Magnitudes and return periods of ether flood 
flows are given in Table 4.
4.42    Western Flood Dyke:
The western flood dyke is designed to protect
the Melka Sadi and Amibara irrigated areas from Awash 
7
River flood flows up to 2000m /sec. A preliminary 
economic analaysis cf the 37 km. of dyke downstream
of Melka Warer has been carried out to indicate the
long-term advantage cf building the dyke high enough 
for this purpose.
The hydraulic OAlcnilations were based on the method used by Italconsult, which is considered basi cally sound, and the economic assumptions were:
(a)      Annual cost of dyke is 10% of capital cost
(b)      Annual cost of flood damage is estimated damage divided by the return period
(c)       The extent of damage varies with the size of the flood, e.g. damage resulting from a 20 year flood is assumed will be 10% of value of the cotton crop. 20% for a 100 year flood.(d) Cost of dyke earthworks BS2/m§
The resultaof the analysis, shown cn Fig. No. 3, are indefinite but indicate that an average height of
dyke of 3-4 m. would be economic. However, preliminary calculations also indicate that clearance of trees
beside the river would lower flood flews by 0.5m. - Xm Both economic and hydraulic calculations will
be repeated as mere data becomes available. In the meantime, design will proceed assuming an average height of 3m. to be the economic optimum if the river-side of the dyke is cleared of major obstructions, (leaving 
a 10-20 6trip of trees in front of the dyke to reduce
flood velocities at the edges and hence minimize erosion)
4.43      Bastern Flood Dyke:
Italcon6ult used the Arba river flows for calcula ting the eastern dyke flood flows by a comparative area
2 
method, U6ing a flow of 244 1/sec/km . The Arba catch ment is net exactly comparable to the catchments flooding the eastern side of the project in area, topography 
or altitude. While this method is not excluded, attempts to determine the flood flows using the updated rainfall
intensities referred to in section 4.5 and calculating times of concentration and run off coefficients will be made. It is also proposed to compare these results
with the empirical ltyves formula Q= 675 A ’ (where Q is flood flow in cumecs and A is area of catchment in sq. miles) which was originally developed for hilly terrain in India but has been found most useful in Africa.
0 67 Italccnsult gave adequate and satisfactory reasons why the flood flows should be kept separate from the drainage flows on the eastern side of the project area. 
It is not envisaged that the cost of construction will
be high for the eastern dyke as most of the fill required can be obtained from the excavation of the eastern main drain and of the main canal from Melka Sadi. Attempts will be made to compare costs with floods of different return periods, but these will only be as reliable as
the short hydrological and meteorological records allow. An economic evaluation of the benefits will then be carried out.
4,44     Dyke cross-sections:
The dyke6 will be designed on the basis of soil samples taken before, and during, construction. It
is proposed to adopt a 3m. top width to allow just enough width for a small vehicle - to encourage regular inspection of the river-side of the dykes. Preliminary cross-sections of the dykes are shown on Drawing ” . No. DR. 7.
Drainage Requirements
4.51     Storm Runoff:
tecent examination of the autographic rainfall recorder charts at Melka Warer Research Station indicates
several higher rainfall intensities than Italconsult observed in the short period of records available to them. Maximum intensities recorded up to 19th August 1973 are 6hown on Fig. 2A<It is not possible from the seven years of rainfall records to carry out a reliable statistical analysis
of probable rainfall intensities. Based on the data available an intensity cf 50mm/day is considered appro priate for drainage design.
It is anticipated that with irrigation applications of about 80mm. every three weeks, storm run off in
24 hours from a field under irrigation, or where irriga tion has just been completed, could be 80% of the storm rainfall* For a field about to receive its next appli cation this figure would be reduced to about 10%. Thus tertiary drains will be designed to carry 80% of the
50mm storm, but secondary and primary derain will have a design flow of only 45% of the storm, This is conservative as the relationship between the percentage run off and the time since the last irrigation is not
linear. ' Using the above figures the average storm run 
off will be 2.6 litres/sec/ha for primary and secondary drains, and 4.6 litres/sec/ha for a tertiary drain.
This represents a flow from a typical 250 ha. unit of
0.7m /sec, excluding allowance for storage, The combined capacities of the east and west primary drains cn this
basis, will be about.36m /sec, excluding the effect
of storage.
4,52    Kejection Flows:
The amount of rejection flow diverted into the the drainage system will depend on the irrigation control proposed for the project. This is discussed in 
section 4,2. Close control would enable surplus irriga tion water during storms to be retained in the primary canal until it discharges into the primary drain .atthe north end of the project area and the only increased drainage capacity needed would be in the short discharge section of the main drain at the lower end of the project. Loose irrigation control could result in the individual concessionaires being able to reject the total irriga
tion flow into the secondary canals by closing the
control gates into their concession. The water would 
then flow along the secondary canal into the secondary 
or primary drains. The capacity of these drains would 
have to be increased accordingly.
It is not anticipated that 100% rejection could 
occur at the same time as the peak outflow from storm
rainfall; nor is it anticipated that peak storm outflows
would occur at the time of pre-irrigation. For normal
irrigations the main canal flows for the Melka Sadi
and Amibara areas are approximately 2/3 of the maximum
j
i.e. a total of some 16m /sec for the Melka Sadi and Amibara areas. Preliminary estimates indicate that
the difference between tight and loose control of irrigation could result in between 10% and 75%
of this flow being diverted into the drainage system
at the time of peak storm runoff, with corresponding changes in drain size and therefore cost. A close
control on irrigation water and an efficient means
of communication between the water guards and the head works can therefore result in considerable saving.
Further studies are planned in connection with this problem.
4.53      Water Table control:
The data available on the water table under Melka Sadi and Amibara gives incnrclusive results.Some additional information will be obtained from the boreholes included in the site investigation contract
but it is the network of piezometers which will be installed, .under the main contract, both at Melka Sadi and Amibara that will enable the movements of the sub surface water to be adequately monitored. Regular readings of these piezometers compared with records
of irrigation applications, rainfall and flood flows
will indicate the effect of irrigation and the need, 
if any, of additional drainage measures.
Should sub-surface drainage be required in the future, it is proposed that the discharge be pumped 
into the drainage network and no allowance will be
made in the design of drains for future deepning.
A possible interpretation of the ground water situation at Melka-Sadi and Amibara is discussed 
in Appendix B.
4•6 Channel Design
Canals:
It is proposed.that unlined earth channels be used for all canal8 and drains, Where 60il conditions indicate some form of sealing to prevent excessive seepage an appropriate lining, will be specified.
The following design parameters have been used; these correspond to present design practices in Victoria, 
Australia and the United States of America (U.S.B.R.).
All canals are designed 
formula/ The maximum design velocity adopted is
using Mannings
0,6 metres/sec. The design sections compare favourably with lacey* regime sections.Mannings n - For canals discharging up to
1 cumec, value of 0.0275 has been used to allow for
the retarding weed growth that may develop in the smaller canals. For the main primary and secondary canals, 
a value of 0.025 has been used.
Side slopes - The shape cf the canals will be trapezoidal with side slopes for tertiary and flelr’ canals cf 1.5 horizontal to 1.0 vertical, and for
main and secondary canals of 2.0 horizontal to 1.0 vertical. Outer bank sides will be 1.5 : 1.0. Soil samples will be tested to confirm the applicability 
of these proposed slopes to all parts of the project
area .
Longitudinal slopes - These will be selected to suit the ground slope and give balanced
The bed slope will be flattened, by the
to restrict the velocity of flow to the maximum.
Free Board - The canal free board dependant on canal size and location with
regulation structures are given below:
Canal capacity (cumecs)
up to 0.8
0,8 to 2.5
2.5    to 7,5
7.5     to 20
Minimum values of
Min, freeboard for earth canal (m)
0.45
0.60
0.70
0.85
earthworks
use      of      drops, allowable
will be respect to free board unlined
Top width of bank - minimum top widths will depend on requirements for roads and for a maximum hydraulic gradient of 1 in 5 from full supply levelto outer toe of bank.
Radius of curvature - The radiu6 adopted for the
outer            bank is   equal tc          5 times the             bed width          of    the
canal    •
Typical
sections
are shewn in
Drawings
No.
DR. 8
and DR.
9.
4.62
Drains:
In general dr
ain
design will
be similar
to
canal
design except as detailed below:
Manning's formula will be used for drain design with a maximum permissible velocity of 1.0 m/sec for peak storm flow.
Roughness coefficient - Manning’s ’a’. Por small drains discharging up to 1 cumec a value of 0.035 will
be used. For primary and secondary drains a value of 0.030 will be used. It is assumed that there will be some weed growth in all drains, and that trimming will not generally be carried out for Grains. Weed growth and irregularities will have a correspondingly greater effect in smaller drains.
Side
slopes
-
All
drains
will
Ibe
trapozoidal
with 
side
slopes
gene
rally 
of
1.5 
horizontal
to 
1.0 
vertical
depending on soils.
Longitudinal      slopes      -      These      will      be      selected      to minimize      the      need      for      drops,.and      also      to      minimize      the earthworks.       It       is       not       anticipated       that       crops       will       be required except possibly in the primary drains..
.Freeboard - Minimum values of freeboard for drains are given below. It is not considered that freeboard 
in drains need to be as high as in canals as drainsgenerally flow with water levels below ground and overtopping may only cause slight damage at time of flood ,
Drain Capacity (m /sec)
0 to 2,0 
2,0 to 15.0
Above 15,0
3
Minimum Freeboard (m)
0.25
0.40
0.60
Typical sections are shown on Drawing No. DK. 7
4.7       Irrigation Structures
In the structures maximum use will be made of
6tone masonary for headwalls, cut-offs, foundations etc. 
Small culverts and syphons will utilize concrete pipes
but the larger ones will of necessity be concrete box 
structures,.
Several typical structures are shown on Drawings
No, DK, 8 and 9 and a description of the more important types of structure is given below:
(a)        Combined check drop and offtake structure on the primary canalj.the check drops will be s imilar
to those proposed by Italconsult but the drops on the secondary canals will be formed by a rubble check and filaci6 as shown on Drawing No, DRi.9.
Provision will be made to insert stop-logs for maintenance and also where required the offtake culverts will be long enough for access roads to pass over them.
Measurement flumes will be incorporated where required.(b)         emergency Spillweirs on Primary Canal Four emergency spillweirs will be incorporated 
in the primary canal to relieve any overloading of this
canal resulting from out-of-phase shutdown of supply 
or accidental increase of headworks gate opening.
These weirs will be either syphons or drop inlets
and will discharge into the drainage system.
(c)       Head Regulator and Drop from Main C^nal to Primary Canal.
In order to provide pondage in the main canal
a major regulator is required at the downstream end, followed by a drop to primary supply level. These two structures will probably be combined into one structure which will be located upstream of Melka Harer,
(d)         Cross Drainage Structures
