1 :1: «2 “:11: ‘1 .'. l r' :,: .» . 1:13.111114'11'1‘1! '3‘ a 4-. u ‘ . ".4." >2“. 1 I -‘H ‘ _ ‘ - 4 o ‘ 2-. . X ' - . < - 4' . - 5' 7 __A4- . < 4 .0. A 4 z . . . ‘ ‘ A.‘ A . £54 7 . . ‘ - ' ' - .. ' ; - -. . . .1 - .‘ a, p . . ‘ u ccccc a o ~44 --.‘.‘- . 0-4a...- . o. 9.09. a 3;" :: vvvvv ”11:; 1:1 ’ '1. . ‘2'. 11,2. 11 1113!”: ’1'] | 1:11' '11. {Ht}. P‘l‘:1\“." . ‘ : 9:111:11'111‘11 ’ ' 11171111"11'11111'Alfvlg11:11:1111111’111‘111Elf'1‘1111‘1mi1 I {‘1"; ‘111‘1I‘1’! : .3] [6. [Pi 1171 111,1 g': :1 1"...Jp'f 1"1‘11 -" 1":11: : I .. II: ' “#4111... a“. (‘13.1'111: :11 {i H M: 1:: WM 1‘:::::: ' 11' : 11 :"111 '11"! 11112:: :3'11111:"" ‘1’::“:" 11:11 {:1 3’11‘111111:1:i:1'111 t1.“ 1::.:1.1 ii] ..','::".:. 1!}!1 i... ii ' 31511;“! W, ‘: . .i: '....:..,::::..:: w: ‘- 4—-“- ‘m L lMW[I’E’WWl ,.{ , '. r. ' u ! v‘é-Cn L-‘ I 9 l‘ I. o ..‘.'n..u4.Prl.--' pf (1“. JII3‘1‘1;.’.~.- Guy-cw 7--3 e Lil—1919:3231 This is to certify that the dissertation entitled FEASIBILITY EVALUATION FOR IMPROVED LIVESTOCK SECTOR OF THE AGROPASTORAL PRODUCTION SYSTEM OF THE UPPER CASMANCE. SENEGAL presented by Abdou Fall has been accepted towards fulfillment of the requirements for M.S. degree in Animal Science e MS U is an Afl'mnafivc Action/Equal Opportunity Institution 0-12771 MSU LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped below. FEASIBILITY EVALUATION FOR IMPROVED LIVESTOCK SECTOR OF THE AGROPASTORAL PRODUCTION SYSTEM OF THE UPPER CASAMANCE, SENEGAL BY Abdou Fall A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree MASTER OF SCIENCE Department of Animal Science 1986 ABSTRACT FEASIBILITY EVALUATION FOR IMPROVED LIVESTOCK SECTOR OF THE AGROPASTORAL PRODUCTION SYSTEM OF THE UPPER CASAMANCE, SENEGAL BY Abdou Fall Animal productivity in the agropastoral production system of Upper Casamance is plagued by ecological, biological, and socioeconomic constraints. The feasibility evaluation phase of the "sytems approach " was conducted to generate alternative solutions to alleviate system constraints. The goals and objectives of all persons and institutions involved in the system are analysed. The system structure and determinant variables are specified. Microcomputer Spreadsheets evaluating the feed balance and a demographic simulation model including an animal traction feature helped specify problems. The study concluded promising alternatives to improve the livestock system in Upper Casamance are: l. the use of biuret to treat or supplement dry season roughages, 2. a better conservation and use of groundnut hays, 3. the intercropping of cereals and cowpea, 4. the utilization of cottonseeds for oxen and lactating cows, 5. a group-growing out production system, 6. Stall feeding combined with limited grazing of oxen and lactating cows, 7. improve small ruminant production. ACKNOWLEDGEMENTS I express my deepest appreciation to Dr Robert Deans who served as my major professor. He critically reviewed this thesis throughout its preparation. I am also grateful to Dr. Manetsch and Dr. Magee for their help. Dr. Manetsch offered me valuable suggestions. My appreciations go also to Dr. Bersten for his support and constructive critisism. I am also deeply indebted to the US agency for International Development (USAID) and the Institut Senegalais de Reserche Agricole for giving me the opportunity to enroll in a master's degree program at Michigan State University. I finally express my profound thanks to all my friends at MSU specially Nana Makaula and Elizabeth Brabbs for their help in editing this thesis. ii To the memory of my brother THIERNO FALL iii TABLE OF CONTENT CHAPTER PAGE I INTRODUCTION 0.0.00......OOOOOOOQOOOOOOOOO0.0.01 Objectives Of the StUdYO I O O O I O O O O O O O O O O O O O O O O 4 Data Gathering O O C C O O O O O O O O O O O O O O O I O O O O O O I O O O O 4 MethOdOIOgYOOOOOOOOOOOOOOOOOOOOOOOOCOOOOOOOOOS Organization of the Study....................7 8 II NEEDS ANALYSISOOOOOOOOOOOO00....00.000.000.00... Decision-making levels.......................8 Farm level................................9 Village area.............................ll Agropastoral region......................12 Nation level.............................l3 Objectives..................................l4 Farmers' objectives......................l4 Nation objectives........................18 Conflicting objectives......................20 Men vs. women............................20 Meat vs. milk production.................21 Consumer vs. producer welfare............21 III SYSTEM IDENTIFICATIONOCCOOO.COOOOOOOOOOCOOO0.0.23 System definition...........................23 System design parameters....................25 Livestock species and herd size..........27 Land use.................................29 System desired outputs......................29 System undesired outputs....................30 Exogeneous inputs...........................34 Weather variables........................34 Feed resources........ ..... ..............35 Controllable overt inputs...................42 System mangement............................45 Herd management..........................45 Oxen management..........................47 Offtake..................................48 iv System performance criteria.................50 Iv PROBLEM FORMULATIONOOOO0......0.00.00.00.000000053 Nutrition budget.............................54 Description...............................54 Technical assumptions.....................55 Results and discussion....................58 Productivity levels..........................63 Milk production...........................63 Meat production...........................65 Reproductive parameters and mortality...........................66 Growth traits...........................68 Energy input to agriculture..................72 Ndama draft potential and oxen availability..............................74 Actual use of animal traction.............78 Animal traction constraints...............81 Manure Production............................82 Demographic model............................83 Objectives................................83 Description of the model..................83 Validation................................92 Base run..................................96 Sensivity analysis.......................101 Summary of constraints......................109 V SYSTEMS ' ALTERNATIVE SOLUTIONS . . . . . . . . . . . . . . . . lll Improvement of crop residues and agroindustrial by-product utilization.................................112 Treatments...............................ll4 Physical treatment...................115 Chemical treatment...................116 Supplementation with high quality forage.......................ll8 Utilization of cottonseeds...................120 Fooder production............................122 Improvement of small ruminant production...................................124 Alternative management strategies...................................126 Group-growing out........................127 Other management issues...................................129 V Other required investments ...... . ............ 130 Promising alternatives.......................l3l Appendices .......... .... ........... ..........135 Bibliography................... ..... .........153 vi Table 1. Table 2. Table 3. Table 4. Table 5. Table 6. Table 7. Table 8. Table 9. Table 10. Table 11. Table 12. Table 13. Table 14. LIST OF TABLES Page Cattle Herd size and structure in the three districts of Kolda.........28 Pastures at Kolda: Types, Productivity and Season of EXPlj-otationoooo000000000000...0.0.00.037 Estimation of Energy (Mcal ME) Supply from Crop Residues.............39 Pasture QualitYOOOOOOOOO0.00.00.00.0000042 Monthly Weight Change of the Foundation Herd at KeldaIOOOOOOOOOOOOOOOOSB Montly Energy Offer, Demand and Balance( Million Mcal).. ......... ..60 Estimation of Per Capita Milk Consumption at Lenguewal, in Upper casamanceOOOO00.00.00.0000000065 Ndama Reproductive Performance and Viability at KOIdaOOIOOIOOOOOOOOOOO67 Ndama Weight at Different Age in Village Herds at Kolda..................69 Seasonal Liveweight Change of 4 Animal Groups under Different Nutritional Planes............70 Cow Productivity under Village Circumstances.0.0.0.000...00.0.00000000071 Number of Draft Animals in Upper Casamance ........... ..... .. ...... 77 Potential Availability of Draft Animals from Kolda Herds...............78 Percentage of Mechanized Agricultural Operations at Kolda.....................80 vii Table Table Table Table Table Table Table 15. 16. 17. 18. 19. 20. 21. Total Herd Poulation at Selected Points in Time versus DT step Size (Delay7=8).00.0.000000000000000094 Comparison of Simulated and Theoretical Deaths over Time............97 Simulated versus Actual Offtake and Population Growth Rates.............97 Model Outputs: Base Run ( Oxen Population, Number of Males Sold, Oxen Balance)............99 Cattle Herd Size under Different Scenarios at t=lO............lO4 Estimation of Crop Residues Availability...OOOCOOOOOOOOOOOOO0......113 Nutritive Value of Groundnut Hay According to Harvest and Conservation Methods....................119 viii LIST OF FIGURES Figure l. Feasibility Evaluation Figure 2. System Identification as Part of the Systems Approach ( Feasibility Evaluation) Figure 3. A General Description of the Agropastoral Production System of Upper Casamance Figure 4. Description of Pastures in Upper and Middle Casamance Figure 5. Dabo Energy Balance ( Energy Demand, Offer, Balance) Figure 6. Dioulacolon Energy Balance ( Energy Demand , Offer, Balance) Figure 7. Medina Energy Balance ( Energy Demand, Offer and Balance) Figure 8. Traction Power: Schematic Representation of Influential Factors and their Interdependences Figure 9. Block Diagram of the Demographic Model Figure 10. Decision -Making Process for the allocation of young males from Delay 5 to Bulls and Oxen Delays. Figure 11. Model Outputs : Base Run ( Oxen, Male sold, Oxen deficit/surplus) Figure 12. Model Outputs ( Sensitivity Analysis) Figure 13. Improved Management and Feeding Systems of the Cattle System of Upper Casamance ix Page 24 26 4O 61 61 62 75 86 87 99 103 134 CHAPTER ONE INTRODUCTION Senegal is experiencing large deficits in food production and relies on huge imports of cereal and animal products. The share of animal production in the primary sector has increased from 10 percent in the 1960's to 30 percent in the 1980's. In spite of such a substantial increase, this sector has been subjected to dramatic changes induced by a series of climatic disasters as well as pre and post- colonial policy orientations which resulted in the neglect of livestock production. Futhermore, livestock development schemes implemented so far appear ill-conceived with a lack of systematic planning. The livestock distribution in Senegal follows a pattern related to climatic conditions. The Bos indicus type of cattle, the zebu Gobra, has its natural environment in the northern, drier part of the country. Ndama cattle, known because of their outstanding trypanotolerant attribute, are found in more humid, tsetse infested zones: Casamance, Senegal Oriental, and the southern part of Sine Saloum. The bovine population in the intermediate zone consits of the " Djakore " type of cattle which is a crossbreed Ndama-Zebu. The Ndama cattle make up 30 percent of the total bovine population in Senegal and 10 percent of the total trypanotolerant cattle in Africa(ILCA,l979) In recent years, the trypanotolerant livestock have become more acceptable. In the past, their productivity potential was underestimated due to their small frame as compared to more susceptible large-frame breeds. For instance, Trail and Wissocq state: ” Over the last ten years, the emphasis has been changing from the view of humpless cattle in Africa , as a historic relic and their trypanotolerance as a biological oddity, to concentration on their economic possibilities". (Trail and Wissocq, 1981, p.57) Ndama cattle are a central component in the mixed crop/livestock production system in Upper Casamance. This mixed farming system is characterized by a completely intertwined and closely integrated livestock and crop production systems. Many livestock species ( cattle, sheep, goat, equines) are raised and many crops (millet, sorghum, maize rice, groundnut, and cotton) are grown. The interdependence of the livestock and crop production sectors is facilitated by the convenient symbiotic interaction of these two components of the agropastoral system. For example, the crop sector receives power and manure from animals while the livestock sector benefits from crop residues. The mixed production system provides jobs and subsistence means to thousands of Senegalese in the subhumid zones. Farmers secure their subsistance and cash needs from crop and animal related activities. Above all, this production sector significantly contributes to the national meat supply. Despite the key role played by the livestock sector in the agropastoral system, it has received little attention from decision makers. The need to increase the supply of red meat and other animal products has been realized yet the only livestock improvement efforts known to have been brought to the level of farmers were inoculation campaigns. Most research efforts undertaken in Upper Casamance are removed from farmers' problems and realities. Research activitites are conducted under a controlled environment, at a research station. Important socioeconomic variables and the physical environment of the production system are ignored. One of the most important obstacle facing livestock production is the inadequacy of the feeding system marked by a long period in the year where the feed quality is so low that it cannot even ensure animal maintenance requirements. The consequent tremendous weight losses are not conducive to high herd productivity. There is a need to investigate means by which the livestock system can be improved in order to enhance the productivity of the agropastoral system. The starting point for this study is to focus on how agropastoralist activities are presently conducted and to discover the underlying objectives and motivations of these activities for all parties concerned. 1.1. Objectives of the Study The objectives of the study were to : l. analyse the objectives and motivations of people and institutions involved in the agropastoral production system of Upper Casamance. 2. describe the livestock production system by Specifying its structure, input and output variables and management characteristics. 3. define major constraints limiting animal productions 4. generate alternative solutions most likely to remove or alleviate identified constraints. 5. specify the most feasible alternatives The general aim of the study is to investigate means by which the feeding and management systems can be enhanced. 1.2. Data Gathering Data for the present study was gathered from a review of literature written about Senegalese livestock system particularly the Upper Casamance mixed farming system. There was a great scarcity of written materials on livestock production in the Upper Casamance. This paucity of information results from the afore-mentioned neglect of the livestock sector. The researcher moved from office to office searching for documents and interviewing people working in these institutions to supplement scant and scattered data. In addition, thirty randomly selected farmers in Upper Casamance were also informally interviewed so as to gain a holistic perspective of the livestock production system and a perception of farmers' objectives and motivations. 1.3. Methodology The framework utilized to carry out this study is the " systems approach". It is a problem solving process with five major phases: 1. Feasibility Evaluation, 2. Abstract Modelling, 3. Implementation Design, 4. Implementation, and 5. Operation. It is only the feasibility evaluation of this methodology which will be used in this study. Feasibility evaluation has as its goal the generation of a set of system "alternatives" aimed at satisfying needs that have been identified and selected for satisfaction (Manetsch,1982). It is an iterative process moving through six phases as illustrated in figure 1.: 1. Needs analysis, 2. System identification, 3. Problem formulation, 4. Generation of system alternative solutions, 5. Physical,sociaL.and political realizability, 6. Economic and financial feasibility. Stop Primitive Need 1 l I Needs .1- :msera t i on A i y.- —- -- O ys COM nalys I Alternatives ‘ r; u 3 u 0.0 NO ll! N0 . Valid Need? 5 Complete? Complete? 3'3 . E,'8 C {5'5 yes 1) In Alternatives ‘ '5: System V I: I) S ys‘ca videntification|—' + "" "’ 331?“??? . ' - I Realizabil t l . ) Complete?- | Realizable? ' N0 Adequate? ' Complete? was I . ' ; YES in .o I Rea izab e H Definition ' Alternative ’ ‘ terminatTon» -Problen “ ' » . _ f Economic and Fornuhmq "L - "' Financial Feasibility H l Profitable? _ Financible? . Statement . 70 Abstract Model l inq _‘ v Figure l Feasibility Evaluation Sources Henetscndfl? w 1.4. Organization of the study The organization of the study follows the different stages of the feasibility evaluation process. Chapter two identifies the needs of all persons and institutions operating in the system. Chapter three identifies the system. Its structure, input and output variables, and management practices are specified, as well as the criteria capable of assessing the system performance under different alternatives. Chapter four defines the problem confronting animal production. Productivity levels are also evaluated. A microcomputer spreadsheet model evaluating the feed balance and a demographic simulation model will help identify constraints. Alternative solutions will be generated in chapter five and their feasibility examined. CHAPTER TWO NEEDS ANALYSIS Feasibility evaluation begins with the analysis of the needs the system under consideration must satisfy (Manestch, 1977). The motivations of all persons and institutions involved in the system are taken into account. The large number of operators in the agropastoral production system in Upper Casamance who are pursuing different goals, dictates the analysis of needs at each decision-making level. This approach has the advantage of providing useful guidance for the introduction of innovations as the specific goals of each target group are considered. 2.1. Decision-making levels. The stratification of decision—making levels is based on the utilization of production factors, e.g., capital, labor, land, and the existing social environment. External factors have significant impact on the system performance, therefore, exogeneous decision-making levels need to be taken into consideration. Four decision-making levels are identified: farm, village area, agropastaoral zone, and nation. 2.1.1. Farm level The farm is an important decision-making level because that is where the production process takes place. The complex use of production factors such as the collective use of labor, farm outputs utilization, and the notion of extended family, make it difficult to define a farm in Upper Casamance. SOMIVAC (1980) views a farm in the Casamance context as a family unit using land for agricultural purposes, autonomous in the utilization of production factors (land, capital, and labor) and in determining the destination of farm outputs (sale, consumption, gifts etc.). The farm can be represented by a single relatively independent management unit or a set of management units ("Principal farm") made up with many households ("Secondary farm") whose members are blood related and headed by the head of family. The most important features of the farm organization are: - a sexual division of labor, - the apportionment of reSponsibilities according to age, - a collective production of cereal for family consumption and individual production of cash crops, - an individual ownership of livestock under a common management. 10 Due to this complex socioeconomic setting, farm management is marked by the following sub-decision-making levels: head of family, other men and women. Head of Family He is responsible for the farm food security. His management responsibilities include the allocation of land to different crops and to different members of the household, labor utilization and hiring, the timing of cereal growing tasks, farm equipment acquisition and family herd management. Other household members One important aspect of the development of the monetary economy in the Upper Casamance zone resulting from the introduction of cash crops is the drastic change in the traditional environment of the production system. The most obvious aSpect of this change is that the production system has evolved from an essentially communal basis to a production system characterized by the separation of subsistence and cash agriculture. (Achterstraat,l984). Each member of the household makes personal use of revenues generated from his/her individual parcel of cash crop and from other off—farm activities to satisfy his/her own needs (consumer goods, livestock and sometimes farm equipment‘ purchase, wedding expenses etc.) Besides their domestic responsibilities, e.g., taking care of children, cooking meals, females' main activities 11 are directed toward rice production for family consumption, vegetable and sometimes cereal for family use or market and cash crop production. 2.1.2. Village Area The village area is an important decision-making level when it comes to decisions regarding access to resources such as water, crop and pasture lands. The village is composed of a set of households. Typically, villages are located near a valleybottom. Village land resources are of different potential and use : - the valleybottom areas are allocated to rainfed rice production and are an important source of forage during the dry season, — the uplands have two components. The cleared land surrounding the homesteads are used for cereal and cash crops production whereas the woody lands provide natural pastures communally accessible to village herds, — fallow areas are maintained to restore soil fertility and are used as rainy season grazinglands. The spatial and temporal delineations of land use are key features of the production system which mark livestock management characteristics and the interaction between crop and animal components of the system. The village chief is the main authority controlling land use at village level. His permission must be sought 12 before the clearing of new crop land from the woody lands can be done. Even though all herds, including those from other villages, have common access to pastures, the village community exercises control over pastures located between valleybottoms and crop fields. Their use for cropping purposes is controlled by the village community. In Spite of a new land tenure system officially designed and implemented since 1978 in Casamance through "rural communities" and giving power to the "rural counsel" for land allocation, the traditional land tenure system still prevails. Similarly, other official institutions such as the "Service des Eaux et Forets" guardian of rangeland resources preservation, are not effective in controlling the utilization of rangeland resources. 2.1.3 Agropastoral Region The environment of the agropastoral region is similar throughout in regards to rainfall, water resources, social and economic patterns of production, herd structure and management, population density, marketing of inputs and outputs. These different elements will be examined later. Many development and research institutions are Operating in the Upper Casamance mixed farming zone. These are aimed at implementing particular development schemes or at executing specific research activities. The "Societe de Development des Fibres Textiles: SODEFITEX" is the most prominent regional development 13 institution which operates in the zone. It is responsible for the promotion of cotton production , its marketing , and the integrated rural development of Upper Casamance and Eastern Senegal regions. It controls the whole process of cotton production i.e. supply of production factors to farmers, extension services, ginning, marketing of cotton fibre and cottonseeds. Even though a food crop component has been introduced in its program since 1971, cereal production remains a secondary activity. The "Service de 1'Elevage et des Productions Animales", a governmental institution, has in its charge regional health and production programs. Because of financial constraints its activities are limited to annual campaigns of vaccination. The "Centre de Recherches Zootechniques de Kolda" conducts breeding plans on Ndama cattle and Djallonke sheep aimed at the genetic improvement of these breeds for meat purpose. Its research programs also include the testing of the productivity of grass and legume forages of the genera Stylosanthes, Panicum, and Andropogon. 2.1.4. Nation Level Important issues such as agricultural policies and development schemes, are designed at a national level and implemented by regional development institutions. The goverment whose concern is the balance of payment, public finances etc., sets pricing policies for agricultural l4 commodities and inputs. It determines producer prices for cash and cereal crops, input prices, subsidies and taxes as well as consumer prices for meat. However, it does not intervene in livestock producer prices. The identification of these different decision-making levels enables us to scrutinize objectives at each level and to identify conflicting interests. 2.2 Objectives. 2.2.1. Farmers' objectives Farmers' motivations have been documented in many studies (Simpson, 1984; Sandford, 1982; Girardot—Berg, 1982; Munzinger, 1982). Information secured from the review of literature and interviews with thirty agropastoralists for this study led to the identification of the following goals and objectives : 1. increase in monetary revenues, 2. improvement in standard of living, 3. short term and long term viability under unreliable physical environment, 4. decrease in inconveniences such as daily drudgeries, 5. improvement in health care, 6. capital formation. Agropastoralists have learned to adopt a variety of strategies to cope with the unpredictable environment in which they live in order to achieve their goals and 15 objectives. Many crops are grown. Cereals collectively produced are given priority for food security. However, most of the agropastoralists interviewed experience a food Shortage usually in the first part of the cropping season when the previous year's harvest is depleted. This is the time when small ruminants play their greatest role in supplementing farmers' revenues for food purchase. Presently, a common strategy to combat this periodic food crisis is to emphasize maize production during the early rainy season. The maize is harvested in mid - September when other cereals are not yet mature. The prevailing cattle herd structure with 72 percent female and the reluctance of farmers to destock even old females is motivated by the high value they place on family milk supply and their concern to insure herd existance under harsh conditions, e.g., diseases, poor nutrition, etc, which cause low reproductive performances. Thus, this is a way of ensuring their own long term viability. "Animals constitute capital assets in every sense of the term, they embody past investments, provide services, require maintenance, and they can be liquidated and reinvested" (Panayotou and Ruangrai, 1982, p. 67). Investment in livestock is the most important means of capital accumulation for farmers. The objective of most of the people interviewed for this study was to increase_their herd size; those without livestock were aiming at building up a herd when cash generated from crops or other sources 16 allows it. However, cash surplus left after the satisfaction of basic needs seldom enables farmers to increase or create a herd. Generally speaking, ambitions expressed by farmers are not different from those of people outside of their Sphere. They aspire to achieve "development" that would mean an improvement in their standard of living i.e. better nutrition, health care and education accompanied by the conservation of their value system. The above discussed objectives are general and do not provide useful guidance about particular perspectives at different decision making levels. Each individual on the farm strives to gain monetary revenue to secure consumer goods supplies, e.g., bicycles, radios, clothes etc., or services such as health fees, and to purchase production factors ( livestock, and farm equipment). It is the ambition of each young man to become head of a household and marriage means being able to cover the dowry requirements expressed in terms of heads of cattle. Usually, livestock is acquired through heritage, purchase or as a reward of years of looking after herds of cattle. The pursuit of monetary gains is indicated by the fact that personal plots are usually utilized for cash crop production. The farm and village milieu are marked by important differences in wealth, cattle ownership, and ethnicities. Studies conducted by BOAD(l984) revealed that 30 percent of 17 Fulani households, traditionally known as prominent cattle keepers, did not own cattle. The difference in cattle ownership is more noticable between ethnicities; Fulani agropastoralist raise more livestock than Mandingo people. This has a significant impact on certain farmers ability to acquire draft animals and farm equipment and to improve their soil fertility with animal wastes. Most people interviewed expressed their needs to have draft animals and/or farm animal traction implements, especially weeding materials. Women interviewed were concerned with the demanding domestic tasks and the drudgeries associated with rice production operations and cereal processing. Almost all rice production tasks as well as cereal processing to obtain flour are hand operated. Cereal processing machines and nurseries as well as the utilization of animal energy for rice production, are the women's prime needs. Two other categories of persons involved in the livestock production system are herdsmen and traders. The former are usually strangers in the village and are pursuing capital accumulation (livestock, money). They are hired by the head of household to graze cattle herd and keep them from damaging crops. Their renumeration is in milk which they market, wage earnings determined according to the herd size, and accommodation. The latter are middlemen seeking monetary gains. They go from village to village collecting slaughter to supply consumption centers. 18 Children play an important role in farm activities by significantly contributing to the farm labor force. Functions such as releasing and tethering cattle before and after grazing, milking, guiding animals during ploughing and seeding operations, protecting cereal crops from bird invasion etc. are performed by children in the farming endeavours. School often competes with their essential contribution to the farm operational tasks. The objectives of official institutions operating in the system and those of persons using part of the system‘s output (meat) are confined in national objectives. 2.2.2. Nation Objectives "Government, as a trustholder of society, inherently embodies numerous objectives ..." (Simpson, 1984, p. 11.). The national goals pertaining to the livestock sector are stated in the 6th plan for social and economic development as follows: 1. to improve population diets by increasing animal protein consumption, 2. to reduce animal products imports necessary to meet national needs, 3. to be able, in the long run, to export animal products in order to diversify export earnings and to improve the national balance of payment. Nation objectives also include the protection and 19 conservation of the environment. These goals are translated into government's policies and actions on pricing policies, taxes, production circumstances, and development schemes. The main strategy to develop the. Senegalese livestock sector is based on the stratification framework. Five eco-climatic zones with comparative advantages have been identified and each one assigned one or more of the layers and stages of the stratification process, i.e breeding cow/calf herds, growing out, fattening and processing. The sylvopastoral zone, in the northern part of the country ,is designated a breeding cow/calf zone. The following stage of growing out will be carried out in the groundnut basin which is endowed with substantial quantities of crop residues. The Senegal river valley and the Cap-Vert region will specialize in fattening operations. The different stratification layers and stages are planned to be conducted within the Casamance and Senegal Oriental regions. These strategies reflect an attempt to move Senegal toward self-sufficiency in red meat which will be achieved by increasing production. In fact, during the last two decades, per capita meat consumption has dropped from 21 kg to 13.5 kg per year and Senegal is a net importer of cattle and small ruminants (Roush et a1. ,1982). The per capita consumption of 15.7 kg targeted in the 5th plan for social and economic development was far from being achieved and has been reset for the 6th plan. 20 The diversity of people and institutions active in the livestock sector induces conflicting objectives . 2.3. Conflicting Objectives Conflicting interests in children enrolling at School and their contribution to the farm labor force has been already mentioned, this section will review contrasting perspectives between men and women, conflicting interests between cattle owner, manager, and herdsman, tradeoffs associated with meat vs. milk production, and those related to producer vs. consumer welfare. 3.3.1. Men vs. Women Men and women have different perspectives toward cattle. The head of family is more interested in the short term economic functions of cattle as input to agriculture and as readily disposable and convertible goods into cash or other valued things enabling him to satisfy family food needs, taxes and medical fees. On the other hand, women place more importance to the saving function of cattle. They accumulate livestock in order to ensure a bright future for their offsprings by providing them with cattle when they get married. Ironically, although women own most of cattle, rice production, a culturally designated role of women, does not use animal power. The ”male argument" being that it is not worth committing such a scarce resource to rice production that yields low output. 21 2.3.2. Meat vs. Milk Production Management practices of excessive milking have numerous adverse effects on herd productivity. It negatively impacts the reproductive ability of females and calf growth by respectively delaying the post partum estrus resulting in longer calving intervals and by deviating milk from calves consumption. An entrusting herd management exists betweem Fulani and Mandingo farmers or a particular farmer or a city dweller can entrust his herd to a herdsman paid in milk. This arrangement is obviously conducive to the maximization of milk offtake by herdsmen. The over-milking shortcomings and its adverse effects are minimized where the owner and the manager are confined to one person. Careful monitoring of milk offtake is then applied. 2.3.3. Consumer vs. Producer Welfare The objective of promoting farm production and incomes is contradicted by the desire on the part of the government to provide consumers with affordable meat products. In Senegal wholesale and retail prices of meat are set up by the Ministry of Commerce and the Ministry of Rural Development whereas there is no policy associated with producer prices. AS stated by Sullivan (1984) and Girardot- Berg(l982), government has promoted a pricing structure that works to the disadvantage of livestock producers. This pricing structure has economic and social implications which serve consumers to the disadvantage of producers as it leads 22 to low producer prices. The pricing policy for meat, as it stands today, has distributional benefits which accrue to urban sectors but works to the disadvantage of the agricultural and livestock sectors (Sullivan, 1984). The pricing policy causes a transfer of income from livestock producers to urban consumers and an under valued product. Serious policy conflicts are then apparent between improved production through producer price incentives and supplying affordable meat to city dwellers. It is not, however, obvious that the price incentive paradigm would work under current production circumstances. What would the livestock keepers response toward increased prices be? Are they going to increase animal offtake or engage in purchasing more inputs for improved herd productivity? AS observed by Delgado (1984) supply elasticity is very low in sub-Sahara Africa. A structural change paradigm consisting of improved research, input delivery and infrastructure systems is necessary. It is evident that the system this study has elected to examine has a multitude of objectives which can be in conflict. This system needs to be identified to find out its determinant variables and structure. That is ~. the purpose of the next chapter. CHAPTER III SYSTEM IDENTIFICATION This phase of the systems methodology identifies system input and output variables, and parameters that define aspects of the system structure as illustrated in figure 2. This is carried out after the system has been precisely defined. 3.1. System Definition " A system is defined as a group or a set of objects united by some form of interaction or interdependence to perform a specified function" (Shannon, 1975, p. 15). The system of concern in this study is the livestock sector of the mixed crop/livestock production system in Upper Casamance (Senegal) as illustrated in figure 3. Its physical boundaries coincide with the £1 and f2 sub-zones identified by SODEFITEX /SONED-Afrique(l980) that classifies the eastern Senegal and Upper Casamance regions into sub- zones based on soil characteristics and potential. It also overlaps with the administrative region of Kolda. Important features of the livestock system are typified by : - a wide variety of animal raised : Ndama cattle, 23 24 W I lavisonaeatal H :: (controllable overt top fl 2'38"") H H p H . Desired Outputs (uncontsollable overt low . Iyetee Deena tau-eta" — Ilov e! 'eeueelsty. seaoosoee 3333. "Week” musty. Figure 2 System Identiri.. . syStems-AnnroaCh (Feasibilifv O the Source: Manetsch, I977 25 Djallonke sheep and goats, equines and poultry, - an extensive animal production system with limited movement of animals. - an individual ownership of animals which are grazed on natural pastures communally accessible. - a multitude of goods and services provided by animals : milk, slaughter animals, power and manure. 2 The f1 and f2 zones occupy 13 718 km and support 206,870 people mainly Fulani and Mandingo who constitute 71 percent of the population (SODEFITEX/SONED ,1980). The population density is 18.7. 21.1 percent of soils ranking between ”medium to good”, are either cultivated or fallow (Landais,1984). The system variables and parameters are categorized into exogeneous inputs, overt inputs (controllable and uncontrollable), desired outputs, undesired outputs, system design parameters, management inputs and system performance criteria. System design parameters are first examined to secure a coherence in the structure of the study. 3.2. System Design Parameters These are relatively fixed variables that serve to specify the system structure. ”However they can be subject NAT ION AGED- PASTORAL ZONL VILIAGE AREA FARH L . ‘ T ! URBAN I [POLICIES | DESIGN 0' i RESEARCH H! A? | - Input! or vuor'rumI Opium-110:4 I DEM-wI Output SlQATEGIES l l . _-'—,‘ pm es ——fi—a , : .