These have been avoided where possible but will be required at the crossing cf the Zimibara Primary Canal and the Melka Sadi Primary Drain, and at a few miner supply and drainage crossings.
They will consist of reinforced concrete boxes
or pipe barrels depending on size.
(e)       MAESCO offtake
A major offtake is required at the southern end of Melka Sadi to supply the MABSCO main canal. This structure will be required to pass about 5.5 cumecs
and will have to be constructed out of reinforced concrete because cf its size.
(f)        Regulator Tail escape of Primary Canal
The tail escape of the Amitara Primary canal will be constructed as a regulator which will feed Angelele when required.(g)        Tail Waste to Secondary Canals
These will be similar to the drop structures in the secondary canals but will incorporate a small drainage pipe so that the canal system can be completely dried out for maintenance.4.8         Roads and Bridges
4.81 Classifications:
Roads within the project area will fall into theso classes:
(outlined on Drawing No.IS* 7)
i) Primary RoadB will connect the main administrative areas
with the Awash-Mille Highway and will also provido the main arteries of tho communication system within the project. They will be gravel surfaced roads of width 
to enable vehicles to paBS n* reasonable speeds.
ii)   Secondary Roads will link the individual concessions with 
the primary road system. They will also be grave l-rsur faced and of width to allow large vehicles to pass at slow speeds.
iii)       Tertiary Roads, unsurfaced, will serve the internal communications of the concessions down to field level. Large
vehicles will not easily be able to pass on these roads except at junctions and at a limited number of passing places, It is assumed that maintenance of tertiary roads will be carried out by the concessionaires.
When plans are finalized for the Project Administration the townshipi ginnery, etc., attempts will be made to forecast vehicle densities on primary end secondary roads. It is anticipated that heavy vehicle densities will bo groatest during tho cotton picking season, when raw or lint cotton is being 
transported out of the project aroa using either tractors and trailers or lorries. For primary roads whore
vehicle densities exceed 250 vehicles per day consideration Hillbe given to the economics of sealing the road Burfaco 
4.82 Surfacing material:
Surfacing material f r roads will generally be naturally oocuring non-graded stone or volcanic ash. The Report on Engineering and Investment Aspects by D.S. Polo, and tho subsequent IBRD document on the Appraisal Mission (197*0 Annex 2 give guidlincs for road design, which include gravel surfacing with scoria over
a width of 5 and 4 metres for primary and secondary roads. However, experience in the lower Awash shows that Btone stands up to heavy traffic better than scoria and docs not crush to 
duBt so easily, resulting in lower maintenance costB. Thus, preference will be given to natural stone or gravel as a
surfacing material provided sufficient quantities suitably graded, are provided, and they are not substantially more expensive than volcanic ash.
The link road will be built using natural stone surfacing and its performance will be carefully watched during its early life so 
that improvements can be made in the design of other project roads constructed later under the main contract.
4*83 Road widthe:
A road width of 5 m. with 1 m. wide shculd.Gr was previously considered adequate for the primary roads. However, it is difficult
for two large vehicles to pass t.L reasonable speeds on a road less than 6 m. wide and Boft shoulders will quickly become damaged if vehicles arc continually pulling onto them. It is proposed therefore to surface the full width cf tho carrigewayWith natural graded stone. The resulting 7 m. wide hard surface
for primary roads conform? to the existing Imperial Highway 
Authority’s standard for Ethiopians Secondary Roads (though this
is shortly to be increased) and if subsequent scaling is proposed 
a 6m. wide sealed surface would complete the road to existing 
IHA Secondary Road Standards.
For project socondary roads the same process is prcpcscc. but the
overall width will be reduced by 1 metre in line with IBRD
proposals. This will allow large vehicles to pass at reduced 
speed. Tertiary roads will be unsurfacod and 4 m. wide overall
oxcept at selected passing places.
4.84 Road Bridges (Drawing No. MSix/Dr. 7)
Three      different     types      of     bridge     construction     would     be     suitable, viz:
(a)    steel girder
(b)    reinforced concrete
(c)     prestressed precast concrete
as (b) and (c)  will utilize local materials they alone have been
considered.
It has been  shown  during  the construction  of the Awash-Wille Highway  th.  t prestressed  concrete is cheaper than  reinforced concrete.  This is mainly  due to tho high ccst Qf shuttcring and scaffolding required for reinforced concrete.
Although somo problems have been encountered in obtaining gcod quality material for tho production of hi h stronsth concrete
eI
(poorly gradod Band hcin- the noct persistent) anticipated that thio will bo so difficult in area.
it io not
tho Amibnra prejoct
Tho     Imperial     Highway     Authority     have     standard     doaigne     for     verious bridge typeo, spans and widtho baaed on tho AASHO Highway
Dridgeo    1961•    It    io    proponed    to    uoc    those    standard    bridge    dosigns no a basin for dooign at /unibora.
Tho propooed carriageway width for doublo lano briclgoo io 7 m. 
and for Dingle lane, 4 m. wido- Bridge longthc will be
determined by nite conditions but standard longth spans will be
uoed for economic ccnotructicn.
4.9       On-farm Works
The proposed conccBDionnire form layout shown on Drawing No#
DR. 3 was prepared on the basic of the metre contour sheets of the
area. However, for the on-form irrigation layout, wo will bo uoing
10 cm. contouring at 1:5>000 scale. Thio will enable the on-farm
reticulation layout to tako account of subtle changes in level
hence minimizing land preparation prior to irrigation. In addition, 
the more accurate survey data will enable us to design fields with a
minimum variation in furrew slopo thus permitting a uniform application 
of water end alBo to minimize the labour requirements. Overall, the
use of 10 cm. contouring will result in cheaper and more efficient
irrigation.
It is hoped that concessionaires will be appointed before their concession iB developed, so that they may be consulted cn••
i
sites, layouts and their requirements incorporated as far as good engineering practice will allow*4.10        Settlement Villages and Housing
Settlement Villages:
As previously referred to- in Section 3.8, further research is necessary to determine the optimum sizes, locations, layouts and composition cf the Afar Settle
ment Villages.
Detailed consideration and re-appraisal is also necessary in respect of the layout and construction 
of their houses^
Todate AV/i has budgeted Eth.152,400 per family house and has produced four designs for such houses. Three designs are for two room houses and one for
three rooms; they are all provided with 8 cm. concrete floor slabs, blockwark walls manufactured from local scoria and river aand, and simple uninsulated corrugated iron roofs. Each family house unit includes a separate cooking enclosure(6 sq.metr) and an enclosed latrine
with a common ZSesspool for every two houses.
Water should be providedat properly constructed communal water points .
The AVA has been planning the construction of seven trial family houses, tut as yet no construction
work has been commenced.
Housing:
At nmibara two ’A  type (2 bedroom) and eight
’ E’ type houses have been erected as staff accommodation These are based on designs prepared in 1971. However, concern has been expressed regarding the adequacy of
the accommodation provided by these houses, and their ability to attract and retain supervisory staft
14.10        Settlement Villages and Housing
Settlement Villages:
As previously referred tc in Section 3*8, further research is necessary to determine the optimum sizes, locations, layouts and composition of the Afar Settle ment Villages.
Detailed consideration and re-appraisal is also necessary in respect of the layout and construction 
of their houses.
Todate A
VZi has budgeted
Eth.32,400
house
and has
 produced four designs for such houses
Three
designs
 are for
two room houses and
three
rooms;
they      are      all      provided      with 
floor
slabs,
blockwark walls manufactured
per family
one for
8      cm      cone rete from local
scoria        and        river        s^xnd        ,        and        simple        uninsulated        corrugated 
iron roofs. Bach family house unit includes
cooking enclosune(6 sq metr) and an enclosed
o
with a common Cesspool for every two houses
Water should be providedat properly constructed communal water points.
The AVA has been planning the cons truetion of
a separate latrine
seven trial family houses, tut
work has been commenced.
Housing:
At Amibara two 'A' type (2 bedroom)
as yet no
cons true tion.
type houses These are based
a a v e been erected as
and eight staff accommodation
on designs             prepared          in 1971.              However,
concern has             been expressed
regarding            the adequacy of
the         accommodation         provided         by ability to attract and retain
these              houses, supervisory staff.
and.theirconcerned, with both the project management and the
farming
It is therefore proposed to recommend the adoption 
of new designs for enlarged ’Type B’ house. Sketch 
plans for the two and the three bedroom ’Type A ’ have
been prepared, and from information available, it is
anticipated that their basic costs will conform to the
budget included in the IBRD draft appraisal report of
1971,       Sketch       plans       for       an       enlarged       ’Type       B’       house       will be prepared in the near future.
1
The design and detailing of the
into consideration the local climatic
i.e insulation, shading, orientation,
dust and insects etc, It is hoped that the areas for living accommodation, will be improved, although they will still not be lavish by any standard. The construction will be largely determined by financial limitations and the dictates of available local materials, i.e concrete blocks or stone, local tirabera, corrugated iron sheeting, etc.
houses        will        take conditions, 
protection fromSBCTION No               5
DESIGN PROGRAMME 5.1             GENERAL
5.2  DESIGN PERSONNEL5« DESIGN PECGRARMB
5.1 General
The design programme is 6hown in bar chart fcr^n on Fig - 4.* It may be viewed as being in four stages.
Stage 1. Review of data and finalization cf project layout and basis of design; (this report summarizes this stage which is expected tc be generally complete during October)
Stage 2. Drafting of the contract documents and drawings for the preliminary works; (this is already 
in hand - most of the documents and drawings should be in final draft during October),
Stage 3. Preparation of the main contract documents
and drawings; (this overlaps with stage 2 and will take about 4 months).
Stage 4, Preparation cf working drawings; (this follows stage 3 and will include any modifications arising out of the preliminary findings of
the Angelele Eclhamc study).
It is intended that all the work should be carried out in the Addis Office except fur the design cf the barrage, and the finalization (for AVA/IEKD approval)
cf the technical specification; these will both be Gene
in London whre access to wider facilities is available.
It is proposed that the printing of all tender documents be done in Addis, though these for the main contract could be dene in London if Addis facilities
are found to be limited.
It is intended that the >S tudy Team for Angelele —Bolhamo should report their preliminary findings to the AVA in April 1974, so that the Amibara drawings can be modified by June, when tenders for the main contract are expected* Such modifications (and any resulting from site investigations) would then be incorporated in the main contract before award .
5 . 2 Design Personnel
A programme of Halcrcwe expatriate staff is
6hown on Fig. : 6 .
This updates that included in
the consultancy agreement, to which miner modifications have been made; the chart is restricted tc staff
employed on Amibara and Melka Sadi (i,e. excluding the Angelele Eclhamo study team).
It should be noted that two engineers from the present team will be transferred tc the study at the
end of October - though they may still work part-tire
on the Amibara project. A further member of the team
will transfer to 6ite early in 1974.
It is proposed to continue the present weekly technical meetings between the AVA and Halcrow engineering personnel.6.1 LATO
UR
SBCTION No. 6
USE OF LOCAL LABOUR AVAILAEILITY, COSTS AND
OUTPUT
6.2
RECOMM
ENDED DEGKBE OF LABOUR INTENSIV
ENESSUSE OF LOCAL LA13OUJR
6,1 Labour availability, costs, and output
It is clearly desirable tructed using labour intensive methods
that the works be cons- so far as this
is practicable, bearing in also be the most economical, costs and operating costs.
methods should be employed, of the likely availability,
mind that the works should 
having regard both to first
In considering the extent to which labour-intensive examination         has         been         made
cost, and output of labour
for
the Amibara work
6.11        Labour availability: There are a number
of construction projects
currently in progress, or nearing completion, in the
Middle Awash  Valley.  These include the MAESCO irrigation development,  the construction  of the Awash-&ille highway, and the AVA Farm development.
The majority of construction workers employed on these projects are highlanders although some local Afars are employed as guards and unskilled labourers. There appears to have been no difficulty in recuiting adequate numbers of labour fur these projects, although semi-skilled and skilled labour have mostly had to bo trained on the job. From discussions with various employers of labour in the Awash Valley, indications
are that there should be no difficulty in recruiting sufficient labour for the Zwnibara works..
6.12         Labour costs and Data has be
output:
on collected on labour costs andoutput for comparable works
Valley including works
carried out in the Awash at the MABSCO and N.itchell Cotts
developments*
Although it is not possible to relate
these costs specifically to 
do indicate that the excavation by hand should be competitive and the construction of parts of
the i.mibara works,
they
o
f the smaller canals
with 
structures in masonry, should construction in concrete.
be
This
machine excavation, some of the smaller competitive with
is borne out by
experience elsewhere, out by others.
and by
investigations carried
6 .2       Recommended degree of labour inlensiveness
6.21 Local Contracts:
In general there is sufficient incentive for local contractors to make maximum economic use of labour, e.g high cost cf imported plant. Thus, it is not
proposed to take any particular steps in the preparation of local contracts to encourage the use of labour.
6.22 The main' contract
Since the main contract will be carried out by an international contractor carrying heavy overheads (SO - 100% on prime cost), it is likely that the exten sive use of labour, in preference to mechanical plant, throughout the works could result in an increase in 
the total cost of the work, as illustrated in thetable below:
Time of construction Cost of labour
Normal-Contract 2 years
20
L-I contract 3 years
40
" ” plant 30 10
” ” materials 10 10
Prime cost
60
60
Overheads ( 10 4- 10 pa )                                     40                                        50
Total Cost
1OO_
iio=
It is therefore recommended that labour intensive methods be promoted primarily in those parts of the works where they are likely to be competitive, namely the excavation of rmaUer ^canals and the construction, of minor structures.
It is suggested that, in the documents issued to interna tional tenderers, the use of mechanical plant be prohibited for the following works which could be carried out economically by hand: (a) excavation of the secondary and tertiary canals, and (b) trimming 
of the excavation for the main and primary canals. 
In order to assess the effect on the tender price, tenderers would be invited to submit alternative rates for these operations using mechanical plant. Similarly, for specified parts of minor structures, tenderers wculd be asked to submit alternative rates for construction in masonry, and construction in concrete. On receipt of tenders, it would be possible to assess the effect
on the tender price of the different methods, the AVA remaining free to accept those methods considered appropriate.In addition to the above direct methods of promot ing the use of labour, there are a number of indirect
methods which may be considered. These indirect methods would have effect only where the difference in cost
between labour-intensive and plant—intensive methods
is marginal, and should not appreciably increase the
total cost of the works. These methods may be grouped 
under two headings:-
Sutsidies on Labour
(a)        Reimbursement to the contractor cf increases in the cost of labour.
(b)       Direct subsidies paid to the contractor per capita.employed.
(c)       Indirect subsidies such as the prevision of a labour camp.
Restrictions cn plant:
(d)
The limitation of advances tc the contractor
on account of plant.
(e)
Requirement for the Contractor to pay duty on all imported plant, instead cf duty-free
importation as is often permitted.
(f)
No reimbursement in respect of increased costs relating to plant.
As
regards the subsidies on labour,
it is proposed
that the contractor be reimbursed the majority of the increased cost resulting from rises in labour rates
during the construction*.
Further, it is proposed that
thi6  be extended  tc include employed,  not just workmen, cf these increased costs be
all Ethiopian personnel
It is suggested that 90% reimbursed tc the contractor.A per-capita subsidy is not recommended due tc possible abuse by the contractor and probable socio political problems. It is net considered that the complications resulting from the direct prevision, and running, of a labour camp by AVA would be justified but rather that the contractor be called upon tc provide, and run, such a camp to specified standards.
It is proposed that the discouragements on use of plant outlined above are incorporated in the contracts i.e. no plant advances, and no reimbursement of increased fuel costs. Further, ccntractcrs should be required 
to pay normal import duties, though the sums concerned should be refunded to AVA by the customs so that they do not become a burden cn the ID/1 credit.SECTION No. 7
METHOD OF IMPLEMENTATION
7.1  DIVISION OF PROJECT INTO CONTRACTS
7.2             PRELIMINARY WORKS
7.3             MAIN WORKS
7.4             SUPERVISION7. KBTHCD OF IMPLEMENTATION
7 . i Division of the Project works into Separate Contracts
7.11 Division of construction work;
In order to meet the requirement that the project works shall, as far as possible, be open to international tendering, it is essential that a substantial part of
the work be offered as a single contract, Only by so 
doing is it considered that sufficient interest may 
be generated amongst reputable international contractors capable of carrying out the works satisfactorily, and 
an adequate number of competitive tenders obtained »
There are however parts of the project works which could, with advantage, be carried out in advance of
the main works. These preliminary works include the construction cf a proposed link road from the Awash-Mille highway to Melka Harer, site investigations including exploratory drilling, construction of housing and 
office accommodation at Amitara and Melka Warer, and the first stage cf bush clearing.
The remaining project works, comprising barrage construction, canal works, minor structures, dyke6 access roads, land clearance, and the remaining bush clearing, could be incorporated in a single contract, and it is
this contract that would be offered to international tendering,
The diverse nature of the preliminary works makes them unsuited to a 6ingle contract, and separate contracts are proposed. Thus the project works would be divided into the fcllcwing contracts:—Preliminary and Miscellaneous Works
Bush Clearing
-
Contract
1A
Site Investigations
-
Contract
IE
Improvement to access
-
Contract
1C
Link Road
-
Contract
ID
River Cleance
-
Contract
LB
Buildings and Services
Settlers Housing
-
Contract
2A
Settlers Villages
-
fl
2E
AVA Housing - Stage 1
-
It
2C
AVA Housing - Stage 2
-
11
2D
Services
-
If
2B etc
Main Work6
Main Works Contract
-
Contract
3
Barrage Gates 7.12 Types cf contract
-
Sub-Contract 3A
The   types of contract proposed for each cf the
works are   outlined in App. F. The different forms envisaged   are:
(a)      Force account, to cover improvement to existing access, and some bush clearing.
(b)     Competitive tenders, by local contractors, 
to cover the link road, all buildings and services, bush and river clearing*
These contracts will be based upon Ethiopian government departments conditions of contract, or those for minor works (xlpp< G) as appropri ate.
(c) Competitive tenders, by internationalcontracto
rs, 
for
the
site
investiga
tions
and 
th 
main 
works. 
These
contracts
will
te
based 
upon 
the international FIDIC conditions of contract.
Generally, the IEkD guidelines for procurement will be followed. The procedure for each contract is outlined in Appendix F.
7.2        Preliminary Works Contracts
7.21        Lush Clearing:
In order to minimize disturbance to grazing lands within the project area, it is considered advisable
to phase the bush clearing with the implementation cf the main project wcrks. However, it will be necessary 
to carry out some bush clearing in advance cf the main works to permit survey work to be carried out. This preliminary clearing includes the general clearance
of the area comprising phase I of the main project works, and clearance along the lines of the primary and secondary canals within the phase II and III areas.
The clearing within the phase I area amounts to some 2,200 hectares^ cf which some 1.300 hectares is is light bush which, it is believed, could be readily cleared without the use cf mechanical plant. Some
900 hectares is medium tush, which includes some large trees. It is considered that mechanical plant would 
be required to supplement hand labour to clear the medium bush. The remainder of the Phase I area has little or no bush, to Le cleared «
The      clearing      along      the      lines      of      the      main      and secondary  canals in  the Phase II and  Ill areas includes medium and  heavy  bush,  for which  mechanical plant.would be required •
7.22        Site Investigations (Drilling):•
In order to prepare the designs and ccntract documents for the main Amibara works it is necessary to carry out site investigation work, including boring, sampling, test pumping, and soil testing. These investi gations are required to be undertaken for the design of *
i) Hydraulic structures
ii ) Road bridges
iii) /access roads
iv ) Canals
v) Dykes
vi ) Water supply
Similar investigations are required tc be carried out for the study of the Angelele Eolhamo areas and would be included in the site investigations contract.
7.23        Link Road:
The proposed rcau will provide access from the Awash-N.ille highway. However, it will take seme menths to construct and it is desirable tsr JurFprbve access
-
immediately for contractors carrying out the surveys preliminary contracts and investigations.
To achieve this it is proposed that access be provided in two stages. The first stage would involve
up—grading the existing track and the construction of
a minimum standard access, by means of a foi'ce account (up to BS2O,OOO). The second stage would be theconstruction of an all-weather road to a standard suit able for construction traffic, by a measured works , contract.
These works could comprise contracts IC and ID. 
The main works contractor would be required to take
over the all-weather road at the commencement cf the main contract, and to maintain it for the duration 
of the contract.
7,24       Luildings and Services;
The building works to be carried out comprise housing for settlers, and housing and offices for
AVA and Halcrow site staff, and for AVA admin, staff.
It is propcsed that the settlers housing be constructed in two stages. The first stage would be
a pilot scheme to confirm the type and layout of
buildings best suited to the needs of the Afar settlers
The pilot scheme would be constructed near Arnib^ra and the numbers and types of dwellings would be as approved by the AVA. This would comprise contract 2A and will
be carried out'assoon as possible.
Contract 2E would comprise the construction of settlers housing, in 3 or 4 villages, and would provide accommodation for the remainder of the settlers and 
their families. This contract will be phased to suit
the rate of settlement. The type and layout of the
dwellings for contract 2E would be determined on the results of the pilot scheme. The provision of water
supply and sewage disposal will be planned by the