‘taxE' I 41 I ' ”Sub‘ it“! ‘ I . .. — -|. . - Iahd i I I l ' L Lawn I ' l | -——- - I I I ‘ ' l I I , I ' I i l L I l _—-—-- ' WYSIUL I mount NTAT ION RISE chu I ' ENKIE‘U'HFNT IO! DE JILOPHL-Ill ACTH/Inge ' I -cIWHate SCHEMES BY AND ' I I ~uat er DE VELW Hi NT OUTCOMES ' ., resources INSTITUTIONS I / FT- " ”T 7 I I I | I ’ I I I I ’ I | I 1 I ’ I | I I / I . ' ' IT lRADITIOMALI . I / 1 I I L‘ND IENUQ‘E; ‘ I / I l I svgvgn I f ‘\3 / I I l . _J / .”~ , I I ’ ' " I I l I '. .’ I I 'f : I PASlUDEL . _. 1_1°. icuor ' I Law. 5 , .' 7’ lL I”? ' ' I 3 I ;‘ ANDS I ' | {I x’ I | OTFTAVFI . . , ‘eedl \ ' . I, I I I HI’ I I I 26 \ "” lie ' ' pu-r // ’Iflanur 9 ___L1 . labor .aninu' husbandry . 553?‘-‘—-- ~------- - l :1 -l Hrun’slMu—I 2 / --------Luouss‘uom T— I I l I I . ‘ I ANIFA1$ F CROPS I .cattle ~—-\,r———-—J' .millet , .theep | _ .eorpwum | .gost | .By-Droducte _ .rice ' .horsr I .resldue: .melze " .donkey , .cotton ' .poultry . .other: . I . i I I __l ' (a I Ital I Figure 3. A General Description of the Agropastoral Production System of the Upper Casamance 27 to alteration for system improvement in a changing environment" (Manetsch,l977). Livestock species, herd sizes, and the spatial and temporal dimensions of land use can be included in such a class of variables 3.2.1. Livestock Species and Herd Size Ruminant species in Upper Casamance are Ndama (bos taurus), a long horn type of cattle, and Djallonke sheep and goats (West African dwarf sheep and goats). The outstanding attribute of these breeds is their trypanotolerance. Trypanotolerance, however, may be a misleading term since it does not mean immunological tolerance (Toure,l977; Karbe,l980) The latter meaning that the animal's system does not recognize the antigen as foreign whereas a trypanotolerant host does. Trypanotolerance is then defined as a resistance toward the disease without resistance toward infection (Karbe,l980) Trypanotolerance is not an absolute attribute. Any stress, e.g., poor nutrition, work overload, can break down the balance between host and parasite consequently causing the disease out break. 28 Table 1. Cattle herd size and structure in the three districts of Kolda. Animal class Percent Dabo Dioulacolon Yorofoula ( years) Male < 1 10.5 6122 3212 5945 1 5.3 3090 1622 3001 2 4.3 2507 1316 2435 3 3.6 2099 1102 2038 > 3 4.6 2682 1408 2605 Total male 28.3 16500 8661 16024 Female < 1 10.0 5830 3060 5662 1 6.0 3498 1837 3397 2 6.0 3498 1837 3227 3 5.7 3323 1744 3227 4 7.3 4256 2234 4133 5 11.6 6763 3550 6569 > 5 25.1 14634 7682 14212 Total female 71.7 41802 21944 40427 Total 100. 58302 30605 56451 Compiled from SODEFITEX, 1980 There is a growing number of equine species known as trypanosensitive in Upper Casamance. The most likely hypothesis explaining this phenomenon is the 1970's series of drought, coupled with the clearing of new lands which led to a reduction in: tsetse flies density in the area of study resulting in a lower trypanosomiasis risk. Table 1 presents the cattle herd size and structure in the three districts of Kolda. BOAD(l984) indicates that an average family of 7 members in the Bonconto zone has a 29 livestock herd composed of 9.26 cattle, 4.46 small ruminants, and .72 equine. Family herd size varies from 0 to 72 heads and 62.5 percent of households own less than 9 heads of cattle. 3.2.2. Land use A central aspect of the agropastoral system structure is the spatial and temporal organization of land use. A fallow system of crop production is practiced. Soil fertility is restored through bush and/or grass fallows. The most noticably changing aspect of the agropastoral production system is the reduction of fallow periods due to expansion of cash crop cultivated areas and a rapidly growing human population. Alteration of fallow production system in humid zones of Africa has been proposed (Kang, Wilson and Sipkens, 1981) to overcome this problems. Alley cropping to replace the fallow system is advocated. It combines the growing of forage and/or fertilizer tree legumes with the production of cereal on fallow lands. This issue will be dealt with later. 3.3. System Desired Outputs In general, desired outputs are the system's response to the needs specified in needs analysis. In other words desired outputs are fulfilling or satisfying identified needs. (Manetsch,l977). The livestock sector plays a key role in agropastoralists revenues and welfare in 30 Upper Casamance. Yearly revenues from animals account for 30 to 53 percent of the farmers total cash revenues Landais (1984). The varying percentage of livestock contribution to farmers' revenues reflects their strategy in resorting to livestock assets in order to balance unpredictable revenues from cropping activities (Landais, l984).Income generated from livestock could have been greater than the above percentages had other functions such as energy and manure inputs to agriculture been taken into account. Desired outputs can be extrapolated from various functions of livestock as supplier of goods and services: 1. Milk production 2. Meat production. From farmers' standpoint, slaughter animals is a more appropriate term to designate this output. Apart from particular occassions such as religious ceremonies and weddings, farmers seldom eat meat from ruminants. Cash generation may better fit farmers perspectives. 3. Energy input agriculture 4. Manure production These outputs will be examined in detail in chapter 4. 3.4 System's Undesired Outputs Besides the above mentioned desired outputs, the livestock sector also yields unintended outputs. Two aspects of such side effects are the potential rangeland 31 degradation through overgrazing and damages to crops generated by livestock. Rangeland degradation in African pasturelands has been a controversial issue. The mainstream view described by Sandford (1983) stresses the alarming situation of rangeland resources that are suffering severe and rapid desertification. Degradation is the result of overgrazing induced by increased density of livestock. Livestock population explosions were in turn made possible by the eradication of major devastating diseases such as Rinderpest and Bovine Contagious Pleuro-Pneumonia. However, according to Sandford (1983) arguments supporting this mainstream view are very weak. The leading theoretical ground of this view is the "Tragedy of Commons" of Hardin (1968). This theory explains overgrazing as a result of unregulated access to common rangeland resources where private and public interests are in conflict. The problem of overgrazing is said to arise due to a difference in the incentives facing individual livestock owners and the costs and benefits the pastoral society as a whole resulting from grazingland. A body of evidence (Stryker, 1984; Sandford, 1983; Horovitz, 1980) militates against this theory. In fact, the access to land is regulated by traditional institutions such as well ownership. More importantly, Stryker (1984) came to the conclusion that: 32 " the elasticity of supply of herds major inputs i.e. labor and capital, appears to be relatively low and the opportunity cost associated with their use, to be, at the margin, high. This contradicts the common assumption that pasture is the scarce resource and other traditional inputs are available at minimum cost." (Stryker, 1984, p. 182) Boudet (1970) has depicted an alarming situation of high cattle densities in certain districts of Upper Casamance i.e Dioulacolon, Koukande where the actual stocking rate was double the pontential supporting capacity of the pasturelands. One could have expected severe livestock starvation and death occurring because of overstocking. Despite the fact that there has been no external feed supply to the system, that did not happen. As it stands today, due to the present state of knowledge on animal statistics and rangelands accurate conclusions about the rangeland resource condition cannot be drawn. Livestock can directly affect crop yields by eating plants. They have, also, been incriminated in deteriorating soil structure by hardening top soil as a result of animal trampling crop lands. The direct damage of livestock on crops is very well controlled in Upper Casamance because of heavy penalties borne by owners of wandering animals. There are other issues that have potential adverse impacts on animal production. They are related to ongoing and planned agronomic innovations and are : - plowing under crop residues at the end of the 33 cropping season. - dry season rice production - large scale use of animal traction for farm mechanization to boost cotton production Animals would be deprived of substantial quantities of dry season feedstuffs if the technology of plowing under crop residues, aimed at increased soil fertility, is adopted Valleybottoms constitute valuable dry season grazing lands. The use of these lands for dry season rice production would aggravate the nutritional stress by increasing stocking rates in other pasture lands. The wide scale animal traction use leads to an expansion of cultivated areas and a reduction of fallow periods. At Lenguewal village, Achterstraat (1984) reports an area expansion of 17 percent between 1980 and 1981, mainly at the expense of fallow lands. The fallow period that used to be more than 7 years has been reduced to 2 years or less. Ange (1984) indicates that cultivated areas have doubled during the last decade. Apart from negative agronomic effects that could result from inadequate soil fertility restoration, feed availability could be significantly reduced by the encroachment of crop lands on pasture lands. However, agronomic innovations would lead as well to increased crop production which means increased availability of crop residues. The positive effect of such increased crop production regarding animal nutrition would depend on the 34 land allocation to different crops. If cotton production occupies more areas as is the actual case, the increased crop by-product would not benefit animals as cottonseeds are exported out of the system. 3.5 Exogenous Inputs According to Manetsch (1983) exogeneous inputs are environmental variables that have a strong influence on system performance and dynamics and, in turn, are not significantly affected by the system. Weather related variables and feed resources fall into this category of variable. 3.5.1. Weather Variables The Upper Casamance zone has a dry tropical climate of ”Low Guinea” type characterized by the sudden occurance of a 5 months long rainy season followed by a long dry season of 7 months. The determinant weather variable that regulates crop and pasture productivity, is rainfall. Since the early 1970's the zone has experienced a very dry period as a result of the occurance of a series of droughts. The mean annual rainfall dropped from 1216 millimeters (1922 to 1954) to 907 millimeters (1974 to 1980) (BOAD, 1984). The average annual temperature is 27.7 degrees Celcius with a maximum of 34.9 degrees Celcius (April, May 35 and October) and a minimum of 20.4 degrees Celcius (January and August). The average annual relative humidity is 88 percent. 3.5.2 Feed Resources Feed resources are analyzed under the category of environmental inputs because their primary availability is dependent upon uncontrollable weather variables. For convenience, crop residues are included under exogeneous variables in order to prevent a fragmented analysis of feed resources. Natural and crop residues are the primary feedstuffs at the disposal of animals. Their productivity and quality is subject to seasonal and annual fluctuations. Pasture Lands Boudet(1970) did a comprehensive description of pastures in Upper Casamance. According to BOAD(l984), despite the series of droughts of the 1970's, pasture composition did not undergo any major modifications. However, plant densities have declined. Table 2 indicates the different types of pastures and their productivity. Fallow areas, considered the best grazinglands, can be at different phase of evolution, from a recent abandoned crop field to a woodland or clear forest. It takes 30 to 40 years to reach its ultimate phase of clear forest. Fallow areas are used in the rainy season while hydrophile-grass pastures and valleybottom pastures are dry season 36 grazinglands. Other pastures like 'bambou pastures', sciaphile and heliophile-grass pastures are used in both seasons. Different types of pastures , according to Boudet's (1970) classification are portrayed in figure 4. Crop Residues. Animals freely graze residues on crop fields after harvest. Substantial amounts of energy are derived from crop residues as illustrated in table 3. Straws from millet, sorghum, rice, and groundnut are the essential crop roughages remaining after the main products have been harvested. As maize is harvested in September while pastures are still green, maize straw was not included in the feed pool. Since animals are not yet allowed access to crop lands it can be assumed that most important quantities are lost. Feed Quality The typical pattern of growth in range grazing consists of a rapid growth during the wet period , rapid maturation of the herbage and long period with mature material in which no further growth occurs (Balch,1976). As illustrated in table 4, pastures forages have energy and protein contents that decline dramatically to very low level in dry season standing hay. Voluntary intake and follow the same pattern. 37 Table 2. Pastures at Kolda: Types, Productivity and Season of Exploitation. Pastures Site Predominant Productivity Season Plants consumable of exploi- kg DM/ha tation RS DS RS D8 A1: Annual Andropogon grass past. U auriculatus 4. l. + "Bambou" pasrures A2 U Ostrideris stulhmani 1. .5 + + 82 U Diheterepogon Pl U Andropogon tectorum Pennisetum subangustum 1. .5 + + Sciaphile grass past. P2 U Pennisetum herdeoides 3. .5 + + P3 U Paspalum auriculatum 3. .5 + + P4 V Becheropis uniseta 3. .5 + + Heliophile grass past. B1 U Diheteropogon amplectens Schizachyrium sanguineum 2. .5 + + BB U Diheteropogon amplectens Andropogon tectorum 2. .5 + + S3 U Diheteropogon ampkectens Pennisetum atrichum 3. 1. + + 81 V Andropogon tectorum 3. 1. + + Vl V Androlpogon gayanus 3. 1. + + V2 V Andropogon gayanus 38 Table 2. cont'd Diheterepogon amplectens 3. l. 82 Hypparhenia dissoluta Pennisetum atrichum 3. l. Aquatic pastures E1 Anadelphia afzeliana E2 Anadelphia Valleybottom pastures Brachiaria mutica U: upland, V: valley, Bambou: Oxythenanthera abyssinica Compiled from Boudet,l970 39 Table 3. Estimation of Energy ( Mcal ME ) Supply from Crop Residues Millet/Sorghum Groundnut Rice Yield T/ha (l) 1.0 2.0 3.0 % DM 85.0 90.0 85.0 ME Mcal/kg DM 1.13 1.13 1.25 Ha. cultivated Dioulacolon 11000. 9750. 3500. Yorofoula 9900. 13250. 1750. Dabo 10250 12500. 2550. Total ME (1000 Mcal) 2 Dioulacolon 5282.75 15865. 5578.1 Yorofoula 4754.5 26950. 2789.1 Dabo 4922.5 20340. 4064.1 l: straw yield , fine parts (tops,1eaves) for sorghum and millet. 2: a recovery rate of 50% for cereal straws and 80% for groundnut hay have been assumed to allow for losses. Compiled from SOMIVAC,1978, SODEFITEX, 1980 40 San! Inn: a.HU . (gas . :nmu him-mm . ‘3 0m . ~¢\\\\\ Egan‘s IE: 0 =95 .. am 238...: (null-U. 393:3...1 WD :93 :SSCuo 3! mouse .0353 188‘: ' E. o>m_ .umUDOm "OOLDOm oco.cco._\_ um_m0m v mocmemmmu m_UU_z Dcm Loud: or» c. mmtnummd m0 co_uoltommo .v mtjmwu 41 Digestibility is a crucial influential factor of forage utilization. It is dependant upon the intrinsic physical nature of the forage ( Pearce,1982) and upon rumen microflora activities. Rumen micro-organisms require certain levels of energy, protein, and vitamins for optimal growth. Shortage of these nutrients lead to inhibited bacteria activities and consequently reduced digestibility. The cellulose and hemicellulose of plant cell wall are digestible but lignin, silica and silicates as well as other cell wall components reduce their accessibility to attack by the rumen microflora. Nutritional constraints imposed by the rumen functions and associated with roughages from pastures or crop residues have been stressed by Preston(l982). The rumen functions cannot fully supply the essential amino acids and glucose precursors with high levels of production. “Microbial protein synthesis in the rumen approaches adequacy only for late growth and mid-pregnancy" (Preston, 1982). High levels of productivity require supplementation of preformed protein and glucose precursors, in such a form that arrive, in part, at the sites of metabolism. (Preston,1982). 42 Table 4. Pasture Quality Class Mcal/kg DM "MAD" Dry Rainy g/kg DM season season 1 2.519 75 aquatic fallows past. 2 1.889 45 valley- Al'VI'VZ'SZ’ bottom Bl,BB,P2,P3 P4,S3 3 1.731 33 E2 A2,BZ,P1 4 1.574 25 V1,V2,Sl, $2,El 5 1.417 25 Bl,BB,P2,P4 'MAD':'Matieres azotees digestibles'I : Digestible nitrorgen matters Compiled from SOMIVAC,1978; Boudet,1970 3.6 Controllable Overt Inputs Controllable overt inputs are those that can be managed to alter performance. Since we are dealing with a low-input system, inputs that have the potential to bring about appreciable performance change, will be considered. They are basically related to the extent at which investment in animal feed supplementation, veterinarian supplies, research and extention could be undertaken. It is difficult to quatify the impact of research and extension on system performance. However, they are included in this class of variable because they are policy variables which have the 43 potential to impact the performance of the system in the long run. The livestock sector in the Upper Casamance has been a domain forgotten by policy makers. In contrast to other ecological zones, the Upper Csamance did not receive any major public investment geared toward its development until in 1985 when a livestock component was introduced into the SODEFITREX's programs. This project is essentially designed to secure animal power input to cotton production. Taxes on cattle heads have been abolished since 1970 but the legacy of this policy is still alive. Farmers perceive any livestock population census as an attempt to restore taxes. In contrast to the crop sector, credits are unknown to livestock producers. Government policies regarding land tenure and animal commodities pricing were mentioned in chapter 2. However, free annual vaccination campaigns are executed to combat major deseases like Rinderpest and Contagious Bovine Pleuropneumonia. The afore-mentioned SODEFITEX‘s livestock project intends to reinforce vaccination programs. Despite the positive effect of health programs, disease hazards such trypanosomiasis, internal and external parasites and tick-borne deseases are still prominent limiting factors of animal production. Better knowledge about the desease profile is necessary to set up health care priorities. 44 Animals do not receive any extra feed except some mineral supplementation consisting of ordinary salt (Nacl) or a traditional recipe composed of roots and barks. The necessity of an economically sound supplementation alternative cannot be overemphasized if productivity is to be improved. Investment in research needs to be increased and better oriented. The significance of ongoing breeding programs on the livestock system performance is doubtful for many reasons. The annual number of bulls generated by the station is very limited relative to the size of the cattle herd in the region. High mortality rates and low reproductive parameters as noted in chapter 4 are not compatible with rapid genetic progress. The overriding factor that hinders genetic improvement programs is the low level of nutrition prevailing at village level. Improvement by breeding has little value unless supply of feed energy are equivalent to at least 1.5 times maintenance needs (Mcdowell, 1977). Research activities should place more emphasis on the design of alternative feeding systems applicable with village herds and on issues associated with herd and pasture management. The marketing system also deserves particular attention. Other essential areas also need a great deal of investment. The reinforcement of the extension service with personnel and financial means is fundamental for adequate technical assistance to farmers . A fire control strategy, 45 for instance the development of fire breaks which could serve as roads the same time, is deemed necessary in order to combat the devastating effects of fires on dry season feed availability. The careful development of new watering facilities would help reduce the dry season feed pressure through a more efficient use of rangeland resources. 