time contract 21 is implemented and will be -xi.ncJ.udcd in that contract.It is envisaged offices foe AVX and out in two stages, 
the construction of AVA and Halcrow site as soon as possible.
that the construction of housing and
Halcrow staff would also be carried
The         first         stage,         contract         2C,would         be houses at Melka Warer to accommodate
staff *           These houses must be
provided
The second stage, contract 2D, would include the construction of further housing and an office etc. at
Melka Warer, to accommodate further staff.
It is al6O envisaged that the main works contractor would site his plant yard and stores buildings at frelka Warer, and that these will be available for the AVA
project administration after completion of the main works contract.
The prevision of cervices at Belka Warer, including water and electricity supplies, would be arranged through contracts running concurrently with the buildings contract.
7.3 Main Works
It is envisaged that the
comprise the construction of:
Barrage and headworks at Melka Sadi Main and primary supply canals Primary drains
rr.ainworks contract would
Secondary and tertiary supply and drainage canals Control structures
Access roads and bridges Flood protection works Lush clearanceadministrative, organizational and financial aspects. Such a committee would meet monthly in Addis and, it is suggested, should be attended by the AVa Project Director, their Chief Engineer and the UNDP Project Director in addition tc Halcrow’s Project Manager.
Other senior staff would attend as called for by the agenda.
It is suggested that the AVA Project Director should not become involved in the day-to-day problems, but keep in mind the broader aspects of the development; he would act as the main link between AVA and Halcrcfc. On the other hand, trainee oorMterparts to senior Halcrow staff would be deeply involved in the day-to-day problems^
7.42 Site Supervision:
The supervision of construction will be carried out by the Resident Engineer and his staff which would eventually comprise three expatriate engineers, four Ethiopian engineers and two expatriate inspectors supported by the necessary surveyors, draughtsmen, clerks inspectors, and checkers.
During the next twelve months, the Resident Engineer, supported by one expatriate inspector and local staff, would supervise the preliminaryyworks. 
His staff would be increaed on the award of the main contract, as shown on the staff programme, fig. 6.
Though 6ix expatriates are shown on site during the construction, it is hoped that it will be possible to carry out adequate supervision with four, (three engineers and a chief inspector), the total number ofKough land levelling
Miner associated works
The contractor would be required to arrange his work to give completion in three phases. Phase I would comprise of the completion cf the barrage and n.ain canal works, and the completion cf all ether works for over about 4,200 hectares, including the AVA settlement farm area. Phase II would comprise the completion of an adjacent area cf approximately 4,400 hectares including the remaining existing concessionaires and Phase III
the remaining 1,500 hectares. lush clearing cf the
first area would have been completed under a preliminary contract, most of the remaining bush clearing being carried out under the main works contract as part cf
the phased completion .
River clearance and other minor works would be carried out concurrently with the main works contract.
To ensure effective supervision cf the implemen tation, close liaison between AVA and Halcrows must continue. To facilitate such liaison, and to minimize
delays and misunderstandings as the momentum of the
should be
extended
project increases, the In addition
being          held hoped that
between AVA a liaison
present arrangements
to the weekly design meetings and Halcrow engineers, it is
committee can be set up to review
the broader aspects of implementation, including thesupervising staff remaining unaltered. However, the final decision must depend on the quality of supervision by the contractor’s staff, and on the experience and competence of the AV/i staff.SECTION No. IMFLEKENTATION PROGRAMME
8.1 PROGRAMME OF WORKS8,. IN PLENLENTATION PKCGRAMiE
8,1 Programme of Works
8.11 General:
The proposed programme for implementation is shown in bar chart form on Fig. No. 5. The main criteria
leading to these proposals are:
(a)
The earliest realistic availability of
irrigable land
(b)
The employment of an international contractor through competitive tendering, tc carry out
, 
the bulk of the works.
(c)
1
The optimum utilization of the ’’design ’
period for preliminary works on the site.
(d)
Adequate provisions for the existing land users, i.e. the Afars..
Having considered the restrainte.it is thought practicable to supply water to the Melka Sadi area by June 1976 and, by the same date, to have some 4,200 ha. of irrigated land available in Amibara. Irrigation 
to the remaining 5,800 hectares would be made available over the following 12 months, allowing 10,000 hectares of cotton to be planted in 1977.
Experience indicates that about a year is needed to let a large contract to an international contractor
(4m. design and contract drifting, 2m. for approval and printing, 3m. for tendering and 3m. to award). It is proposed to utilize this period for various preliminary works - access, clearance, housing - and data collection - topographical surveys and drilling.8.13 Main Contract:
The principle feature of the main contract is the proposed phasing of the development,. The works on the concessions (’’farm works") would not be started until 6-9 months before the land is made available
to concessionaires. Such works include clearance, survey, design layout, land planing and construction of tertiary supply ana drainage canals and farm roads.
Attention is also drawn to the proposed early award of a nominated sub-contract for the barrage gates tc ensure an adequate delivery period.8.12 Preliminary works
Particular attention is drawn to the following
(a) Access to site:
Since a contract for the link road under competitive tender is unlikely to be awarded until early 1974, with the road available for use in the second 
half of next year, it is proposed thatja limited amount of work be undertaken this year to improve access to the site for the preliminary works contractors »
(b) Push clearance
To retain grazing for the Afars until other provisions can be made for them, and to minimize problems from dust, it is proposed that the site
be only partially cleared of bushes in the preliminary period — the clearing cf many of the concessions beinj dene under the main contract.
(C)        Site investigations
The drilling
contract is
(drilling )
indicated as starting
in March          1975 which means that the data obtained would be too late to incorporate in the tender documents (and would also extend the period for
the Angelele Eclhamc study which starts this year). It is thus recommended that means of short-circuit ing the AVA/IERD procedures be discussed to allow a start to drilling in January 1974, e.g immediate invitation to tender or, quicker, negotiate a
contract with a competent international firm working in Ethiopia .SECTION 9 DECISIONS REQUIRED FROM AVA
9.1         1GENERAL APPROVAL OF PROPOSED LAYOUT
9.2          COMMENTS   ON  PRELIMINARY DESIGNS
9.3 APPROVAL OF IMPLEMENTATIOPNROPOSALS
9.4              MATTERS FOR    DISCUSS .ON9. DECISIONS REQUIRED FROM AVA
Ruport Reference
General Approval of Proposed Layout
9.11 One river diversion structure, not two.
9.12 Regional layout and infrastructure
9*13 Layout of canals, drains and dykes
9»14 Layout of concessions
Comments on Preliminary Designs and Approval of Criteria; 9,21 Irrigation water duties
9*22 Barrage and headworks
9»23 Floods and protective works
9,24 Irrigation works
9*25 Housing, settlement and miscellaneous
Approval or Comments on Implementation Prooosals
9.31 Comments on method of implementation
Preliminary works
Main works
9*32 Comments on proposed programme
9 «■ 33 Approval of list of contracts
9*34 Approval of ES20,000 expenditure by force account to improve Bite access
9»35 Approval to shorter tendoring procedure on site investigation contract
9*36 Approval of contract procedures
Matters for Discussion and Agreement
9*41 Arrangements for liaison between AVA and Halcrows
9.42 Confirmation of Periods allowed for AVA/lBRD approvals
9.43 Procedures for approvals by IBRD
Drg. No,
Text.
DR 1
DR 2
DR 3
DR 3,4
Fig. 1
DR 5
DR 7
DR 7,8,9
DR 10
—
-
-
Fig.5
-
-
-
—
Fig. 4
-
App. B
3.1,6.9,1c 3.3 - 3.5
3.1, 3.7
4.2
4.3
4.4
4.5,4.9
4.10
7.1
7.2
7.3
8
7.1
7.. 23
App. F.l App. F
7.41
-
-APPENDIX a
KX IS TING LEVELOHENTS IN THE AKIDARA PROJECT IgA Much of the iMd is currently used by the hfars to graze their
livestock*
The following is a brief description of the existing development
a.     Melka Harer Research Station has been established since 1966 and now occupies some 220 hectares adjacent to the Awash River, (See Drawing DR.2), It is one of five research 
stations set up by the Institute for Agricultural Research 
to study crop growth in different climatic regions of
Ethiopia. The area allocated to the Research Station within 
the Project area is 400 hectares. It is hoped that it will
be able to act as the research station for the developments
at Amibara, Bolhamo and Angelele. Labour employed at the Research Station now lives in Mclka Warer Village.
b.     AVA’s first settlement farm in the project area covers 125* hectaiBB adjacent to the Awash River downstream of the Me 1kg. Warer Research Station. The eastern boundary of the farm is
the proposed alignment of the primary canal of the Amibara Project. The farm headquarters are situated near -Rot ri Pharr, Hill. To date, over 200 Afar families have been settled on the farm. A start has recently been made on the first 250 hectares of the second settlement farm which will cover about
1000 hectaros between the primary canal alignment and the eastern boundary of the project. The proposed location of the farm headquarters is shown on Drawing DR.2. The two farms will in due course provide 2.5 hectares for each of
S00 Afar families.c.       Algeta Concession is situated adjacent to the Awash River downstream of AVA’s No. 1 farm. Its area is 660 hectares.
d.      Downstream of Algeta, Ambash Concession covers SCO hectares*
1
c. The last in the ’string  of existing concessions adjacent to 
the Awash River, is the Sublalo Concession which covers 320 hectares•
£11 five existing developments aro contiguous and between them occupy nearly all the available land with a frontage on the Awash River. Frontage for only one concossion of 320 hectares remains. The main crop grown on the AVA farm and the throe concessions is cotton. There is, in effect, a sixth development in the Amibora Project area, which is the present occupation and use of much of the land, not described abovo, by Afars and their livestock. Work is now in hand to provide water in the Alledcghi Plains to enable the Afars to keep their livestock there, but until this is completed tho Afars will continue to use the Amibara Project area.3
and daily supply* Preliminary considerations indicate that the project should be capable of storing at least 6 hours of average demand flow
3
(20 cumecs) i.o. some 500,000 m . It is proposed that this pondage
be provided in two places, some 100,000m  in the headpond formed by 
3
the barrage and up to 5OO,OOOm  in tho main canal upstream cf Amibara.
It is now considered that the small dry weather flows in the Kcsscm and Kebena Rivers will be fully utilized by the irrigation developments on the banks of these rivers, and therefore that only 
one river diversion structure is required for the implementation of
the Melka Sadi, Amibara and Angelele areas. This structure would also be able to supply the Bolhamo area if required. The major factors determining thi6 decisions are:
i. Saving in costT
ii« Alleviation of probable drainage problems at the northern end of Melka Sadi,
iii« No impedance to flood flows from Kcsscm and Kebena Rivers.
iv.        Simplification, and improvement, of water control.
v.          Unfavourable site conditions for a river diversion structure at Melka Harer,
vi.        The avoidance of disrupting the built-up area at the
Melka Harer Research Station by the conRinictiop of
the main canal from the Melka Waror Barrage.
Potential problems do arise from the adoption of one diversion structure, but these can all be overcome. They are:
vii* Decrease in pondage. This has been discussed above.
viii. Dry Beason flows in Kessem and Kebena Rivers will not
be acquired.
Division of responsibility and costs between MAESCO and AVA-.APPENDIX B