3.7. System Management Some aspects of pasture land use have been already mentioned, this discussion will therefore focus on the following aspects of herd management: grazing time, animal husbandry, oxen management and marketing. 3.7.1. Herd Management An important aspect of herd management in the Upper Casamance context is the intervention of many operators. Cattle owner, herd manager and herdsmen all have control, to a certain extent, over herd management. The village herd can be partitioned into sub-herds belonging to a particular family or to a group of people with particular affinities. Cattle are owned by individual members of the family or of the group. Frequently the cattle owner is an individual external to the village. Economies of scale dictate the gathering of small size individual herds under a commom management. Herd grazing labor requirements are usually hired. These features of cattle ownership and management 46 are important because of the conflicting interests and the complexity of the decision-making process they generate. Part of the herdsmen renumeration is in milk of which he extracts as much milk as he can, no matter the adverse consequences on herd productivity. The herd manager needs the permission of the cattle owner who may be far away to undertake any important decision on herds. The manager is usually unwilling to incur expenses such as veterinary or feed inputs for animals other than his own. This delayed and complex decision-making process is a serious impediment to the introduction of innovations. Overall pasture forage intake is limited by reduced grazing times. Animals are released around 10:00 AM after the milking of lactating cows and tethered again at about 6:00 PM. Because of high temperature occuring at certain periods of the year these 8—hours grazing time can be significantly reduced. Calf weaning occurs late. Castration is not systematic, nor is there a breeding season. Animal mate throughout the year. However, there is a natural grouping of births; 86 percent of calvings occur between June and December (CRZ/Kolda, 1983). No housing facilities exist for cattle. Sheep may have shelters established within the homesteads. New born calves are, however, exposed to rain, high temperature and heavy parasitic infestation and therefore have high mortality rates. 47 Oxen Management Sound oxen management is essential for an appropriate animal energy input to agriculture. Important aspects of oxen management deal with nutrition, health care and age at time of sale. Oxen are returned to the cattle herd after the cropping season. They go through the nutritional stress of the 7., dry season which adversely affect their traction power at the onset of the cropping season. The required health care such as deworming and chemical treatment against trypanosomiasis, essential for an optimal energy output, is seldom provided. This is mainly due to the lack of veterinary supplies and extension service staff. Farmers are otherwise aware of the potential benefits of such practices and are willing to take on these health care expenses. The holding age of oxen is a determinent variable regarding trade-offs which exist between meat and energy outputs. Different management options have been advocated. A short oxen holding period of three years is viewed as an appropriate fattening scheme that could promote the rate of meat output from the system. Longer holding periods of 7 years have been also suggested. Advantages and disadvantages are attached to each of these options. Early age of oxen sale is likely to yield more slaughter animals and to enhance farmers revenues through increased animal sales. The effectiveness of such a policy 48 hinges upon the capacity of herds to generate draft animals and the ability of cattleless farmers to acquire credit for the purchase or rental of oxen . A strong argument against this option is that animal do not perform efficiently in the early period after training and have light weights. Therefore, heavy plowing energy requirements would be difficult to meet. Longer holding periods do not display the above disadvantages and reduce capital requirements for oxen acquisition. Considering the large number of oxen needed in the zone, there will be a long delay process associated with the male animal slaughter age. Optimal slaughter age of draft animal is difficult to set. It will depend on individual farmers' circumstances, meat prices and farmers ability to equip themselves. Due to poor herd reproductive performances and high mortality rates, the second scenario of long holding periods is likely to prevail. Offtake SOMIVAC(1978) and De Reviers(1979) indicate offtake rates of 8 and 6 percent respectively including autoconsumption and sales. There is no planned pattern of animal offtake. Livestock are marketed when farmers confront an urgent problem requiring liquidities, e.g., food shortage, medical fees. Depending on the degree of the problem either small 49 ruminants or cattle are sold. Livestock can be traded at village level or at weekly collection market where intermediaries or ”diula" intervene to collect animals to supply large consumption centers. Due to a lack of information , the efficiency of the marketing system is difficult to access. Costs incurred and benefits gained by different operators at the different stages of the marketing process, are not available. However, it is a general concensus in the livestock marketing system in Senegal , that the producers are the big losers. They are very vulnerable and do not have a great deal of bargaining power vis-a-vis intermediaries. Prices agreed upon may be paid at the time of transaction. Frequently farmers get their money a long time after transaction and often partially. They may never receive the remainder (De Reviers,l979). Long distances from consumption centers, unorganized farmers and poor 11 access to certain villages contribute to the difficulties of farmers in profitably marketing their commodities. Any technological innovation aimed at the improvement of the animal production system should be concomitantly carried out with the reorganization of the marketing system. Farmers should organize to increase their bargaining power. If the returns expected from innovations are to be depressed by an inefficient marketing 50 system, farmers incentives to adopt them will be reduced. 3.8 System Performance Criteria These criteria are set to evaluate system alternatives (Manetsch,l977). The system dealt with has a multiple of outputs and many operators. These criteria should reflect as much as possible the objectives of individuals or group interest. Potential criteria that could be used to assess the likely effect of system alternatives on system performance are as follow: 1. Number and weight of slaughter animal marketed. The cyclical pattern of livestock marketing should be taken into consideration. This criteria is set to allow for the meat output function of livestock reflecting national and farmers objectives. 2. The quantity of milk produced per unit of time 3. The amount of animal energy supplied to the cropping sector. This can be expressed in term of horse power produced . Such an expression may be more appropropriate than the number of animals since it incorporates animal weights. 4. The productivity of herds on a per cow basis or on per cow weight basis. This criteria 51 includes cows' offspring weight, milk yield, reproductive performances and mortality. It is a holistic criteria that enable the comparison of alternative technologies on overall herd productivity. 5. Return to labor may be used to evaluate the profitability and potential acceptability of innovations by farmers. 6. Return to land would reflect the efficiency of rangeland resources utilization. 7. Farmers' revenues need to be considered. Since the ultimate goal of improving the system is to enhance farmers' revenues and welfare, the contribution of livestock on farmers' income is to be evaluated relative to other sources of income. 8. The rangeland conditions is also crucial to consider since it determines the long term capability of range lands to support animal production . Numerous difficulties accompany an evaluation of these criteria. The basic information related to herd productivity ( animal weights, milk yield, reproductive parameters), which are necessary to evaluate such criteria, will require a great deal of resources, time and monitoring for their availability at the farmers' level. Futhermore, as pointed out by Bersten(l983) livestock systems are 52 characterized by nonmarket inputs and outputs which do not lend themselves to easy evaluation. CHAPTER IV PROBLEM FORMULATION Constraints handicapping animal production in Upper Casamance can be broadly classified as ecological, biological, and socioeconomic. (Bersten et al., 1983). Ecology impacts livestock through many factors including climate ( rainfall, temperature, day length), land form and soils, and vegetation (Pratt, 1984). Biological factors include disease hazard and animal genotype. Socioeconomic constraints are associated with management practices, input/output prices, marketing, policies, and labor availability. (Bersten, 1983) Factors of overriding importance will be examined in this chapter. Two models have been developed to help identify major problems facing the livestock sector in Upper Casamance. The first is a nutritional budget using microcomputer spreadsheets. The second is a simulation model. Certain important variables could not be included in either of these frameworks because of the scarcity of information. However, a reference to these variables will be made in this discussion. 53 54 4.1 Nutrition Budget 4.1.1 Description For the purpose of this study a nutrition budget Spreadsheet has been developed for each of the three districts of Kolda: Dabo, Dioulacolon, and Medina yoro foula. The objective of the nutrition budget is to evaluate the extent of the nutritional problems for each district. It will also help in designing an appropriate feeding system. The analytical framework used has been adopted from Hart, Onim, Russo, Mathuva, Otieno, and Fitzhugh. (1984). They describe the framework as a nutrition budget that follows the flow of feeds ( in metabolizable energy units) from different sources to a pool of available feed that is drawn upon by different livestock herds. Feed demand and offer are evaluated on a monthly basis for a year. Pasture areas, productivity, and metabolizable energy (ME) content are used to evaluate total available energy in megacalories (Mcal) on a dry matter basis. Energy from crop residues is also included. Total energy from different sources is, then, summed up each month to give the monthly available energy. Feed demand is evaluated on a monthly basis for each livestock herd ,i.e. cattle, small ruminants, and equines. The cattle herd is partitioned into cohorts according to 55 sex and age. Energy demand for each cohort is computed each month taking into consideration maintenance requirements and different functions: pregnancy, lactation, work. The energy balance is then calculated by substracting demand from offer. 4.1.2 Technical Assumptions In the next discussion assumptions and assertions about the nutrition budget will be presented. Only ME is considered for the evaluation of feed demand. The underlying assumption is presented here below: " Organic nutrients from different sources of feed available to an animal are used for a variety of purposes, including the maintenance of body functions,the construction of new tissues, the synthesis of milk , and the conversion to mechanical energy used for walking and other work. These diverse functions all require the transfer of considerable quantities of energy , so that in most situations when the energy functions are met it may be assumed that the animal non-energy requirements(protein, minerals, and vitamins) are also met." (Kondadreas et al.,1982,p.14) Grazing areas (in ha), productivity (in kg of consumable dry matter) and nutrient contents ( in "UF": "unite fourragere" ) are derived from Boudet(l970 ). Pasture land areas are updated here to take into account the expansion of cultivated areas., SOMIVAC(1978) indicates that cultivated areas have expanded at least 66 percent in Lower Casamance. Since this information does not exist for Upper Casamance a figure of 70 percent is assumed. This estimate, 56 which may appear understated, is based on the fact that arable land expansion was facilitated in the Upper Casamance because of rapid development of cotton production. Feed quantity and quality reported by Boudet (1970) are used. Using Boudet's values is both bold and of questionable validity. Yet in the absence of updated values, they still remain = baseline values. Cognizant of the fact that even if the pasture composition have not significantly changed since the 1970's, appreciable changes in plant densities may have occurred due to the decrease in the average annual rainfall since the early 1970's. The above mentioned report gives feed quality and quantity on a seasonal basis. In this study, seasonal values are applied for each month of a season. Following the example of SOMIVAC (1978) fallow areas have been given a small feeding value in the dry season. Feed from crop residues have been evaluated from hectarage and yield of different crops. There are great variations of crop yields among years because of climatic changes. The energy derived then from crop residues for a year is indicative of quantities available. The amount of crop residues available is evenly distributed throughout the dry season starting from the month of harvest of that particular crop. The model assumes that feeds of different quality are combined into a pool that is drawn upon by cattle, 57 sheep, goats, horses and donkeys. This assumption might be questioned since it is likely that livestock species have different feeding pattern and type of feed base (Hart and al., 1984) Feed demand is evaluated based on actual animal weights and consequently variations in demand are taken into consideration to allow for seasonality. The pattern of weight change across the year is derived from trials reported by Gueye et a1. (1983) and CRZ/Kolda(l983). Weight fluctuation of control groups fed only on natural pastures were used to project weight variation in village herds. This projection appears valid since animals used were among the initial herd of the station (external then to the station) and did not receive any feed supplementation. Table 5 illustrates weight fluctuation relative to mean annual weight at the research station at Kolda. When highest and lowest monthly weights are compared, animals lose 19 percent of their body weight during the dry season. 58 Table 5. Monthly Weight change of the Foundation herd at Kolda Station. Months n Mean weight(kg) Percent weight change January 59 250 1.6 February 58 249 1.2 March 58 238 -3.2 April 54 227 -7.7 May 53 226 —8.1 June 51 217 -ll.7 July 52 235 -4.5 August 52 255 3.7 September 52 264 7.3 October 51 268 9.0 November 51 263 3.6 December 49 254 3.2 Source: CRZ/KOLDA,1983 4.1.2 Results and Discussion Table 6 presents the energy balance for the three districts of Kolda. Detailed spreadsheets are given in appendices A, B, and C. Figure 5, 6, 7 indicate energy demand, offer and balance. As it could be expected, the feed energy available 59 exceed the animal requirements during the rainy season. In contrast, a negative energy balance throughout the dry season in Dabo, Dioulacolon and Medina yoro foula is observed. Despite the fact that weight loss has been accounted for in the evaluation of energy needs , there is a negative feed balance. This means that the assumed weight loss is underestimated and that animals in the traditional setting lose more than 20 percent of their body weight during the dry season. Other feed resources such as tree leaves and fruits have not been accounted for in the feed pool. Examples of important trees that supply feedstuffs in the dry season are Acacia albida, Pterocarpus erinaceus, manguifera indica, and Khaya senegalensis. However, their input is offset by pasture forage losses caused by fires. According to BOAD(l984), fires destroy 10 to 20 percent of the dry season standing hay. Such losses were not considered in the nutrition budget. The energy deficit can be converted into an equivalent number of animal units which would express the number of animals that exceed the stocking capacity of rangelands. The total number of tropical livestock units(l TLU is a standard animal weighing 250 kg) that are in excess are estimated to be 12347, 6680, 8665 for Dabo, 60 ,_moxcmucm.mm >Omeu A.m0:vucmEOQ xmtmcw A_muzvtmwwo >omem m_. mm..- mN.mu mm.m_ mo.o_ mm.m_ _o.mw oo.m hm.mu LODEmOmo an.rn ¢N.ma am.mn w_.m. -.o_ w_.m_ hm.m_ Nm.n m_.m LmDEm>Oz hm.mm 00.0, o>.m_ mN.NN no... mm.NN __.®v vm.- vm.mm Lmoouuo om.mm mv.o_ Nm.mm _w.NN uo.~_ .0.NN _..mv vm.m~ um.mm .Emuomm vv.m~ mo.Nm >N.o. 00.0. ov.o_ mm.m_ __.mv vm.- vm.mm umama< oo.mm M>.N_ Nv.hm Nh.m_ mm.m Nh.m_ mm.mv oo.- mm.om >_Do mN.NI mo. mm.ms om.m_ mm.m om.m. bo.m. 0v.m ho.Nm mcao mm._s v_. mm.ml wo.m_ ~m.® ww.m_ so.m_ wv.m po.m_ >0: v<.~n o_.n mo.v: N_.w. hm.m ~_.w_ hw.m. mv.m ho.N_ ..LO< mm.ma No.| o¢.m| no.0. mm.m mo.w_ so.m_ ov.m ho.~_ LOLmz mo.m| mv.l om.vn m>.w_ om.m m5.w_ ho.m_ mq.m so.~m .Dtnvu mv.mu __.I wo.vu mm.w_ mm.m m_.0_ hw.m_ wv.m so.N_ .mncmo .mc.omz .m.:o_o ODmo oc.Omz .o.no.o GOOD .OC.OO: .m_no.o coma mcucor 5.00: c0_.__zv mocc.t; U:O UCOEOQ >mLOCm Lowwo >mtocw >_£pc0t 0 263 61 Figure 5. Dabo Energy Balance (Energy Oomnmllofler, and Enhance) 4t) :35 -J ”cal . (Million-I 9.. J- ;l_¥ .t__ u I! I '10 r I I x l I I 7 I J’ ' N A N J J A 8 0 N ' "onfhs D ”on! 0“" + Meal Demand 0 Net M001 Figure 6. Dioula. Energy Balance (Energy Offer; Demand, and Galena) 14. d I ? . u “..‘I‘ .M-*~ ’ \ — \.°_-" . 13 M \ £2 -.-.. I’ -. a: " w I ________ .v,____,’_,___, -.4-VJ¥ \ ) / I Months 0 Energy on" 0- Energy Demand 0 Nat. Snug 61 Figure 5. Dabo Energy Balance (Energy OomandLOffer, and Balance) (I) 35 ‘ SID - «£5 - .. 20 -‘ 3 —-'-‘. _ _ 3:3 15 4 M‘— "—'—" ' 3f: .. -. .1 a r u I! r 10 1 5 - o - 4 -5 35*? F ‘10 r I r I I r r v I J 7 H A N J J A 8 0 N ' Numb! a Non! Ono-r + Meal Demand 0 Net loo! Figure 6. Dioula. Energy Balance (Energy OflcrLDemand, and Balance) 2U - I \ 1H . ..l 1‘3 .. I 14, d I 0.. . " “.b' . 'M-"‘\ I ’9. ~- g ”,4- ant-Q 33 , l x I, a: ‘ '7” '. 2 1 w n Months I: Energy om.- + Energy Demand 0 Nu Enorv 62 Fi ure 7. Dabo Energy Balnace (Energy Offer, Demand’ond Balance) 50 4O 30 8 Meal. (Millions) 10 ‘l d P (D O -10 L) Net Meal 4.44 Months Meal OHM 1» Meal Demand 0 63 Dioulacolon, and Yorofoula respectively. These values are equivalent to 26, 14, and 19 percent of the total livestock biomass ( cattle sheep ,goat, donkey,horse) respectively for the three districts. This dry season feed restriction has dramatic consequences on the level of animal productivity. 4.2 Productivity levels 4.2.1 Milk Production Milk output is influenced by a host of factors such as the genetic potential of animals, the environment reflected by year and season of calving dictating the nutritional plane, the stage of lactation, the lactation number and the length of lactation. Season of calving has an overriding impact on milk production since cows are not supplemented (Nicholson, 1984). "The year of calving and season of birth affect both total milk offtake and total yield, while daily offtake and yield are function of season, stage of lactation and lactation number" (Nicholson ,1984, p. 24) The pattern of milk production described by Nicholson (1984) is consistent with that of Upper Casamance. Generally, Ndama cattle yield low milk output. Milk yield on station averages 588kg(206 days) in Ivory Cost, 488 kg(234 days) in Siera Leone and 416 kg(266 days) in Senegal (Landais,1984). Williamson and Payne(1980) report a range of 64 150-270 kg(150—300 days). These figures give an average daily production of 1.8 kg which is close to 2 kg per day reported by Toure(1977). However this does not reflect the real milk potential of Ndama cattle. Casse et al(1965) report that 800 kg a year have been obtained with selected cows. Selection with a high nutritional plane allowed higher yields. Some cows produced up to 1250 kg in 210 days. BOAD (1984) estimates that, in Upper Casamance, cows produce about 400 kg per year out of which 60 percent or 240 kg are taken for human consumption. SOMIVAC(1980) estimates offtake to be 160 kg per year per cow. Table 7 presents an estimation of milk consumption at Lenguewal village. 65 Table 7. Estimation of per capita milk consumption at Lenguewal, in Upper Casamance. Milk offtake kg/cow/year 240. Number of cows 154. Calving rate % 46. Mortality rate (0-1year) % 25. Total yearly milk offtake(a) kg 15240. Number of people 340. Daily milk consumption 9 125. a: a lactation length of one year is assumed with 25% of cows having a half lactatation length. The average daily milk consumption is 1259. SOMIVAC's (1978) estimate for the Upper Casamance is 1009. This average figure does not depict seasonal variation in milk consumption and variation based on ethnicities and family herd size. Despite the importance attached to milk offtake, consumption is low and provides for only 49 of animal protein per day which is equivalent to 20 percent of daily animal protein requirements. 3.2.2 Meat Production The factors influencing meat production are the genetic potential of animals and the production environment which determines the level of inputs. Determinant parameters that affect meat production levels are the reproductive performances, growth pattern, and death losses. It is unfortunate that most of studies done on Ndama cattle and Djallonke sheep and goats have been conducted 66 under a controlled environment on station. When it comes to productivity evaluation in village herds, data are scarce and fragmentary. Nonetheless, there are data from village surrounding the Kolda research station which can be used to give proxies. Values obtained at the village level call for comments. The population from which these values have been derived is very limited. The data used to evaluate them were recorded over a period of a year. Therefore, fine statistical analysis such as the analysis of variance to investigate the effect of particular factors causing variation - season, year, age-, could not be undertaken. HOwever, these figures have indicative values and provide approximate productivity. Reproductive Traits and Mortality Table 8 gives reproductive performance and deaths rates of Ndama cattle on station and in village herds at Kolda. 67 Table 8 : Ndama Reproductive Performance and Viability at Kolda. units station village Age of first calving months 39.8(n=101) 51(n=47) Calving interval months 16.5(n=357) 17(n=43) Calving rate % 82 46(n=150) Viability % calves(O-lyear) 89 75 young (l-3years) — 90 adult (plus 3 years) 97 96 Cow longevity years - 12-15 sources : ILCA,1982; CRZ/KOLDA,1982; De Reviers,l979 Ndama females, in a traditional environment, calve for the first time at about 4 years and 3 months of age and the interval between calving is estimated to be 17 months. 46 percent of mature females give birth each year to calves 25 percent of which will die before the age of 1 year. 4 percent of the adult animals die each year. These mortality rates imply that 43 percent of females die before calving. The rearing proportion which is the proportion of births that produce a female that survive and is fertile (ILCA, 1982) is 29 percent ((1.-.43)*.5). This means that once in 3.4 (1/.29) calvings a cow produces a heifer that will reach lactation in the herd. Cows are maintained for more than 12 years. Assuming a cow longevity of 12 years and therefore a productive life 8 years (12-4) , a cow will calve 5.7 times (8/1.4) in its reproductive lifespan. Therefore 60 percent (3.4/5.7) of females born are required as replacements to 68 maintain herd size. Almost all of the female born would be required to ensure a certain herd growth. This explains the reluctance of farmers to destock female even when they get older. Growth Traits The mean weight of animals at different age are displayed in table 9. These show a slow growth rate of 2009 per day between 0 and 1 year. Animals attain adult weight at 4-5 years under traditional management. 69 Table 9 : Ndama Weight at Different age in Village Herds at Kolda. Age ( year ) n Weight(kg) Standard deviation Birth(M,F) 50 15.0 3 l M 37 88.0 26 F 37 89.0 31 2 M 41 143.0 32 F 36 137.0 33 3 M 28 208.0 33 F 28 173.0 33 4 M 11 216.0 14 F 34 200.0 18 5 M 8 265.0 33 F 34 207.0 25 6 M 24 265.0 38 F 223.0 30 M: male ; f: female Season affecting feed availability in quantity and quality, is the most significant factor affecting growth. Table 10 presents seasonal liveweight changes of four experimental groups of animals under different nutritional planes at Kolda. The control group (I) fed only on natural pastures ( similar to village conditions ) underwent dramatic weight loss during the dry season followed by an amazing weight gain superior to that of treatment groups receiving concentrate supplementation during the dry season. This phenomenon is known as compensatory gain. Increased dry matter intake has been suggested to explain compensatory 70 Table 10. Seasonal Liveweight Change of 4 Animal Groups under Different Nutritional Planes. Ia IIb IIIb IVb DSC RSd DS RS DS RS 08 RS Liveweight (kg)at be- ginning of season 235 204 232 247 232 275 232 242 Liveweight (kg) at end of season 204 268 247 294 275 305 242 280 Liveweight gain -kg -31 64 15 47 43 30 10 38 -% —13 31 6.5 l9 19 11 4.3 16 Daily gain g/day -277 609 134 448 384 286 89 362 a: control :grazing on natural pastures b: natural pastures plus different level of supplementation with groudnut hay,cereals,minerals c: DS: dry season d: RS: rainy season Sources: Gueye et al.,l979 gain (Sheely and Senior cited by Kearl, 1982), as has been the lower metabolizable energy requirement resulting from lower body weight at the initiation of the rainy season. The most plausible explanation of this high regain after a period of restriction is the very high feed efficiency after the period of deprivation. This the result of: - smaller body mass and consequently less ME requirements 71 - lower fat content of gain; fat reserves have been depleted, - increased efficiency in cellular nutrient absorption. Even though this phenomenon could be advantageous in fattening schemes, feed restriction have adverse effects on overall herd productivity. Heifers growth , for example, is retarded and consequently puberty onset and the reoccurence of pospartum estrus. The low productivitity per cow under village conditions as illustrated in table 11 can be attributed to high mortality rates of calves , retarded calf growth and poor reproductive performance. Under improved conditions cows can achieve yearly productivity of 70.1 kilograms of 9 months old calf compared to 37 observed in village herds. Table 11. Cow Productivity in Village Herds. Cow viability(%) 96 Calving rate(%) 46 Calf viability(%) 75 Calf weight at one year(kg) 89 Liveweight equivalent of milk offtake (kg) 26.4 Productivity: kg 1 year old calf per cow per year 37.4 Cow weight(kg) 210 Productivity : kg of 1 year old calf per 100 kg of cow 17.8 72 Small ruminants play a substantial role in the overall meat output of the system . SOMIVAC (1978) estimates that 13 percent of the T ; meat production in Casamance is generated from sheep and goats. Lack of information on small ruminant production has prevented the analysis of their productivity. 4.3.3 Energy Input to Agriculture Animal traction was introduced in Senegal in the early 19003 and the primary objective was to boost cash crop production (groundnut). The implementation of animal traction in the African farming system has been a controversial issue. Opponents of this technology argue that the labor surplus released by the use of animal mechanization would aggravate the critical problem of underemployment. Moreover, this technology is seen as a retardation of the development process. Advocates of animal farm mechanization base their inclination on various advantages that could be generated. According to Eicher(1982), the rational behind the utilization of animal powered mechanization stems from: 1. the potential increase in yield through seed bed preparation, deeper plowing , more timely weeding and planting, and moisture conservation, 2. the potential increase in acreage cultivated, 3. the income generation through off-farm transport 73 4. the reduction in drudgeries 5. the long term benefits of improving soil fertility through application of manure from animals, deeper plowing and plowing under crop residues Other factors militate in favor of adopting this technology. The mechanization of farming systems brings about the problem of energy sources. Most african countries rely on external sources for their energy needs supply. High energy prices, the reliance on external sources for spare parts, and poor maintenance are strong limiting factors of farm motorization. The utilization of animal energy in agriculture in Upper Casamance has been growing at a rapid pace. Such expansion may be attributed to many facts. Draft animals were readily available, farmers had a long tradition of keeping animals, land was not scarce, and most importantly the introduction of cotton production facilitated the acquisition of animal traction implements. Farmers growing cotton were eligible for credits. The development of that cash crop enabled them to generate cash income necessary for loan repayment. Two elements are determinant in the evaluation of animal contribution to the energy requirement of croppings; they are the draft power of animals used and their avalability. 74 Ndama draft potential and oxen availability Cattle are largely used as well as horses and donkeys as presented in table 12. Cattle are used for all agricultural operations. Due to their lighter weight, equines perform less demanding tasks such as seeding and cart pulling. Ndama cattle proved to be appropriate draft animals. (Munzinger, 1982; Starkey, 1981; CEEMAT, 1975). They are well shaped for work being stocky with short legs and strong muscles. It has been reported that, in proportion to their weight, their power is higher than other breeds. (Traverse, 1972, cited by Starkey, 1981). They are capable of an average traction equivalent to about 14 percent their body weight as compared to 10-12 percent for other breeds (CEEMAT/FAO, 1982, cited by Starkey,l981) Many factors influence the traction power of animals as illustrated in figure 8. The driving variables are weight and management including animal management and work circumstances, e.g., speed, number of daily working hours ect. It has been determined in Senegal that a minimum of 80 to 100 kilogram of draft power was required for performing various tillage Operations: plowing, seeding, weeding and harvesting. (Nourrissat, 1965) 75 LNINNOHMNI WVHINV . mUUCUUCMGmULmPP: cowvmucommtama ,uszfi r3 n13:- >¢3 I 99.32... 3.28 _?3233r Effie. .Ifiiiiif has—.352: 2:28:59 Nisan .san --.-----t g uwtggu A. u r-- ml o_vmem:um ass—(E P< Eux§z(x owwam ' ‘. I—-Af—--°’ O I Cg.— var—Ugh . P “Lmloa mUzQQDOZu ‘ a . . . . . o |-\-- \ D ‘ t_mcu Ucm mtouumb .meucm3_wc_ COwPUmLP Emu—W3 .m mt:m_n. J >00 Jotamat unwrzafln- , \ . Emu a man 76 In Upper Casamance context, applying an average traction equivalent of 14 percent of oxen body weight, a well shaped oxen team weighing 600 kilograms could sustain a work draft of 84 kilograms. There may then be energy constraints limiting the performance of high energy demanding operations such as deep plowing in heavy soils. This suspicion of energy deficit is reinforced by many facts. Draft animals are not supplemented and feed resources available in the dry season cannot assure animal maintenance. Consequently, oxen are very weak at the onset of the cropping season at peak energy demands and cannot perform as effectively. Young males are trained very early at two years old. At this age inadequate body development limits efficient performance. The performance of some agricultural tasks such as deep plowing, plowing under crop residues, seed bed preparation just before the onset of the rainy season and the use of heavier implements, are impeded by these factors. It is therefore clear that attempts to improve the efficiency of animal traction , should concentrate on the improvement of the animal itself. The design of adequat implements is also necessary. 77 Table 12. Number of draft animals in Upper Casamance year Kolda Velingara cattle horse donkey cattle horse donkey 1979-1980 5196 473 852 8695 643 1227 1980-1981 5880 832 2361 8181 1219 1763 1982—1983 7182 1275 3603 7465 1511 2188 1883—1884 8284 2049 4282 8832 2026 3241 Sources : SODEFITEX, 1980; Diao, 1985 The availability of draft animals is another important variable worth considering since it determines the success of animal farm mechanization. Table 12 indicates the number of draft animals in Upper Casamance. Table 13. gives an estimation of the potential number of three year old males that could be generated each year from Kolda herds. Each year 8543 males will reach three years of age and could be used either for draft or direct meat purposes. Such a figure is different from SODEFITEX'S (1980) estimate of 14434 of potential draft animals that could be drawn each year from herds. This latter figure appears unrealistic since a calving rate of 80 percent has been used to derive it which is too high a reproductive parameter for traditional herds. Such an estimate error may have significant negative effects for that institution to achieve the planned objectives. 78 Table 13. Potential Availability of Draft Animals from Kolda Cattle Herd. Medina Dabo Dioulacolon Number of breeding females 28141 28976 15210 Number of males reaching three years 4369 4498 2361 Number of breeding males required 938 965 507 Number of young males available for draft or direct slaughter purposes 3156 3533 1854 cow viability: 94%, calving rate: 46%, calf viability: 0-1: 75% , l-3:90%, 30 cows/breeding male. Actual Use of Animal Traction A look at table 14 indicates that almost all crops utilize to a certain extent animal energy with the exception of rice as previously discussed. Land preparation is the only widely mechanized operation in Upper Casamance. The percentages of mechanically plowed lands are 69 percent for sorghum and millet, 91.4 percent for maize, 76 percent for cotton, and only 12 percent for rice. The remainder is either prepared manually or not plowed at all. 79 Seeding tasks are almost entirely mechanized while weeding operations remains essentially manual. Existing farm equipments are composed of ploughs, Sine Houe weeders, seeders and carts. Because of poor maintenance, most of farm implements are not operational. SODEFITEX (1980) inventory census revealed that 1333 out of 3555 or 37 percent of available ploughs were out of order. The distribution of farm implements among farmers is very skewed. BOAD(l984) reports that in the Bonconto area 66 percent of households own one or more ploughs , 43 percent one or more Sine Houe, 25 percent one or more seeders, and 26 percent have carts. Out of 80 households, 36 or 45 percent possess one or more oxen team. 66 percent of the households use animal traction either with their own or leased animal. 80 Table 14. Percentage of Mechanized Agricultural Operations AT Kolda Agricultural Animal Manual None tasks traction Land preparation sorghum/millet 69 7 24 cotton 70.2 6.6 23.2 maize 90.4 7.8 .8 rice 12 73 15 Seeding cotton 12 88 Weeding cotton 4.5 95.5 maize 13.1 86.9 Compiled from SODEFITEX,1980 ; SOMIVAC,1985 The ideal animal traction farming system includes a complete set of equipment, crop rotation and tillage practices, animal health care and nutrition. (Sargent, Lichte, Malton, Bloom, 1981). The pattern of animal traction use in Upper Casamance is very unbalanced regarding mechanized operations and appears similar to the partial package approach advocating the replacement of implements one at a time with agronomic improvements completing animal traction. A major drawback of this paradigm is that technical benefits gained from timeliness in planting and area expansion can be offset by inappropriate weeding and therefore lower yields. In fact Kline et a1. cited by Sargent et al.(1981) give examples where partial adoption of animal traction equipment has shifted labor bottlenecks 81 from land preparation to weeding. The consequence is reduced yield potential. Interviews of farmers and technicians operating in the system confirmed that the number of weeding operations has been significantly reduced due to area expansion and the scarcity of labor at certain critical period of time. This is also evidenced by Achterstraat's observations: ” Sorghum and millet are manually weeded a single time, and sometime, benefit from a mechanic weeding. Generally, that's very unsufficient because of highly weedy fields. Old farmers are very conscious of this flaw and remember the time when cereals were manually weeded three times. ' (Achterstraat, 1983, p. 12) Animal Traction Constraints The main factors constraining the development of animal traction are the difficulty to mechanize certain agricultural operations, the deficiencies in oxen managements, and the inadequacy of implements. The partial destumping of lands, ridging practices, make impractical the mechanization of weeding operations. The present level of animal nutrition is inadequate to maintain well shaped draft animals which can perform heavy operations. The availability of credit is crucial for farmers to purchase implements. This constraint is more critical for cattleless farmers who have to expend an extra amount of resources to purchase or lease an oxen team. 82 3.2.4. Manure Production Yield increasing effects of cattle droppings is well recognized by agropastoralists . ' If you do not have animals to tether at night in your crop fields, you will not harvest enough cereals”, said Mamadou Diamanka of Sare Samboudiang village. The current practice for cattleless farmers is to lease herds to spend the night in their fields in order to improve soil fertility. The agromomic value of cattle droppings lies in their organic and nutrient contents (Jahnke, 1982). The organic matter improves the soil physical structure by making more effective the mellowing effects of plowing. Coulomb et a1 (1980) stress the essential role of the mineral content of manure in restoring soil fertility. The use of cattle droppings as manure has the additional advantage of transferring fertility from pasturelands to crop lands. However, the actual extensive management system of all categories of animals does not allow a full utilitization of the potential fertilizing effect of cattle droppings. High temperature and humidity lead to a rapid deterioration of the organic matter content of dung which in addition is futher deterirated by the invation of termites. Important quantities of manure are lost or deposited at irrelevant places. 