CONSIDERATIONS _0F ABTRACTION OF WATER  BY . N&JJS OF ONE OR TWO DIVERSION STRUClTJRES
As proposed by Italconeult, the project comprised two contiguous areas on the right bank of the Awash River the Melka
Sadi area - to be supplied by gravity from its own diversion weir -
and Amibara, an area of 12,000 hectares to be supplied by gravity from a second diversion weir at Melka Warcr (see Drawing MSA/DR.l)« The reasons givon for having two diversion structures were the need 
to provide some pondage upstream of Melka Warer to even cut the varia— tions in flow in the Awash River and to ’’catch” the dry weather flows from the Kessem and Kebena Rivers that join the Awash between the two diversion sites.. It was later appreciated that the creation cf a
reservoir at Melka Warcr might impede drainage and cause potential difficulties with ground water control in the northern part of the Melka Sadi area but the decision to construct two diversion structures was not changed.
On reflection and study, the diversion structure at Melka Warer seemed to serve no purpose that could not be fulfilled by a single structure at Melka Warer. Pondage of the river flow would not appear necessary for operation of the Amibara and Melka Sadi projects, since the flow should always exceed the abstraction at this part of the river. Nevertheless, it would seem advisable to provide some pondage to facilitate operation and avoid undesirable fluctuations in supply levels. . The topography cf the area rules out the economic provision of anything greater than daily pondage, i.e. • sufficient stored water to even out fluctuations in daily demandX. Supply canal to Bolhamo along left bank will net be
possible, ncr will road access to Bolhamo from a bridge
at Melka Warer.
Discussing these factors in more detail we have:
i.          Saving in Cost
The cost of the diversion structure proposed for Melka Warer, utilizing a site close to that proposed by Ita.lconsult but capable
of passing a flood of 2,000n?/sec, is estimated to be ES7«5 million, including canal headworks.
The cost of the earthworks for a main canal from the diversion structure at Melka Sadi to Melka Warer is estimated to be Ethb3»5 million (1.5 million cubic metres of earthworks; including haulage). 
In addition there will be the cost of the enlarged headworks at Melka. Sadi, a gated regulator, a drop and. a cross-dra.in structure at Melka
Warer, which together with other minor structures arc estimated to 
cost between Eth©l million and Eth$ 1,5 million. There will possibly 
be some auxiliary costs for diversion of roads and drains which could amount to Eth^O.25 million.
The bridge access to the left bank at Melka Warer will be
lost and will have to be replaced if Bolhamo is developed. A compensatory saving will derive because the potentially expensive
road on the left bank from Melka Warer to Bolhamo will new be replaced by a project read in Amibara. Indoed, in the final design, access to 
the Bolhamo area may possibly havo been re-located downstream to avoid the tortuous route from Melka Warer. Tho additional cost to provide
a bridge might be EthOO.25 million.Thus the total cost resulting from the supply of water to Amibara from Melka Sadi is estimated at Eth^5«5 million compared with an estimated cost of Eth;7.5 million if supplied by diversion at Melka Warer, giving an estimated saving of Etho 2 million.
In addition one structure would lead to an earlier supply of water to the whole project.
ii-      Alleviation of probable drainage
problems in Melka Sadi
The ground levels in the area where Melka Sadi and Amibara meet are as low as 736m- which is 2m. or more below operationa.l retention level with a diversion structure at Melka Rarer. Water
logging and a possible rise in ground water in Melka Sadi could result with all the attendant problems and remedial expenses.
iii.        No impedance to floods: Self-explanatory.
iv.   Simplification of water control
Two river diversion structures would demand a high order liaison between the two sites. For example  should the gates
t
be lowered at Melka Sadi, rapid appropriate aotion would be required at Melka Warer to avoid operational problems.
v,   Foundation conditions at Melka Warer
Faulting and non-uniform strata are acknowledged problems at this site. They could lead to increases in cost during construction 
to cater for unforeseen foundation conditions.vir Disruption at Molka Harer Research Station
The primary canal from a diversion structure at the proposed site would run through the built-up (houses, offices, laboratories
etc.) area of the Research Station, which would have to be re-located.
L possible alternative route for the canal through the Melka Warer plateau would be very costly.
vii.       Pondage: discussed above.
viii.      Dry Season Plows: discussed above.
ix.        MAESCO and AVA
This is not a real problem but is mentioned as a factor which will require discussion and perhaps some re—negotiation. There should be a Baving in cost to MAESO and as it has been recommended that AVA should retain both ownership and operational control cf all Awash 
River structures, the proposed single structure will bo consistent with probable future plans.
x« Left bonk access to Bolhamo
The abondonment cf the Melka Harer diversion structure precludes the use cf the left bank for access and water supply to Bolhamo. However, the terrain between Melka Harer and Bolhamo is very undulating (foothills of Mount Iofan) and cost end unsuitability would probably have resulted in alternative routoa bolas used.APPENDIX C
EFFECT ON AMIBARA DESIBN OF DEVELOF^IENT AT ZiNGELELE AND BOLHAMO For conveniencc  the tender documents for Amibara will be
T
prepared assuming no development at Angelele and Bolhamo. However, 
the likely effect of these two developments will bo indicated on 
drawings and covered by the contract. Final designs will be prepared 
based on findings of the Interim Report of the Angclele — Bolhamo
Feasibility Study®
The modifications to the design of Amibara should these two areas be developed are aB follows. They assume a single river diversion structure at Melka Sadi and a syphon to carry the supply for Bolhamo across
the Awash.
i.          The capacity of the Melka Sadi headworks and of the main canal will be increased by about 17 cumecs.
ii.      The capacity of the Amibara primary canal upstream of the off—take to Bolhamo will similarly be increased by abouu 17 cumecs.
iii.       The capacity of the Amibara primary canal downstream of the off-take to Bolhamo will be increased by 6 cumecs to cater for Angelele.
iv. An 11 cumec primary canal for Bolhamo will be required from the Amibara primary canal to the syphon across the AwaBh River.
An oxtenBion of the Amibara main road system to a bridge across the Awash River will be required to provido access to Bolhamo. Access to Angelele would by a continuation of the Amibara road system*
The design of the Amibara flood
protection        dyke will take
into    account    a    similar    dyke    on    the    loft    bank    to    protect Bolhamo.APPENDIX D
HBSIGN OF WI&TEKN DYKE Tho alignment of I he wet bexn dyke
been chosen to take
account of existing developments in tho Melka Sadi and wnibara areas, and the revised soils mapping. It differs from Italconsult
proposals. Furthor crosseections of the Awash - generally at 1 km. intervals - are available, and they will prove useful in estimating 
river Btagds under flood conditions•
Italcon6ult used values for roughness equivalent to Manning ’n’ values of 0*05 for the main river channel and C.2C for tho flood 
plain* The main channel value is in line with recommendations by Chow (Open Channel Hydraulics). The flood plain figure however appears conservative, and may be reviewed in tho light of further site inspections# In addition clearing of the river flood plain, (not however adjacent to the dykes) could justify reducing this figure to about 0.10. This would result in doubling the flood plain flow for a given river stage. Thus
the dyke height may be reduced by as much as 1 metre. It is not envisaged that tho cost of this flood plain clearing will be appreciable, as the
larger trees could probably bo sold to a contractor, and the remaining bush could be given to the settlers as a concession on the undertaking that they would remove all growth above a certain size.APPENDIX E
Possible Interpretation of GroundwgteJ Situation
The groundwater situation in the Melkn. Sadi and Amioara areas
is no’-, clear. Reference has been made in several reports to the existing grouncwater tables and to their possible future rise in level causing 
problems of salinity, and saturation.
However, to date there is no direct evidence of a water table
per bc. Boreholes through the silty loam subsoil appear tc stay dry 
until they strike an aquifer at between 15 and 40 metres. This aquifer
is under pressure and the water rises immediately to about 8 to 10 m. 
below ground level. It is this water level which appears to be called 
a water table and its fluctuations noted in the somewhat limited records. 
Ko specific aquiclude has been identified which would cause artesian conditions but it could bc the silty loams themselves, which are sufficient impervious to be the aquiclude.
If this is the case, then thcBQ silty loams are behaving ISkn 
a clay as far as vertical permeability is concerned (very possible with 
4
permeabilities of the ordor of io" cm./sec) and therefore no water table
could exist and movement of salt would net tako place. This would also 
mean that tile drainage would be ineffective except to drain small
deposits cf a moro permeable nature.
Piezometers    in    a    clay    will    in    time    fill    with    water water     and     will     register     fluctuations     in     pore     water     pressure, levels do not of course, represent a true water table.
from the pore
These waterAPPENDIX F
TYPES OF CONTRACT
;*
1.        SITE INVESTIGATIONS
2.        LINK ROAD
3.        BUSH CLEARING
4.        SETTLER’S HOUSING
5.  BUILDINGS AND SERVICES
6.        MAIN CONTRACTtypes of contract
1 # CONTRACT FOR SITE INVESTIGATIONS
Tendering Procedure:
In view of the requirement that this contract, unlike the
other preliminary workd contracts, is to he open to international
tendering, the procedures for tendering world be in accordance with 
the IBRD "Guidelines for the Procurement under would Bank Loans
and IDA Credits.”
However, in view of the need to have the site investigations
work under way by January 1974» it is suggested that the procedures
for advertising, prequalification, and submission of tenders should 
be substantially curtailed.
The most expeditious method of letting this work would be
to negotiate directly with a contractor who is known to be competent
to carry cut the work, is established in Ethiopia, and able to 
complete the work within the required time.
Should this procedure be unacceptable, it is suggested that tender documents bo issued to no more than three selected tenderors, 
who would havo confirmed their interest in submitting a tender, and 
who would bo approved by AVA. and IBRD. The time for submission of tenders would be 3 weeks from the issue cf tender documents, and. 
the successful tenderer would bo required to commence the work within
2 weeks of acceptance of his tender.
A decision cn the procedure to be adopted is urgently required QB the preparation of the tender documents is new in hand.Form of Tender Document
It is envisaged that the tender documents will comprise:- Instructions to Tenderers
Description of works
Forms of Agreement and Bond
General Conditions of Contract
Particular Conditions of Contract
Specifications - General and Particular
Bills of Quantities, with preamble and Summary Form of Tender
Appendices
Drawings showing the Location of the work
The forms of tender, agreement, and the general conditions
of contract would be the forms and conditions prepared by the Federation International Des Ingonieurs — Ccnseils (FIDIC) for works
of oivil engineering construction#
The particular conditions cf contract and the specification 
are being prepared to take account of the particular requirements
of the AVA and cf the work to be carried cut.
The Bills of Quantities are being prepared generally in 
accordance with the Standard Method of Measurement cf Civil Engineering Quant it ies, published by the Institution cf Civil Engineers  London.
t
Baymdnt to the contractor would be certified monthly cn the baBio Of the quantities of work executed and the rates in the Bills
•Tuantitice, subject to a retonticn cf 5% payable on completion
°f the whole of the work. Final payment would be based cn measurement the completed work.