83 Manure production from stall fed oxen using crop residues as litter would be an alternative as this allows an accumulation of manure which later could be used. 4.3 Demographic Model 4.3.1. Objective of the Model The purpose of the demographic model is to evaluate tradeoffs between meat output and energy input to agriculture from cattle according to different holding periods of draft animals. A sensitivity analysis on technical coefficients evaluate the impact of key parameters on the two outputs . 4.3.2. Model Description The demographic model, including an animal traction feature, is biologicaly oriented. It is dynamic in that it generates the time path of model variables. It is deterministic because no stochastic variable is included. The model is also non-optimizing in that no optimal slaughter age of oxen is specified, rather trade-offs between meat and energy outputs from cattle are evaluated. The model represents the dynamics of cattle herds given technical coefficients ( birth rates, death losses ) fed to the model as inputs. The herd is partitioned into a series of cohorts differentiated according to sex and age. Seven cohorts are identified : 1/ female calves, 2/ heifers, 3/ mature female, 84 4/ male calves, 5/ young males, 6/ bulls, and 7/ oxen. Appendix E presents the theory and mathematical procedures used to develop the model. Figure 9 shows a block diagram representing the structure of the model. The model assumes that animals are marketed at the end of their work or reproductive careers. Animal sales , at specified time periods, are then the integrated outputs from delay 3, 6, and 7 representing mature females, bulls and oxen respectively. The total number of marketed animal is computed as follow: TAS(t) = SF(t) + SMA(t) (1) where: TAS(t) = total number of animals sold SF(t) = number of female sold SMA(t) = number of males sold Equations 2 through 5 compute the number of female and male marketed. dSF(t) Mature female: = VOUT3(t) (2) dt dSMO(t) Oxen: —————— = VOUT7(t) (3) dt dSMB(t) Bulls: = VOUT6(t) (4) dt 85 Total males: SMA(t) = SMO(t) + SMB(t) (5) where: SMO(t) = number of oxen sold SMB(t) = number of bulls sold VOUT3(t) = rate at which mature females leave the delay process. VOUT7(t) = outflow rate from delay 7. VOUT6(t) = outflow rate from delay 6. The output flow from cohort 5 should be allocated to bulls and draft animals. The decision rule governing the allocation of ROUTS (output flow from delay 5) to delay 6 and delay 7 is illutrated in figure 10 . VIN6 and VIN7 are input rates to bull and oxen delays respectively. As a decision rule the herd breeding requirements, logically must first be met to assure adequate reproductive activities. Thereafter, the maximum number of animals leaving the young male cohort is directed into the oxen population to satisfy the energy requirements of agricultural operations. At each similation time, the actual bull population is compared to the required number of breeding males. If the 86 14030 E" : LNDLD IL] oxxorLt-B >4 0 d t: C Tnsu) ‘ “MHZ + Z + + $4) Vow-Mt) Kdl vour'ut3 (n P193 (— DELL Kn f PL?! , DELi g<)rma BRFfl) PoPRHu) ’ / DEM POW-((0 j E ‘5 .7 ‘ Wat) VINJ‘IU) J l MtochnoN Z t 0F ROOTS E TO DELMG Po nl1 ) mo DELAY? ’ SEE moo Rs ODOR 0:3 8R 4" ROUTslt) PLR; BRT(0 ‘ “ PonH (t) l. ll.) uhrrr'md 5R vounlt} \iIT "F— SR VOUVMS Fame (’0 \L547r' no BRHU) Figure 9. Block Diagram of the Demographic Model. (see foot notes p. 86 for variable directory) 87 Footnotes to Figure 2 : Variable Directory ADOR(t) : Additional number of oxen required at time t. BR : Calving rate. BRF(t) : Number of female calves born at time t BRM(t) : Number of male calves born at time t. BRT(t) : Total number of calves born. C - Hectarage annually cultivated by an oxen team. DELl, DEL2, DEL3, DEL4, DELS, DEL6, DEL7 : Length of the delays of herd cohorts female calves, heifers, cows, male calves, young males, bulls, and oxen reSpectively. K2 : Order of the distributed delay 2 KS : Order of the distributed delay 5 LND(t) : Hectarage cultivated at time t LNDO : Initial hectarage cultivated OXRQT(t) : Number of required draft aniamls at time t PLRl, PLRZ, PLR3, PLR4, PLRS, PLR6, PLR7 : Proportional loss rates corresponding to death losses for delays 1, 2, 3, 4, 5, 6, and 7 respectively. POPAM(t) : Total number of male animals POPBLS(t): Population of bulls at time t POPCW(t) : Population of cows at time t POPFC(t) : Population of female calves at time t POPMC(t) : Population of male calves at time t POPOX(t) : Population of oxen at time t POPYM(t) : Population of young males ROUT2(t) : Intermediate rate for delay 2 of heifers ROUT5(t) : Intermediate rate for delay 5 of young males SR : sex ratio SMA(t) : Number of male animal sold TSA(t) : Total number of animal marketed VIN6 : Input rate to delay 6 of bulls VIN7 : Input rate to delay 7 of oxen 88 B'8=°OFHLS-8noqr J YES no a~a > o z ”0 YES POPBLS H BMRQT ' 0190f Pepin YES no YES RI ) I ’ R2 > I 1 ix) v1ns=o. v1~s= -QOUTS'(l?Rll l~6-m Vll7:QOUlS VII). -ROUlS'Rl :f:§;:33;;3f,o ~n2) l_v:u7.m -o. 5 ‘\\\\\\\L ' //////// Irxt fmuunou\ \. 51:: 411‘ Figure 10. Declslon Haklng Process For the Allocation of Young Hal es From Delay 5 to Bulls and Oxen Delays. 89 latter is satisfied then the actual oxen population is checked. When there is not enough draft animals , the whole output flow from delay 5 is allocated to the oxen cohort. Otherwise the output flow is proportionatly allocated to delay 6 and delay 7. R1 is the ratio between the required number of oxen and the total male population. R2 is the ratio between actual and required number of bulls. These two ratios are used allocate the young males. The surplus of young male after breeding and draft animals requirements are met, are used for meat purposes. The number of oxen necessary is derived from the projection of cultivated area expansion and a coefficient C. The coefficient C is the number of hectares that can be cultivated by an oxen team during a cropping year. Nourrissat (1965) determined values of C for the different agro-climatic zones of Senegal reflecting differences in cropping patterns, soil and equipment used. C is estimated to be 6.29 for Upper Casamance. Since this coefficient has been derived using mature animals at their peak productivity, adjustment is deemed necessary. In so doing, the fact that it takes longer for young animals to attain their peak draft performance due to their lighter weight is taken into consideration. Moreover they are not yet adjusted to their new function. C has been set at a value of 6. which is in accordance with the average area of 5.9 hectares cultivated by an average family owning 1.1 oxen in 90 the Boconto zone of Upper Casamance. (BOAD, 1984). Equines are not included because of their incapabilities to perform heavy operations. The required number of oxen teams is therefore : OXRQT(t) = LND(t)/C (6) where: OXRQT(T)= required number of oxen teams. LND(t) = projected hectarage of cultivated lands The projection of cultivated areas is based on past expansion in the Upper Casamance and on the present trend. Ange(1984) reports that the cultivated areas have doubled in the last decade. With rapid population growth of 2.2 percent a year coupled with the availability of land this trend may continue. LND(t) LNDO * (1. + .10 *t) (7) Where: LNDO = initial hectarage of cultivated areas Equation 7 assumes a linear form of area expansion . Even in the situation of Upper Casamance with plenty of room for area expansion, a linear form of area expansion might not hold . Distance from homesteads, cost of clearing new lands and marginality of land may constrain such trend. However , this form will be assumed because of the rapid population growth prevailing in that zone. Furthermore, this 91 population growth prevailing in that zone . Furthermore, this assumption may be valid for a short period of simulation before area expansion reaches a plateau. Input flows to calves cohorts are : for females: BRF(t) = BRT(t) * SR (8) for males : BRM(t) = BRT(t) * SR (9) where: BRF(t)= rate at which female calves enter delay 1 BRM(t) = rate at which male calves enter delay 4 SR = sex ratio BRT(t) = POPCW(t) * BR (10) where : POPCW(t)= population of mature females BR = calving rate Storage in each delay yielding the cohort population is calculated as in equations 11 and 12 for the distributed and the discrete delays as follow: K D Q(t) = * Ri(t) (11) K . (=1. N Q(t) == DT* Ri(t) (12) where: ‘31 Q(t) = storage in the delay at time t 92 D = delay length K = order of the distributed delay Ri(t)= intermediary rate of the ith stage DT = incremental time in the simulation process Q(t) storage in the delay process N = number of stages in the discrete delay The discrete delay DELAYl and the distributed delay DELAYZ subroutines adapted from Manetsch(1982) have been used to model cohort dynamics. They have been extended to include proportional loss rates which correspond to death rates. Storage is also computed within the subroutines. The following differential equation is used to compute intermediate rates in the modified subroutines: where: dt D PLR dRi 1 ( Ri-l(t) -Ri(t)(l.+PLR*D) (13) ll “I Ri= intermediate rates DEL/K , DEL= lenght of the delay proportional loss rate the discrete delay , intermediate rates are computed as follow: Oi(t) = Oi+l(t-DT)*(l.-PLR*DT) (14) 93 4.2.3. Model Validation One important step in model development is its validation that is " the process of bringing to an acceptable level the user's confidence that any inference derived from simulation is correct. Validation is two faced: determining that the model behaves in the fashion as real life systems ; validating that the inferences drawn from the real life system with the model are valid or correct " (Shannon,l975,p,29 ). The conventional hierarchy of testing stages are: logical consistency tracking historical data expert opinion - prediction of the future In the absence of historical data the validation of this model will be restricted to the testing of internal consistency. In order to check wether or not the model behaves as the real world system, the trend ilustrated by the model outputs will be compared with the logical real world system trend brought about by changes in some parameters. 4.2.4 Internal Consistency There are two types of error associated with model building. The first is related to the representation of the real world by a mathematical model. The computer model approximating the mathematical model generates the second 94 type of error. Difference equations are used in the computer model to approximate differential equations of the mathematical model. Central to the stability of the model is the step size used for the numerical approximation of differential equations. DT is the time step increment choosen. The error in the computer model decreases as the step size decreases. DT must be small enough to make numerical errors in the computer model " negligible" but it should not be smaller than necessary because computer operating costs increase rapidly as step size decreases (Manetsch,l977). The following conditions should be met for stable models containing distributed delays: DELj 2 MIN > DT > 0 (15) Ki Stability considerations require also that: DT < DEL/(2.*(K+PLR*DEL) (16) Different values of DT have been used to run the model. Table 15 presents the model outputs i.e. total herd population at different simulation times. 95 Table 15. Total Herd Population at Selected Points in Time versus DT step size( Delay7 ==8) DT size t=l. % change t=2. % change .2 384.2 402.9 .1 390.9 1.7 409.7 1.68 .05 394.1 .82 408.5 .29 .02 396.0 .42 409.6 .26 .01 396.6 .15 410.0 .10 DT=.05 with an error of .42 percent is used . The additional test of internal consistency is to check weither the demographic model perform correctly. This is carried out by using death losses. Net POpulation change = Births - Deaths - Sales (17) Equation 17 (Jaske,l976) means that at any interval of time the net population change must be equal to deaths and sales substracted from the number of animal born. This equation is solved for the death variable. The simulated death is compared with the theoretical death represented by equation 18 t+DT Deaths = DT *' POPi* DRi (18) t where: DT = simulation time increment 96 POPi the population size in cohort i DRi death rates pertaining to cohort i fed to the model as input Solving equation 17 for death gives simulated deaths in equation 19 Death = POPMCllt+DT)+POPFCl(t+DT) - VOUT3(t+DT) - VOUT6(t+DT) - VOUT7(t+DT) - (TPOP(t)-TPOP(t+DT)) (19) where: POPMCl = storage in the first stage of delay 4 of male calves POPFCl storage in the first stage of delay 1 of female calves VOUT3, VOUT6, VOUT7 = outflows assumed as being animal sold from mature females, bulls, and oxen respectively. TPOP = total herd population Table 16 indicates simulated and theoretical deaths. Between t=.15 and t=.45. A simulation error of 4.8 percent on deaths is made which seems acceptable. Additional validation tests are performed regarding simulated versus actual values of variables such as offtake 97 rates and population growth rates. Results are presented in table 17 Such various tests conducted support an acceptable validity of the simulation model which can be operated under varying scenarios after the base run is performed. 4.2.3. Base Run The model was operated to simulate a situation at a village level. The village appears to be an appropriate level of model aggregation as far as some features related to land and cattle herd utilization are concerned. In fact, there is a rental system of draft animals that allow cattleless farmers to rent oxen in order to perform their agricultural operations. Thus, it can be assumed that the excess draft power from some households can be spread over the entire village. Moreover, it is a common practice that excess family land or fallow areas, be freely exploited by other households. This leads to the assumption that land availability would not be a limiting factor to cultivated areas expansion at village level. Table 16. Comparison of Simulated and theoretical Deaths over time T Simulated Theoretical Error(% error) deaths deaths .15 1.778 1.871 .093(4.9%) .20 1.782 1.874 .092(4.9%) .25 1.786 1.877 .09l(4.8%) .30 1.790 1.880 .090(4.7%) .35 1.794 1.883 .089(4.7%) .40 1.798 1.887 .089(4.7%) .45 1.803 1.891 .088(4.6%) Total 12.531 13.163 .632(4.8%) Table 17. Simulated versus Actual offtake and Population growths rates(de1ay7 = 6) Offtake rate Population growth rate (%) (8) Simulated 5.7 2.7 Actua1* 6.0 3.0 * source: DeRevires,l979 The data base used to operate the model and dealing with herd size and structure, and cultivated areas are 99 adapted from Achterstraat(1984) who studied the farming system at Lenguewal village in Upper Casamance. The model has been run with different values of oxen delay length expressing oxen age of sale. Model outputs are displayed in table 18. Figure 11 is a bar graph representing oxen populations, the average annual number of marketed animals, and the balance between actual and required number of oxen. A positive balance is equivalent to the number of males that could be used directely for meat purposes. It can be noticed that rapid turnovers of oxen lead to more male offtake, less oxen numbers and therefore higher animal traction energy deficits. The accumulated number of males marketed at the fourth year of simulation is 70 for a holding period of 3 years of oxen. This number drops to 30 for a holding period of 8 years. The oxen deficit amounts to 33 and 1 for a 3-year and a 8- year holding periods respectively. Tradeoffs exist then between the herd meat and energy outputs regarding oxen age of sale. Considering the prevailing technical coefficients, it is required that oxen be kept for at least 7 years so that the animal energy requirements of the cropping sector can be met during the fourth year of simulation. 100 Table 18. Model Outputs: Base RUH( Oxen Population. Number of Males Sold. Oxen Balance) Slmlllatlon tlme (year’s) l 2 3 I 5 6 I 8 9 l0 fic>lcilvlcn Perlod (years) Oxen 3 49 39 SI 60 63 66 65 66 65 66 Pooulallon I 54 58 66 75 I9 85 85 85 83 86 5 56 62 I3 84 95 l0I I04 I06 l85 I86 6 58 66 ll 89 I02 ll5 l|9 I24 I?! IIS 3 59 68 8| 9‘ I07 I2l I33 I38 Ill l43 8 60 30 83 97 Ill I26 I38 ll9 l55 I60 Hflcswd 3 I9 36 SI 10 88 |08 l3l l5l llS I97 I I4 28 (I 5‘ 78 86 l07 I28 I38 Ill 5 I2 23 34 ll 55 69 86 l85 l25 ll6 6 IO 20 29 38 ll 5‘ 69 85 IO! I23 3 9 ll 26 33 ll 43 55 69 84 l8? 8 8 I6 23 30 33 ‘2 I9 55 69 84 Oxen 8alance 3 ~22 ~28 ~30 ~30 ~33 ~37 ~45 ~58 ~53 ~63 I ~ll ~28 ~l8 ~l5 ~ll ~I8 ~24 ~38 ~35 ~43 5 ~15 ~15 ~ll ~6 ~2 ~3 ~S ~l0 ~l8 ~23 6 43 42 ~6 ~I 5 I! I8 8 2 ~‘ 3 ~I2 ~9 ~3 3 I8 I8 23 22 28 I5 6 -u -7 ~l- 1 u z: 29 u 33 3| (-) Oxen DeFIclt‘ (+ Hale Surplus After Oxen reoulrements are met. 101 33.5953 .xo W33 22. 3...: WV... .5 U. A9393 1.25.... 9:32: .zaqutnN :lnurg'xv )0 m. s... 0 n... v m p . p _ p p m L _ I. .II I plfi CHI . ... W. cm. . . r 3.. 9 4 I I! 1 mi \& 4 e.\ a w- O \ u 7% pm.“ a.» 3 \1 \ E; .. .P\_ \ml on .\ _ .. _ _. s _ m _ is .. C x I an \ .\. \ .._ ..\ 1. Q. . x. _ w t C . .\L .\L ~._ _ w\_ —\l Ow. . . C. 1.. C. C C . T v... C .L. T... 3 v ”\t. ..‘\J.. V; t; 3 3m _ _ h\ 3 .. Us. . a t a. l _. t. _ .1. "x L t _ _ a; R Q C T as a w e _ e 12.339.6ch .xo .22. 01.2.3; 35m ommm .H H oedema 102 4.2.4. Sensitivity Analysis Sensitivity analyses have been performed on technical coefficients , e.g., death rates, calving rates, and on the young male delay length. Such experiments were conducted to figure out how these parameters, singly or in group, influence the holding period of oxen. Improvement of these parameters indicate an amelioration of the system productivity brought about by technological changes capable of improving health and nutrition. In a first scenario, a 20 percent improvement of calving rates is set. This seems realistic since the Livestock project of Eastern Senegal has achieved 23 percent increase in calving rates after six years of intervention. ( Landais, 1984). The “Project Rural de Sedhiou' aims at 20 percent increase in calving rates in the Middle Casamance zone(De Reviers,l979). The second scenario deals with a 40 percent reduction in mortality rates. A third simulation run combines the two previous scenarios. A fourth run is performed which combines conditions of the third run as well as a reduction of the delay length of the young males. Such a reduction would reflect the effects of a better nutritional plane for young animals resulting in the reduction of the time necessary for young animals to reach sizes at which they can be trained for draft purposes. 103 for young animals to reach sizes at which they can be trained for draft purposes. The model outputs from these different runs are displayed in appendices D1, D2, and D3. Figure 12 represents graghs of oxen population , the average number of males sold each year and the balance between actual and required oxen poulations. Runs two and three yield the same minimum time period of 5 years for oxen utilization on farm. When these two improvements are combined a minimum holding period of 4 years is obtained. The additional shortening of the young male delay length reduces the period to 3-4 years. As indicated in table 19 even though effective in providing enough draft power to the crop sector and capable of increasing the meat output, these strategies are conducive to huge cattle population increase given the assumed offtake pattern. This may not be desirable as the availability of feed resources may not allow or support such a population explosion unless there are concomitant measures capable of promoting increase in animal offtakes. The problem of population increase through increased health care deserves a great deal of attention. Farmers behavior regarding animal offtake under such conditions of herd increase is difficult to predict. Are they going to market more animals? Pressure on feed resources and reduced productivity are the ultimate outcomes in the case Amflm>amcm >uw>amcwmv AmsHmu:m\uHOflmmp.xo . Upaom mamz .GOHumasom .xo V musmuso Homo: .NH musmflm 104 ‘3‘. *\ o. o . a I C o a .. .cmhhh- “Mme 9:25. 