It iB proposed that the contractor be roquired tc provide formanoe Bond equal to 10% of the tender sum. A tender bend is
1  considered necessary if the proposed method of procuement iB adopted2. CONTRACTS FOR LINK ROAD
2,i Improvement to Accesst
It is proposed that work cn the link road be carried out
in two stages. The first stage, comprising upgrading the existing 
access, would be carried cut under a force account contract arranged 
in accordance with AVA practice#
2.2 Tenders for Link Road Construction
The tender documents for the second stage would comprisei Instructions to tenderers
Description of works
Ferm of tender
Conditions of Contract
Specification
Bill cf Quantities
Drawings shewing the works to be executed*
The form off tender and General Conditions cf Contract would be based on the General Conditions of Contract for Minor Works, a
copy of which is included as Appendix G to this Report. Particular conditions of contract wculd be included to take account of particular requirements of the AV A and of the work to be carried cut.
The specification would define the form of construction, materials, workmanship, and limitations on methods to bo employed in carrying out the work. The bill of Quantities would provide for a Bimplifiea method of measurement, e.g. a rate per linear metro of road construction, with additional items fcr earthworks and culverts.Payment to the contractor would bo certified monthly on 
the estimated value cf work dene, subject tc a retention cf 5%* 
A performance bond equal to 10% of the tender sum would be required. Variations in price of labour and materials would not be reimbursed*
A tender notice would be published locally at least two 
weeks before the invitation cf tenders, and interested contractors
would be required to submit information to demonstrate their ability 
to carry cut the work. On the basis cf the information submitted, 
a list cf selected tenders would be submitted tc AVA for approval*
Tenderers would be required to visit the site and to submit tenders within 3 weeks of the invitation. The successful tenderer
would be required to commence work within two weeks cf the acceptance of his tender®
3. CONTRACT FOR BUSH CLEARING
It is envisaged that the documentation for this contract
will comprise:
Instructions to Tenderers
Description cf Works
Form of Tender
Conditions cf contract
Specification
Bill of Quantities
Drawing Bhcwing extent of the works.
Tho form cf tondor and general conditions cf contract would bo based on the general conditions of contract for miner works, a
CO PY of which is included as Appendix G. The specification would ^ cscribo the requirements for oloaring, and limitations on the
the ds which may bo employed tc carry cut the work.Tho Bill of Quantities would provide fcr payment for three different types of bush clearing, i.e. light, medium and heavy bush, 
at rato6 submitted by the tenderer. The areas cf each type of bush 
to be cleared would be entered in the Bill, and would be as indicated 
in the field prior to tender. Payment to the contractor would be
certified monthly on the estimated value of work done, subject to a retention cf 5%, which would be payable on satisfactory completion 
of tho whole of the work. It is envisaged that a performance bond would net be required, and that variations in price of labour and materials would not be reimbursed.
A tender notice would be published locally as early as possible before the invitation cf tenders in order to assess the
level cf interest in this work which enquiries to date suggest is
limited. If there i6 an inadequate response to the tender notice
it may be necessary to carry cut the bush clearing by force account contract.
Tenderers would be required to visit the site, and the different types of bush would be indicated tc them. They would be required to submit their tenders within two weeks of the invitation, end they would be required tc submit a description of the methods
plant they propose to employ to carry out the works.
Selected tenderers would be interviewed to ascertain their ability to carry out the work satisfactorily. A report cn the
tenders received, and a recommendation for acceptance would be submitted to AVA.
The contract fcr river clearing would bo arranged in a similar anncr to tho bush olearing contract, the work being carried cut at a
later
of tho project.4. CONTRACTS FOR SETTLERS HOUSING
It is envisaged, that the tender documents for the construction of settlers housing, contracts 2A and ZB, would comprise:
Description cf Works
Form of Tender
Conditions cf Contract
Specification
Bill of Quantities
Drawings.
The form of tender and condlticcc of contract would be based on the General Condition of Contract for Minor Works, (sec appendix G) suitably modified to suit AVA requirements and of the work to be
oarried out. The specification and drawings are being prepared taking account of local building methods and materials. Tho bill of Quantities would provide for payment on the basis cf a fixed sum per dewellmg. Payment would be certified monthly cn the estimated value of works done subject to a retention of 5%» payable cn satisfactory completion 
of the work. A tender bond and performance bond are net considered necessary, and variations in price of labour and materials would not
be reimbursed. Liquidated damages would be payable for late completion.
A tender notice would be published locally about two weeks
before the issue cf tender documents. Prospective tenderers would be required to provide information to demonstrate their ability to carry out the work satisfactorily. Tenders would be invited from a selected list of contractors approved by AVA.
Tenderers would be required to visit tho site cf tho work and to submit their tenders within two or three weeks cf the invitation.Subject to the satisfactory performance of the contractor
for contract 2A  there may be seme advantage in negotiating with 
t
him fcr the construction cf the further housing under contract ZB.
CONTRACTS FORBUILDINGS AND SERVICES.
It is envisaged that procedures for these contracts will
be based on these employed by the Ministry of Public Works and Water Resources fcr similar works»
The tender documents would comprise:
Instructions to Tenderers
Description cf Works
Forms cf tender and agreement
Conditions cf contract
Specifications
Bill of Quantities
Drawings•
Completion dates fcr each contract would be fixed and liquidated, damages would be payable for late completion.
The specification end drawings fcr the buildings would, as far   aB possible, take account of local building ccnstructicn materials
and methods.
It is envisaged that the Dills cf Quantities would include lum^sxun items fcr each building complete, with measured items fcr
°undationB, and external works and sorvicos. Tenderers would be lc mircd to price a schedule cf Rates for use in the evaluation of
y variations to tho work which may be ordered*General conditions of contract
Particular conditions of contract
General specification
Particular specifications
Preamble to Bills of Qualities
Bill of Quantities and Summary
Schedules (to be completed by tenderers)
Form of Tender
Appendices.
The instructions to tenderers would includo information
such as the date, place and manner of submission of tenders, arran— 
Cements for site visits and the submission cf queries, dotails and 
check-list of documents to be submitted with the tender, arid any 
other matters tc which tenderers’ attention should be drawn.
The General Conditions of Contract would be the FlUlC
international conditions for civil engineering works. The particular
conditionswould include any amendments to, cr amplification cf, the
General Conditions to meet the requirements cf AVA and cf the work 
io be carried out.
The    specifications    would    describe    the    standards    cf    materials vid   workmanship,   to   be   used   in   the   works,   and   any   other   matters   rcla- iing to the manner in which the works were tc be carried out.
The Bills of Quantities would be prepared generally in6,1
Payment tc the contractor would be certified mcnthly on 
the estimated value of Work completed, subject to a retention of
5^> payable on completion. A performance bend equal tc 1C$ of the
tender sum would be required, variation in price of labcur and 
materials would not be paid.
A tender notice would be published Jccall.v at least two 
weeks before the invitation cf tenders, and interested contractors
would be required to submit information to deinc nstrate their ability 
to carry out the work satisfactorily» A list cf selected tenders
would be submitted to AVA for approvals
Tenderers weald bo required to visit the site and to submit their tenders within three weeks of the invitation. The successful
tenderer would be required tc commence the work within three weeks cf the acceptance cf his tender*
The contracts for services, including water supply and electricity would be arranged in a similar manner, having regard 
to appropriate local requirements and regulations.
6.         IUIN CONTRACT
of Contract
It is proposed that the main works contract be open to international tendering, and that the tender procedures and documen tation be generally in accordance with the IBRD “Guidlincs for
l* ccurement under World Bank Leans and IDA Credits.-’
r
The tenderer documents would ccmpriso: Instructions to tenderers Description works
FormB of Agreement and BondThe schedules would include:
a.      the Schedule cf Basic Prices of Labour and Materials
on which the tender was based,
b.      a list of sub-ccntracters to whom the main contractor
may propose to sublet parts cf the works and
c.      a li6t of alternative national standards.
The appendices would include items relevant to the tender document, 6uch as a tender programme form, list cf national standards etc.
It is proposed that tenderers be required to provide a tender bond equal to 5$ cf the tender sum, and. the appointed contractor
would be required to provide a performance bond equal 10$ of the tender sum.
Payment tc the contractor would be certified monthly subject to a retention of 5%, payable on completion, alternatively the cont ractor would be permitted to provide a Retention Guarantee.
In accordance with the recommendations in Section 6.2, advance payments in respect of plant provided for the works wculc. not be payable, but a general mobilization advance cf up tc 15a
ihe tender sum would be payable.
Liquidated damages would be payable by the contractor cn late completion cf any of phases I, II, or III. There would be no bonus for early completion.
Variations in cost of local personnel and of specific basic
m ateri i  would be reimbursable to xhe contractor, with the exception 
aB
°£ plant fuel as proposed in Section 6»2.Materials and equipment would be free of import duty, with the exception of mechanical plant as proposed in Section 6.2. The Ethiopian Government would arrange for refund, to AVa, cf the customs
duty paid by contractors on plant,
6.2     Tendering Procedures
A pre-qualification notice will be advertised in appropriate national and international journals and newspapers at least 60 days before the formal invitations to tenders. Copies of the notice will
be 6ent to appropriate Embassies and Government agencies. On the basiB cf the information submitted, a list cf selected tenderers will be submitted to AVA for approval.
Tenderers will be required tc visit site and tc submit their tenders within 90 days cf the issue of formal invitations to tender.
6.J Barrage gate Subcontract
The gates Tor the barrage «uld »e supplied and installed
under noninated sub-contract. BnguW
» 1U                s“' “ rC’"Ubl°
gate manufacturers in USA, Canada, Europe and Japan, and 
eeleoted to tender. It is envisaged that the order will be placed 
prior to the award of the main contract - the subcontract later being 
taken over by the main ccntractort
Modifications from Angelele-Bolhamo Development
The main contract will include those works required tc irrigate
the Amibara
area. Possible extensions cf these works to cater for deve-
.          , x
lopment of Angelele and Bclhamo ( / i hpa  c.g. ldweXeorks and main canal)
c
will bo indicated i
Bill of Quantities,
cf
in the contract and would be covered by a separate which will bo priced by tendorers. Before the award 
the
in the
of the
contract, the extensions will cithor be deleted or incorporated centract according to the decision following the interim report ^olele-Bolhamc study.APPENDIX G (coni’d)
TENDER FOR
Date
•••••••••••••«••
Vie offer to  undertake and  complete this Contract for the sum of *>•••••••• 
...................................................................  (£>...•) (or such other sum as may be ascertained in accordance with the General Conditions of Contract)®