32 o. t a .6 EU 3.83:5...th. Mums: 9:22. 33 .23.. E :6 RN 5. o m c n c s o n o n p . DH M f 2. we . e e. t. \ n u x a-.. u & A r - 2. M n \ w - s \ On M Q \ \ w . _ t \ 3.. x \. _\ _ 3 w. v \ \ T v u \ w. \ LK .\ v“ 0. W. . x m x . .. e u \‘ \ N X \ X .t ..c e a s l e . \ re A LL .6. R $ t I; a .- (l afl— W a j m \mv . is C 3 W. \U .53; \u.u.‘o ..O . 3 o n . .. ) ... a.._e.7n\ .2er .8 30¢ no...) no ..fi A'slvhw mus... “2.20: ‘_ O. o n R81 . .ILICM.$IL .¢.v_o= a B a e. e e . a a r m n a A l h I l \ fl 0“. .r II L l 3“. _ D. On- . - 3 O—l H4 .61 1“ fl \L / I_d .J m L. . /.\ .1 1 E w. .m/w . r; #1 :3 e x i v x K t :1 % Fur: H» . X. T t. L .1 \r V v m , 4+ \ OM m . . x V. \ X 3 . W ¥ \. \ . . 8 .r x. \e \ x v. w. w). . C. V. x 41 On w... .M \. r v. V4 H c. w . . N. . x T m x x x x x . .1 L K. I. X s .4 3 .. c t \ \ L S w . 1 x t .2 S- e... w x ‘\ a\ \ \v 30 we \ A. VA ML . Wu H \. .\ ..\ \ I \ YA .L L x .1. $9 w x .x \u \H \ L- :n I “O \ 1. \ H or. cl C 4.. r f A a. f. I Q A ,m o s t .. t s a r x . t x t L 3 e. m e . 2 \r m A 93. a D: \L . 3"_ .n-._$ ...n\¢.v.~tc .-0 4:8 08-OSIOV m. .— n :3: \n N ::z \c 105 that the improved health care is not concomitant with increased offtake rates. This question needs to be raised since a livestock project financed by the World Bank and Table 19. Cattle Herd Size under Different Scenarios at t=10 Oxen holding period(years) 3 4 5 6 T a Total POpulation Base run 439 459 479 498 516 533 Run 2 579 606 631 654 675 693 Run 3 622 649 677 703 728 748 Run 4 829 862 897 928 957 980 Run 5 811 849 885 918 949 977 implemented by "SODEFITEX", put emphasis on combating disease hazards facing the livestock sector in Upper Casamance (Banque Mondiale, 1983). An attractive alternative to avoiding population explosion resulting from improved health care would be the use of females as draft animals and increasing male offtakes. Such an alternative has the advantage to release 106 a great number of males that could be directly oriented for slaughter. Increased offtakes and thus the alleviation of the pressure on feed resources may be expected from such a strategy. However this option may not be viable under prevailing production conditions in Upper Casamance where females are stressed by excessive milking and deficient nutrition. When an improved nutritional level can be assured, then, the use of females as energy supplier can be combined to their conventional functions. Important biological and economic variables influencing the age of sale of oxen were not included in the model because of the fragmentary nature or absence of information related to them. Growth curve marked by seasonality is a determinant influential factor of energy and meat outputs. The present state of knowledge about the animal growth under traditional environments does not allow the incorporation of animal weights in a quantitative framework such as a simulation model. Landais (1983) ,after fitting liveweight growth data from village herds in Ivory Cost into classic growth curves models ( Compertz, Von Bertalanffy models), found discrepencies between actual and theoretical growth curves. He concluded that the mathematical modelling of growth curve under conditions of high health hazards and poor nutrition, has little value for breeding purposes. Economic considerations are crucial in analysing 107 Economic considerations are crucial in analysing the age of sale of draft animals. Important variables are : farm gate prices of live animals, interest rates, value of services, and cost of maintenance (feed, labor) (Panayotou, 1982). Benefits such as increased value and/or weight, services, and costs including labor feed, forgone interest, risk of loss, are associated with keeping an animal an extra year. Microeconomic, optimizing models pertaining to the age of sale of multiple purpose animals have been developed (Jarvis,l982; Panayotou et a1. ,1982: Ariza- Nino,l984). Following Panayotou et a1.(1982) the equality of benefits and costs of delaying sale ( marginal benefits and marginal costs) determines the optimum age at which the animals are sold. The mathematical formulation of the optimal condition is : V'(t) +S(t) = r.V(t) + a(t) +b.V(t) where: V(t) = p(t).Q(t) = market value of the animal at time t p(t) = price Q = quantity V'(t) = dV(t)/dt is the change in value of the animal due either to a change in weight , dQ(t)/dt , or a change in its price, dP(t)/dt S(t)= value of services provided by the animal during a year ( work, manure ) 108 r = rate of interest of return from alternative investment available to the farmer a(t)= cost of maintenance: labor, feed b = probability of losing the animal to theft or disease In Upper Casamance, the input stream to draft animals is minimal. Labor is the most important one since a zero opportunity cost can be attached to feed from pastures and cereal straws. On the other hand there is a sharp increase in oxen prices. A 25 percent annual increase of oxen prices is reported by Achterstraat(1984). Individual farmers circumstances have a lot to do with the slaughter age of oxen. Low income, cattleless farmers may have difficulty in acquiring animals or keeping them for long period of time . They may be compelled to sell their animal in order to purchase food. The sale of animals for such purposes occurs frequently in the rainy season when farmers experience food shortages. Herd size and its capacity to generate male animals is also important for farmers decision to sell oxen. The possibility of oxen rental is another variable that can influence farmers' decision. Retarded growth due to a poor nutrition environment, high death losses and poor reproductive performances explain the low animal energy supply to the crop sector, the delayed slaughter age of animals and resultant low rate: of meat output from the livestock system. Attempts to improve these 109 two outputs solely through health programs may be counterproductive because of potential population explosions which may result. Priority should be given to enhancing the feeding system which correlated with basic health care and an improvement in the marketing system should increase animal offtakes. - 4.4. Summary of constraints Seasonality in food quantity and quality is the single most important constraint facing the livestock sector. Consequent dramatic weight losses are not compatible with high herd productivity. Remedies to the the dry season nutritional stress must be investigated. Although more gifted than the northern Senegalese zone regarding water resources, insufficient watering facilities, in Upper Casamance, is a serious obstacle to improved livestock productivity as this limits grazing areas in the dry season. Internal parasites (tryponosomiasis) external parasites such as ticks and the diseases they carry, infectious diseases ( anthrax, brucellosis ) are major health problems. Animal genotype is not a principal constraint unless environmental stesses are removed or alleviated. (McDowell, 1977: Bersten,1983) Even though the overall system management efficiency cannot be accurately assessed at present , one can notice that practices such as excessive milk offtake and reduced 110 grazing time may have adverse effects on overall herd productivity. Farmers' lack of bargaining power, and consequently their vulnerability vis-a-vis middlemen prevent them from profitably market their commodities. Rainy season labor bottlenecks would be a serious impediment to the introduction of new technologies during that period. A negative labor balance is reported by BOAD(l984). An average household in the Bonconto zone has a theoretical requirement of 185.6 man-working-days to perform plowing, seeding and first weeding operations , out of which 160 are available. At a macro level, goverment policies play a key role on system's performance. The pricing structure, as it stands today, works against improved system productivity. Controlled meat prices set up to protect urban areas meat consumers is conducive to low producer prices. Consequently, this lessens farmers' incentive to engage in more productive production systems. Moreover, the absence of credit earmarked for improved livestock sector reflects the absence ofg commitment on the part of decision makers to the livestock sector. In addition, the design and application of new technololies is limited by insufficient investment in research and extension and an inadequat research orientation. An attempt to find solutions capable of removing or alleviating such constraints is the purpose of the next chapter. CHAPTER V SYSTEM'S ALTERNATIVE SOLUTIONS As discussed in the previous chapters numerous constraints plague animal productions in Upper Casamance. The generation of alternative solutions capable of overcoming such constraints, is the purpose of this chapter. A wide range of pontential alternative solutions are proposed. They are then screened for the most feasible at the village level. Each solution will be subjected to a first test of physical and social realizability and a second for financial and economic feasibility. The information necessary to carry out cost/benefit analysis of alternative solutions is not available, rather simple calculations and economic rules of th thumb will guide choices. This illustrates the information gap on crucial socioeconomic variables. Some of the solutions proposed in this chapter are an attempt to tackle the dry season nutritional stress while others are directed toward improving management practices. Certain solutions will require that some aspects of the system structure be altered. Others are intended to enhance the input supply to the system and on ways of enhancing farmers' bargaining power. Four pontential classes of alternatives come out of the analysis of contraints confronting the livestock system. They are: 111 112 l. improvement of the utilization of crop residues and agro-industrial by-products. 2. increased fodder production 3. improvement of small ruminants productions 4. innovations in management practices 5 .1 Improvement of Crop Residues and Agro-industrial By- products Utilization. Numerous variables need to be considered in attempts to use or improve the use of fibrous agricultural residues (Parra et a1,l984; Sundstol,1984; Naga et a1,1984). Variables to be considered are: amounts, quality and distribution of the materials, seasonality of occurrence, alternative use of the materials, alternative feeds and feed supplementation, level of technology in the agricultural sector, pontential improvement due to treatment, toxic/anti-nutrient factors, and use cost. Various feed resources from crops are available in the Kolda zone. Cereal straws from millet, sorghum, maize, rice, crop legume hay from groundnuts, and cottonseeds are most abundant. As illustrated in Table 20 appreciable amounts of cereals straws and groundnut hay are derived. The cotton processing plant at Kolda has produced 6725, 4835, and 5048 tons of cottonseeds respectively in 1982, 1983, and 1984 out Of which 1427, 0, and 853 113 Table 20. Estimation of Crop Residues Availability. Maize Sorghum Millet Rice Groundnut Grain Yield kg/ha(l) 880 706 713 751 1215 Conversion factor (2) 2 6.5 6.5 2 2 Residues yield kg/ha 1760 4589 4635 1502 2430 Hectarage (3) Dabo I Dioula. l Medina Residues , Avai6ebility _ I8040 25698 21552 3830 30375 Dioula. 19360 25240 25492 5257 23693 Medina 17424 23109 24565 2629 32198 Sources: 1: Diao,!985: 2: Mongondin, 1979: 3: SODEFITX. I980 114 tons have been used as seeds (SODIFITEX, direction industrielle,l985). There ares no major transportation constraints associated with the use of roughages from crops. The same management units producing crops also rear the animals using the residues. Moreover, there is no major alternative use of these residues. Only small portions of the cereal straws are used as construction materials. From a nutritional point of view, the feeding value of fibrous agricultural residues is limited by their deficiencies in crude protein, minerals and vitamins. In addition, they are high in fiber, have high lignin and/or silicon contents, low digestibilities and low voluntary comsumption by animals (Parra et a1,1984). There are two approaches commonly used to enhance the use of cereals crop residues: treatment and/or supplementation of deficient nutrients. 5.1.1. Treatments Many treatment procedures have been devised to improve the digestibility and voluntary intake of roughages. They include physical, chemical and biological treatments or their combination. 115 Physical Treatments Physical treatments include grinding, high pressure - high temperature steam, pelleting, drying, soaking, etc (Kiflewahid,1983). These methods result in a reduction of particle size and an increase in surface area and density (Parra et al.,1984). Grinding with subsequent pelleting markedly increases voluntary intake, reduces digestibility and have no effect on lignification, but increases rate of passage (Donefer,l977; Jackson, 1978). Because of high operational energy in processing these procedures are not suitable in the Upper Casamance context. A possible physical treatment would be chopping and/or soaking of cereal straws prior to feeding. It has demonstrated to yield positive reSponses in terms of increased intake and digestibility of rice straws in Southeast Asia (Khajaren et al. ,1983). The improvement of rice straw by any means other than through the grazing system at Kolda is constrained by some socio-economic factors. Paddy is harvested in such a way that only spikes are cut and stems and leaves are left standing in the fields. Improving the use of this roughage through any treatment would require alteration of the harvesting technology. For example stems would be out near the surface of the soil and stacked therafter. It is not apparent if the additional labor requirements associated with this technological change will be available. Moreover, 116 because of the sexual division of labor, rice production is a reserved domain of females and they are not responsible for cattle feeding. Thus as of the present moment, any improvement of rice straw utilization other than one based on the grazing system can be ruled out in Upper Casamance. The utilization of maize stover is made less practical by the fact that its harvest occurs in September when agricultural tasks are at their peak and the rainy season is still on. This makes conservation difficult. As a result of the above contraints limiting the use of rice and maize straws, attention should be focused on millet and sorghum straws if the feeding system is to be improved. Chemical Treatments Chemical treatments consist of alkaline hydrolysis of roughages using chemicals such as sodium hydroxide (NaOH), calcium hydroxide (Ca(OH) ), ammonia (NH ). Tarkov and Feist cited by Parra et a1. 2(1983) describe3 the mechanism of actions of the alkali. They contend that alkali treatments lead to the solubilization of cellulose and hemicellulose and this is accompanied by a break down of the ester bonds which link lignin to the structural carbohydrates, as well as by an increase in the saturation points of the fibre. The concensus is that among the various chemicals used, sodium hydroxide and ammonia have proven to be the most 117 successful (Kategile, 1983; Sundstol, 1984). There are many factors which make the applicability of these methods at small farmers' levels at Kolda unfeasible. The chemicals are expensive. They also pose human and animal health hazards and enviromental pollution' problems which make them difficult to apply. The most promising method that needs more investigation is the ammoniation of crop residues with urea or biuret. Potential benefits can be gained by using such ingredients. They incorporate non-protein nitrogen in the diet. These chemicals are extensively used in South Africa(Topps, 1972) and in Australia (Alexander,1978). Toppsll972) found substantial reduction in weight loss of cattle existing on low protein forage and eating regular and suitable amounts of urea or biuret. The immediate practical aim of feeding such compounds was to provide or to approach closely a maintenance level of nutrition. Biuret appears to be a more appropriate chemical compound for treatment. It is hydrolyzed more slowly in ammonia and carbone dioxide in the rumen than is urea, therby obviating toxic accumulation of ammonia in the blood plasma. Above all it does not exhibit any undesirable palatability or toxicity characteritic. (Fonnesbeck, 1975). According to Kowalczyck(l976), supplementing diets of relatively low digestible orgarnic nitrogen is useless as the bacteria are unable to utilize the ammonia released in the 118 rumen and it is advisable to do so with diets containing large amounts of readily digestible carbohydrate but small amout of protein. However, as pointed out by Jackson (1978) and Topps (1972) concentrate supplementation depresses the digestibility of of the cellulose in the diet and the benefit of additional dietary energy on the retention of body nitrogen and protein is relatively small at low level of nutrition. Biuret and urea can, therefore , be of practical use in situations where dramatic weight losses resulting from low feed quality are observed. These two chemical compounds deserve particular attention in Upper Casamance as a pathway to partially solve the dry season weight losses. 5.1.2 Supplementation with High Quality Forage It is a concensus that an efficient way to improve energy intake and animal performance on crop residues, is by suplementing these with high quality forage. ( Preston,1982; Khajaren,l984; Parra et al., 1984). Forage legumes which have proven to efficiently supplement fibrous crop residues are Leuceana leucocephala, Gliricidia eSpium and diverse varieties of Stylosanthes. In the Upper Casamance context the most viable alternative would be at present improving the utilization of groundnut hay. It is a good quality forage but harvest and storage methods are conducive to high losses in quality as illustrated in table 21. 119 Table 21 Nutrive Value of Groundnut Hay According to harvest and conservation methods. Harvest and Percent ME "Disgestible Conservation DM Mcal/kg DM Nitrogen Methods Matter:MAD" g/kg DM Harvested before Thershing 93 2.52 102 Stored after Threshing 92 1.89 40 Left in fields after threshing 92 1.10 40 Souce : SOMIVAC,1978 , MAD : matieres azotees digestible Substantial benefits could be gained if groundnuts were lifted after harvesting the forage. The feasibility of this technological change needs investigation. A proper storage is a prerequisite for an adequate conservation of the groundnut hay feeding value. With the current harvesting method this feed resource should not be left in the field after theshing because of the deteriorative effect of the sun on hay quality. Storage of groundnut hay in huts or under shadow of trees and its protection by pasture hays and thorny branches is feasible. Animal tracted carts can be used as means of transportation. Legume forage production is another alternative that 120 needs to be considered. It will be examined later. 