Vie will complete the Works wjthin •••••«••>•••»•••••••••«• weeks and we agree to the following.
Defects Liability Period Third Party Insurance
Rate of liquidated damages
••••«•■•«•>•••••• weeks
£ ................................. for any
one aooident £per week
Thls ‘tender is valid for ten weeks.
Yours faithfully,APPENDIX C
GENERAL CONDITIONS OF CONTRACT FOR MINOR WORKS
CONTENTS
Form of tender
CI.1.            Definitions
2.          The Contractor’s responsibilities
3.           Liabilities          and     Insurances
4.           Progress of        the        work 5«         Payment for       the        work 6.          Variations of the work
7.          Adverse conditions 8.         Expulsion
9.         Disputes
Sir William Halcrow 4 Partners Newcombe House
45 Notting Hill Gate London W11APPENDIX G (cont'd)
GENERAL CONDITIONS OF CONTRACT FOR MINOR WORKS 1, Definitions
"Employer"
"Contractor11
’’Engineer”
“Contract”
“Works
'Payment Certificate” Lefe°ts Liability Period”
" Pinal Certificate”
Law
Language
The tenderer whose tender has been accepted by the 4&npl oyer
Sir William Halcrow and Partners
The general conditions, specification, draw ings, priced Bill of Quantities, the tender
and any pre-Contraot correspondence
The work to be executed in accordance with 
the Contract
Issued by the Engineer to the Contractor when the Works are substantially complete
The period, shown in the tender, following the issue of the Completion Certificate
Issued by the Engineer to the Contractor and Employer after the Defects Liability Period The Contract shall be construed according to the law of the country in which the Works are carried out, and the language used shall be English.
2. 2.1
The Contractor’s Responsibilities
The Contract comprises the execution and completion of the Works, and also making good defects arising before the end of the Defeots
la  xlity Period. The Contractor shall provide all labour, mater
s’ Plant, temporary works, superintendence and everything,
Gr  of a temporary of permanent nature, that may be required to co mplet  the Contract.
eAPPENDIX G
(cont’d)
2.2 
It iB aBBUme(i that the Contractor has examined the site and satis fied himself before tendering as to the correctness and sufficiency of hie tender. It is also assumed that the rates and prices in the
tender cover all his obligations, whether specified in, or reason ably to be inferred from the Contract
2.3 
The Contractor shall execute the Works to the satisfaction of the
Engineer, and shall comply with the Engineer's instructions cn all
related matters,
2.4 
The Contractor shall provide all assistance that may be required by the Engineer. The Contractor shall also provide the Engineer with such returns as the latter may require. Showing details of the
labour, plant and/or material employed on the site.
2.5      The Contractor shall be responsible for the setting-out of the Works
and for the correctness of the positions, levels, dimensions and 
alignment of all parts of the Works.
2*6 The Contractor shall be responsible for making good any defeots aris
ing from the use of materials and/or workmanship not in accordance
with the Contract. These shall include defects which become apparent during the Defects Liability Period. The Contractor shall execute all
such repair work at his own expense and with all reasonable speed, so • that the Works shall, at the conclusion of the Defects Liability Period, 
be in good order, fair wear and tear excepted.
p ■y 
The Contractor shall not assign the Contract. The Contractor shall
not sub-let any portion of it, without the consent of the Engineer.
Liabilities and. Insurances
3,1  The Works:
The Contractor shall be responsible for the care of the Works and shall . x his own cost make good any damage which may happen to the Works or
temporary works prior to the expiry of the Defects Liability Period Trom any cause, except one solely due to the Engineer’s design of the Works, or one due to war or civil commotion or one solely due to 
"the use Of works by the Employer during the Defects Liability 
fe riod. The Contractor shall insure against any loss or damage for w hich he is responsible under the Contract. The insurance policy is
Le in the joint names of the Employer and the Contractor.APPENDIX G (cont’d)
Third. Party:
Third Contractor shall indemnify the Employer against all claims for
injuries or damages to any person or property which may arise out of
the execution of the Works, except those resulting from the actual
default of the Employer or his agents or servants. The Contractor
shall insure against such liability, for at least the amount shown 
in the tender. The insurance policy is to be in the joint names of
the Employer and the Contractor.
3.3 Employees:
The Contractor shall indemnify the Employer against all damages or compensation consequent upon any accident or injury to any person employed by the Contractor or his Sub-Contractors in connection with 
the execution of the Works. The Contractor shall insure against such liability.
The Contractor shall, whenever required, produce to the Engineer any 
of the insurance policies, and the receipts for payment of the currant premuims.
Progress of the Work
The Contractor shall commence the Works within 14 days of the Engineer s order to commence. The Contractor shall then proceed with the Works with due expedition so that they are oompleted within tho time stated in the tender calculated from the date of that order■
4.2
The Contractor shall also, within 14 day- o mit to the Engineer for his approva
order of procedure and the methods y w the work. The programme is to be in sue a easily monitored. If the Engineer considers
f the order to commence, sub— o 
t
shQW the
proposes to carry out orm that progress can be
^
he ra e
t  of progress
ctQr shali
 take such
is    too    slow    to    ensure
steps as may be approved by the Engineer
If the execution of the work shall be delayed y 
timely    completion,    0      vnPaite    the    work, 
, a bv any causo boyond the
thc  time
reasonable oontrol of the Contractor, o * reasonable having regard tor completion by Such period as ho considers ro»
”t° all the circumstances*APPENDIX G (cont’d.)
4.4                  If the Contractor fails to complete the Works within the time stated., 
or extended, time, then the Contractor shall pay to the Employer liqi- dated damages at the rate stated in the tender, up to a maximum of 10$ 
of the tender amount.
4.5                  The Final Certificate, giving the final cost and signifying that the Contractor’s obligations have been fulfilled, will be issued after the
end of the Defects Liability Period when all defects have been rectified
5.                        Payment for the Work
5.1                    The Works will be measured by the Engineer, and final payment to the Contraotor will be based on the sum of the products of the actual ; quantities and the appropriate tender rates, and/or variation rates.
5.2                   Interim payments will be made monthly? • The Contractor shall submit a monthly account to the Engineer showing the value of work done and the make-up. The Engineer will agree or amend this account and issue the Payment Certificate to the Employer.
5.3                   The Employer shall pay the Contractor within 14 days of receipt of the Payment Certificate,retaining 10$ of all payments.
5.4                   The amount retained will be paid to the Contractor in two halves, thu. first within 14 days of the issue of the Completion Certificate and the second within 14 days of the issue of the Final Certificate.
6.
6.1
6.2
Variations of the Work
The Engineer may order any variation in the form, quality or quantity 
°f the work that may, in his opinion, be necossary for the satisfactory completion of the Works. The Contractor shall execute suoh variations.
If appropriate, such variations shall be valued at the rates in the
•^tll of Quantities. If no such ratos arc appropriate, then valuation 
v, ill be by agreement between the Engineer and the Contractor. If agree— merrt is not reached, the Engineer shall fix reasonable rates, deducod 
f rom the rates in the Bill of Quantities as far as is practicable.APPENDIX G
(cont ’ d)
Provisional and prime cost sums in the tender shall only be used on 
the direction of the Engineer. The rates for provisional sums will be
determined as for variations, and the rates for prime cost sums will
be those derived by substituting the actual cost for the prime cost
sum.
!•             Adverse Conditions
If the Contractor shall encounter physical conditions or artificial
obstructions which could not have been reasonably foreseen by an ex
perienced Contractor, he shall immediately give written notice there
of to the Engineer. If, in the Engineer’s opinion, such conditions
or obstructions do exist, then the Contractor will be entitled to 
reimbursement of the reasonable additional costs arising from such con
ditions or obstructions.
Expulsion
If the Contractor shall go into liquidation, or shall persistently neg
lect his obligations under the Contract, the Employer may expel hin 
iron the site and employ another Contractor to complete the Works ana may, without prejudice to any other method of remedy, deduct from mcney due to the Contractor any additional costs resulting from this procedure
Disputes
If any dispute or difference shall arise between the Employer and the Contractor in connection with the Contract, it shall be referred t« and settled by the Engineer who shall state his decision in writing to the Employ   and the Contractor.
er
If  the Employer or the Contractor is dissatisfied with any such decision °f the Engineer, then either may requiro, within one month, that the matter be referred to the appropriate Ethiopian court, to be agreed bcW Ween the parties or, failing agreement, to one nominated by the President ®f the Institute of Arbitrators. The award of the arbitrator shall be
final and binding on the parties.TABLE X
Comparison of Seasonal Consumptive use Requirements
Planting Period
I
II I
II
IV        V      VI
Blaney - Criddle Total U(mm)
480
457
439
433  423  410
Max, ^-monthly U(mm)
97
86
86
84    84    80
Penman, Total ET(mm)
468
459
451
446  441  437
Max. ^-monthly ET(mm)
95
88
88
91    91    89
Thornthwaite. Total ET(mm)
360
364
333
321  307  289
Max. ^-monthly ET(mm)
79
71
67
69    69    62
^argreavcB - Christiansen. Total ET(mm)
557
535
520
517  516  513
Max. -^-monthly ET(mm)
105
93
90
95    96  100
Planting periods are:               I  15th May - 3lBt May
II      1st June - 15th Juno
III      16th June- 30th June
IV   let July - 15th   July
V   16th July- 3lBt July
VI 1st August - 15th AugustTABLE 2
SECONDARY CANAL DESIGN DUTIES AND DISCHARGES
!io.
Gross area
Commanded
(ha)
design Max. 
flow
(l/sec)
Approx, length
Caned
(kz)
Approx.
Seepage loss
(l/seo)~
Gross design 
canal flow at head regulator (l/sec)
10
20
JO
40 50
6C
70
8C
9C IOC
Uc
435
375
815
1095
950
490
710
970
575
395
630
590
510
1110
1485
1295
665
965
1315
780
535
855
1.9
1.1
2.5
3.65
2.75
1.85
5-4
1.45
2.35
3.70
2.10
22
13
29
42
32
21
63
17
27
43
24
612
523
1139
1527
1327
686
1028
1332
807
578
879
»
-1
* 
EeeiGTi for seepace l°ss at ra^ , i. haV0+o of 1000 m /i . / Jan./ j „ cdafy-. fjnibora where  been obser
losses between 300 and 1500 ™ (1973)
testsSCHEDULE OF PRIMARY
TABLE 3
CANAL OUTLETS AND DISCHaRG^
00 CM
xT
H
H
H
to
Cx
3i
CM (T> H  0) k0 ID oj rx
CM
ID
KT
<0
00 <0 cx  sr to  10
(0
to xr CX ID ID
to
<O 00 Cfi 0) to \T 10 CM
0> 0 ST ID H ST
H
10 CM to
H
d® c® du OH >» 0 dd E CL
•H Qj U Q) fc ®
3
0C C0 d0 4J 3 !0 u
Q O oo
T ID to
0* ▼S
O o>
Cx  H
HH
cn
I H Cx
H
to 00 10 •7 Cx 1 ID 1 CM I to 1 H
H           *d        H
© P   ->
□E
H
P
3
0
O     ID     ID              O     ID
00 CX H
•■ •
CM  CM
* O • •
ID o 00 Cx
CM vT to o
•
CM   O      H              to   CM          CM    H                       H
O•            •         O•
o*
o • o • o •
ID  H       GJ
®
01
CM CM P
cj
P cj
0
£ GJ
c
U
c
•d GJ E
p GJ
0
p
•d
3 cr
©
P
3
O
•d
©
0
©
E
3
o
cx
II
3
U GJ
X)
•d
E
P
0
04
p
0
p GJ
H
3
no
Q)
2369
1285TABLE 4
FLOOD MAGNITUDES (CUMECS) AND THEIR RETURN PERIODS (YEARS) FOR RIVERS AWASH (AWASH STATION) AND KGSEM (AWARA BELKA)
T 
(Years)
AWASH AT AWASH STATION
KESEM AT AWARA NELKA
Gumbel 1Log-Pearson
Boldakov
Gumbel
Log-Pcarson
Boldakov
1
23
4
5
6
7
(2.33)
10
50
loo
200
500
looo
loooo
200              — 305           299
405   '       398
440    1     443
485           489
540           540
580           585
-              740
--
—
295
-
370
-
-
460
513
670
415
710
990
1100 1220 1400 1500
-
-
669
949
1039
1152
1300
MOO
1800
-
-
665
-
1120
-
-
1200
1350
1830V—
AccNo. —
Copy No-----------//
Call No.
AUTHOR TITLE
CALL NO.
EDITION
VOLUME
CAcc. NO

< <

Summery

Project: Amibara Irrigation Project, Awash Valley, Ethiopia

Date: September 1973

Prepared by: Sir William Halcrow & Partners for Imperial Ethiopian Government

1. Project Overview

The Amibara Project is a major irrigation development in Ethiopia's Awash Valley, designed to irrigate 10,100 hectares of land (with potential expansion to include adjacent areas). It forms part of a larger regional development plan for the Middle Awash Valley.

Key Components:

River diversion structure at Melka Sadi
Main and primary supply canals
Primary and secondary drainage system
Flood protection dykes (western and eastern)
Road network and bridges
Settlement villages and housing
Regional township and administration center

2. Design Highlights

Water Supply System

Key decision to use a single diversion structure at Melka Sadi rather than two structures as originally proposed, resulting in estimated cost savings of Eth$2 million.

Irrigation System

Component	Design Specification
Crop Water Requirements	Max 105mm half-monthly for cotton (1.68 l/sec/ha design duty)
Primary Canal Capacity	17 cumecs for Amibara (expandable to 23 cumecs for future developments)
Secondary Canals	Following natural topography rather than rigid geometric layout

Flood Protection

Western dyke designed to protect against 2000 m³/sec flood (50-year return period). Economic analysis suggests optimal dyke height of 3-4m.

Drainage System

Designed to handle 50mm/day storm intensity with 80% runoff from irrigated fields. Primary drains capacity approximately 36 m³/sec.

3. Implementation Plan

Phased Construction

Phase I: 4,200 ha including AVA settlement farm (target completion June 1976)
Phase II: 4,400 ha including existing concessions
Phase III: Remaining 1,500 ha

Contract Structure

Contract Type	Description
Preliminary Works	Bush clearing, site investigations, link road construction
Buildings & Services	Settlers housing, AVA staff housing, administration buildings
Main Works	Barrage, canals, drains, roads, flood protection (international tender)

4. Key Appendices

Appendix B: Rationale for single diversion structure at Melka Sadi
Appendix C: Effects of potential Angelele and Bolhamo developments
Appendix F: Contract types and tendering procedures
Appendix G: General conditions for minor works contracts

        Critical Decisions Required from AVA
        Approval of single diversion structure design
Confirmation of irrigation water duties
Approval of flood protection criteria
Acceptance of implementation program and contract structure
Budget approval for preliminary works (including Eth$20,000 for site access improvements)

    