5.2 Utilization of Cottonseeds Cottonseeds which are available in the same zone of livestock productions, constitute an immediate and appriopriate feedstuffs to supplement poor quality forages and crop residues. Two aspects of cottonseeds utilization are of prime importance. They are the animal response to this feed and the opportunity cost associated with their use as animal feed. Adegbola (1982) cites experiments, in Nigeria, where pasture forages were combined with different levels of cottonseeds supplementation. Gains were lower with higher levels of cottonseeds and these animals showed low dry matter and crude protein digestibility. The lower performance of animals on high cottonseeds level was attributed to high levels of oil in the rations. At present cottonseeds are officially marketed with a price of 18 FCFA per kilogram to livestock producers. This is a policy specially aimed at supporting starving animals during the dry season. Mongondin(1979) estimated the forgone profits resulting from diverting cottonseeds from plants producing vegetable oil which is exported. This estimatation yields an opportunity cost of 37 to 43 FCFA. Vegetable oil and oilcakes are major export commodities of Senegal which derive most of its export earnings from these products. It is then more 121 appropriate to use the opportunity cost of this resource rather than the official price to access the economic soundness of using cottonseeds as animal feed. Assuming a situation where cattle meet their energy maintanance requirements from natural pastures and are supplemented with cottonseeds, they will require 3.5 "UF“( 1 UP = 3.149 Mcal ME ; Munzinger,l983) per kilogram of liveweight gain (Riviere, 1977). Such energy demand is equivalent to 3.5 kilograms of cottonseeds ( 1 kg of cottonseeds provides 1 UF ; Calvet, 1978) The total cost of cottonseed supplementation would be equal about 140 FCFA per kilogram of liveweight gain. The producer price of liveweight animal varies between 120 to 140 FCFA per kilogram in Upper Casamance. It therefore appears as though the supplementation cost would be prohibitive relative to producer prices. Even with the use of official prices, the feed cost is high ( 63 FCFA) , and more so other production costs have not been included yet. Such calculations though simplistic indicate that it may be neither financially profitable nor economically sound to use cottonseeds in a beef enterprise. However, one can assume that, in situations of restricted feed alternatives and considering potential benefits that could accrue, e.g.. increased crop production through better oxen nutrition, improvement in viablity, calving rates and milk production, supplementing animal with cottonseeds may be worthwhile. Careful choices are to be 122 made as to what level of supplementation is efficient an what category of animal should benefit from it. Draft animals should be given priority. This would enhance the effective integration of the two sectors. The potential extra crop and meat productions might justify the use of cottonseeds as animal feed. Just how effective this option is requires more investigations capable of establishing the benefit/cost ratios. 5.3 Fodder Production from Legumes Fodder production has been suggested as an alternative of intensifying livestock production and integrating crop and livestock productions in the humid zones of Africa (ILCA,1979 (b); Sumberg,l983; Sumberg,l984). Herbacious legumes of Stylosanthes genera and tree legumes such as Leucena leucocephala and Gliricidia espium are proposed as suitable legumes. Kang et a1. (1981) described an alternative production system known as alley cropping to replace the bush fallow system in the humid zones of Africa. This system has been advocated as an attempt to remedy problems facing the farming system in humid tropics where population growth and restricted land lead to the shortening of fallow periods. The reduction of fallow periods limits soil fertility regeneration. Kang et al.(1981) define alley cropping as"a cropping system in which arable crops are grown in the eSpaces between 123 rows of planted woody shrub or tree fallows in which the fallow species are periodically prunned during the cropping season to prevent shading and provide green manure for the companion crop." The use of legume species have additional advantages. Besides soil fertility regeneration, they also provide for stacking materials, firewood, animal and sometime human food. (Kang,1981; Sumberg,l983) The same authors report appreciable dry matter and nitrogen yields: 4 to 8 tons of mulch DM/ha/year yielding 100 kilogram of N for crop production. Many factors limit this alternative because of the radical changes the system design parameters of the agropastoral production system in Upper Casamance which would be nedded. The fisrt is the unavailability of seeds or cuttings for the establishement of legumes. Secondly and most importantly, such innovation will be setback by the necessary restriction of animal movements. This system may be more practical for sheep which are easier to handle. Many problems emerge when one deals with a livestock system where cattle constitute a predominant part of the livestock biomass. Livestock are reared on communal grazinglands. Even though fallows are family owned and their access monitored, the requirement of fences either live or mechanical, may rule out this innovation. Sumberg proposed leucena-fences for sheep. This may not work with cattle. The most viable alternative in terms of forage 124 production is the intercropping of cereals with cowpea(Vigna sinensis). In addition to its contribution to the supply of protein to farmers already faced with food shortages, the forage produced will be valuable as a source of protein for dry season supplementation of pasture forage and crop residues when properly conserved. Futhermore , this alternative does not require all the inputs mentioned for forage legumes establishement. 5.4 Improve Small Ruminants Production Small ruminants have been neglected for a long time and this situation is still prevailing. They significantly contribute to farmers' welfare and to the overall system meat output. They are highly adaptable and resistant to adverse environmental conditions. Their resistance and adaptability is evidenced by the series of droughts which caused more damage to cattle than to small stocks. DeSpite their recognized potential, no development scheme directed toward this sector has been designed yet in Upper Casamance. For example no health care is provided nor is there any purposely planned improved nutrition or management strategy. They have specific advantages: - wide dietary range - small size and therefore lower energy requirements per animal 125 - low initial capital requirement — easiness in handling and marketing The above advantageous characteristics offer more opportunities for rapid turnover of capital from small ruminants. Their meat ouput can be substantial relative to the overall livestock biomass as evidenced by Wilson (1984,p. 98) "the faster reproductive rate of small ruminants, their more rapid growth and early age of offtake enable them to produce meat on an annual basis at approximatly twice the level of their contribution to the herbivore standing biomass". The productivity of the small ruminant sub-sector in the humid zones of West Africa is plagued by diseases. The most prevalent is " Peste des Petits Ruminants ; PPR" which is is a Rinderpest related viral disease causing high mortalities (Sumberg and Cassidy, 1984; ILCA, 1979(b)). Other important disease are Mycoplasma and Pasteurella infections and parasitic infestations. Diseases control and a better management would lead to improved small ruminant productivity. PPR can be effectively and economiacally controlled by annual vaccinations with tissue culture rinderpest vaccine or TCRV ( Mack et a1, 1984) Upton(l984) proposed two possibilities for the intensification of sheep and goat productions in the humid tropics of Africa. He suggested cultivated pasture production for grazing by sheep or permanent housing with cut-and-carry feeding or zero grazing. Problems identified 126 earlier and associated with pasture establishement limit such models of intensification in Upper Casamance. A small scale family fattening scheme such as the one practised in the Senegalese groundnut basin is an alternative for the intensification of small ruminants production, eSpecially sheep because of their popularity. Young males are isolated from the flock and stall fed with groundnut hay and home made oilcakes. Housing facilities are made using local materials. High prices attached to well finished animals during muslem religious celebrations, make profitable such operations. The scarcity of data on all production aSpects marking this sector does not allow the design of a comprehensive plan for the development of small ruminant production in Upper Casamance. However, the ongoing monitoring of village flocks will provide information which will permit to define a research orientation and to set a development strategy. Information on the disease profile, management practices and socioeconomic variables are urgently needed. 5.5 Alternative Management Strategies If promising technological innovations are relatively easy to conceive, their practical application into the prevailing management system appears more challenging. Alteration of management strategies though essential for improved system productivity, requires careful assessement of their acceptability and potential. Various management 127 improved system productivity, requires careful assessement of their acceptability and potential. Various management choices can be devised, the most important of them will be discussed in the next section. 5.5.1 Group-Growing Out Scheme Landais (1984) and Peleton (1977) gave a comprehensive description of an alternative production system that has proven effective in Ivory Cost. The strategy consists of a grouping of young males of village herds and rearing them for a certain period of time under the management of a village level institution similar to a producer cooperative. The growing out process utilizes young males of 18 to 20 months old which are pasture fed and supplemented in order to obtain rapid growth rates. At the end of the process lasting 8 to 10 months animal could be used for different purposes e.g. oxen, breeding males, slaughter animals. Participant farmers introduce one or more animals in the operation and continue to own their assets during all this process. The capital requirements as reported by Landais (1984) are equipment, veterinary supplies and credit. The basic equipments needed are : a park with adequate fences, a storage facility, a house for the hired herdsmen, and a scale to weigh animals. The veterinary inputs are vaccination, drug against trypanosomiasis, internal parasites and pesticides. The availability of credits is crucial for the cooperative 128 to meet initial and operating capital requirements cited above. At the initial phase animals are weighed and given a price according to the prevailing market prices. Owners are protected against any risk of loss or death since animals are insured. Participants will receive from the cooperative an amount of money equivalent to the initial value of the animal in cases of death or loss. At the end of the operation farmers could take their animal back or market them through the cooperative. The marketing of animals will be carried out in an auction fashion with advertizing and the invitation of potential buyers e.g middlemen, those seeking for oxen, wholosalers, butchers etc Still Landaisll984) reported that this system yielded satisfactory outputs in term of improved animal productivity and increased farmers' cash revenues. Such positive results were made possible by the increased offtakes of animals that were heavier and by increasing the bargaining power of farmers who witnessed substantial increases in cattle producer prices. Such an innovation in management practices reduces pressure on pasture feed resources through the increased animal offtakes. In addition, a great deal of benefits could be gained in terms of farmers effective organization which appears to be their pathway to get more out of their efforts. They will be, additionally, exposed to new managerial skills. 129 limitations. The prime constraint would be the availablity of credits earmarked for livestock production. Such financial conveniences are not readily available and this is likely to remain so in the forseeable future. This calls for some alternatives to lessen the initial capital requirements. For instance, all the needed equipment should be derived from local materials. Hallowed-out tree truncks can be suitable troughs. The initial labor investment should be provided by participant in a period of the year when the opportunity cost of labor is low. The local extension service which will provide the cooperative for technical assistance will guarantee the provision of veterinary supplies which will be repaid after animals are marketed. Another crucial issue deals with the availabily of a feed resource capable of ensuring rapid growth of young males. A feeding system based on pasture , groundnut hay and cotton seeds may be a solution. However , if this entreprise is potentially profitable from a financial standpoint, the feeding cost may hamper its economic viability because of high opportunity costs of cottonseeds. 5.5.2 Other Managements Issues The practice of excessive milking raised in chater 2 brings about the dilemma of human versus animal nutrition which is delicate. It would not be wise to divert milk from 130 brings about the dilemma of human versus animal nutrition which is delicate. It would not be wise to divert milk from human consumption considering the already critical nutritional status of farmers. A potential solution would be to promote farmers' supply with fish products which could partially replace milk as a source of animal protein without any major drawbacks regarding human nutrition. Reduced grazing time is an obstacle for higher pasture forage intake by cattle. Night grazing is advocated to overcome this constraint. A major shortcoming of that will work against night grazing is that manure production will be deposited in non useful places. Farm manure production deserves attention. A semi intensive production system consisting of the combination of limited grazing and stall feeding of oxen and selected cows is an alternative way of maintaining oxen in good shape during the dry season and a means of improving manure and milk productions. Cereal straws will be used as litters. The feasiblity of such innovations needs an evaluation on-farm with the preliminary assesssement of the availability of required labor and necessary capital e.g. housing facilities, transport etc. 5.6. Other required Investments The development of watering facilities and a fire control system are necessary investment which should conconmitantly carried out with any other development 131 strategy. However, their undertaking involves a great deal of resources and hinges upon the commitment of decision makers toward livestock productions. They must be carefully designed and implemented to prevent for example the conterproductive effects of the gathering of animals over limited watering points. 5.7 Promising Alternatives A host of systems alternative solutions have been scrutinized and their suitability examined. Is apparent that the complex social setting, the unpredictible physical environment as well as the interaction of the crop and livestock sectors though complementary can take a conflicting form, make it difficult to generate viable alternative solutions which will be applicable into farmers' production circumstances. The task of conceiving development avenues for the livestock sector in Upper Casamance will require a great deal of further research on station and specially on farm. The farming system reserch framework will be a valuable tool for devising alternative pathways. Despite the scattered information related to the livestock, this study has tried to design alternative solutions after the system structure and management have been specified and after major constraints have been identified. Promising solutions stemming from this study are of three classes i.e. 1/ solutions oriented toward the 132 alleviation of the dry season nutritional stress; 2/ solutions suggesting the alteration of some of the system design parameters, and finally; 3/ solutions proposing changes in aspects of the system management. They are illustrated in figure 13. l. The dry season feed restriction could be tackled by the means stated below: - supplementation of pasture forages and / or treatment of milllet and sorghum straws with biuret, - a better conservation and use of groundnut hay for supplementing pasture forage and crop residues, — intercropping of cereals with cowpea. Forage derived from cowpea will supplement dry season feed resources. - selective use of cottonseeds for oxen and lactating cows. 2. The proposed change in the system design parameters deals with paying more attention to small ruminants which could lead to changes in the livestock herd composition. 3. The suggested alteration of management practices are as below: - Group-growing out as described in this chapter. - semi intensive system of production.i.e. stall 133 feeding combined with limited grazing of oxen and selected lactating cows which will be fed residues, and cottonseeds. This alternative is aimed at improving manure and milk production and at maintaining oxen in good shape. These solutions are not definitive. They require that futher investigations be conducted at village level to assess their feasibility. The availability of more information would lead to an evaluation of their economic and financial feasibility and an alteration of these solutions. However they are promising and they open research avenues. The following research activities could be carried out on station: - animal response to different level of cottonseeeds supplementation - animal response to biuret treated or supplemented cereal straws From the above experiments the following on farm trials would be conducted. - testing the feasibility of the growing out scheme - testing of the realizability of the semi intensive production system of oxen and selected cows. - testing of the feasibility of a small scale fattening scheme of small ruminants. HERO HANROEHENT IMPROVED IHPRWED FEEDING SYSTEM 1L34 mature +emale —0 Improved ou tput male selected [:::::] innovation calves cows .-.-.9 (ceding priority Gamma-our ‘ (8-18 months) GRAZING + STALL FEEDING slaughter . animals . Heat Cash; Power : Manure Milk cereal straw conservation ]f I I ‘ I .’ cottonseeds I groundnut and cowpea treatment hay pasture biuret (orages ' supple- mentation CROP PRODUCTION ‘5\\~‘-‘~ intercropping cowpea-cereal Figure 13 Improved Management and Feeding Systems of the Cattle System of Upper Casamance 135 -.-~n~. ma.. ... 9....- ~..v-~_ w... ..— .....- ..w._-o_ ¢.. .._ 9...: .....__. w.. m.- ....¢~. o.»~__... a... v . ..o..~. ~...-v~.. m... m.- ...oo.~ ~...~... a... ..— ..o... ..m._-.. m.- ..— ..e..'~ .._.._.o m.. m.. ..o..~. ..«...... a... w.. ...._.. ~.oo-ono. m-.. n.. ..c.... ~..o.... ... ... ...... ..~o~_.oo ..- u.” ..ooorn a..o-u~. e.g ..~ ..ooo- ...~—..a~ mm.. 0.. ..cco-c ..gm~o_.n. 9.. ..c. ..o.... ~.m..... ... ... ..o... ...o..~.o ..u ..— ..uooo. ....-.- o.. ..~ ..uoo~. . .n ....—~ ......~¢~ nm.. .._ ....o.. ..ow~o~on. ..o o... o...... . PU , me L Q ~..u.o.. ... ... ...... ...o~.—.o o.- ..n —..o.o~ o.¢¢.oo.. ..- ..n 0.....- ..n—~..m~ om.- ..- ..ooo-c ..em~¢.o.. ..- a... ..co... m.-a ... ... ... 3:... 3:... :8... a. m. .2 a.» o.» ..m ..ooo~ ..oouu ..ooen -.~.-c— ~.~.~.- ..ucuccp ... ... ... 3 . .._ ... ..88 2:8 ..88 ~._o~‘~_ ~.~o-~. ~.—o-—— a. a. n. ... E Z :2. 3:. ..3: ..ao.-.. o.mo.-o. ..wo_-.. n. 3.. t. ... ... .._ 0....ou ..cooo— c.0000“ 2:... t=2.. :3... a. a. a. n. n. n. 9...: 3.2. ..so: 3...... 3...... 3...... a. 3.. e. n. e. a. 9...: 5:: 98:. 3...... 3:5. 38:: m... a... w... n. a. a. .6: 9:0. 9‘. 3.5: 9.3:: 38$: «.0 n.. m.. ..~ ..~ ..~ ..8... 9:... ..8... . q o . 000G .PMGUDQ 38. .c. unsh- uu‘.. an axle... .uqno ..seea ..e- .- nu... .g\-. u— .g..... ..a...;..... ..u. .u out.- .nguo a- n8..... files-.2. ..u- 3. ans.- .xs-o on To... assets: ..9- 3. ans.- .sasa o- .1..~.- ~.-...¢ao...- ..u. so on... -x\.- u- 08-..». ..nueu.‘no...- .2. up u-\.— q:\-. o- 2 06s. ..-.- ...‘....a ..a I .5. E... a. .3 .... a... ll. .2. 83...; £8: .t 1.8.. 8:: 2.. on: :95 _ u _ .532 .4 .xwocmaad 136 ..oo~¢m~m- _.¢mo.mmo. ..OMmor. ..a..._. o.o.¢¢m¢. ..o.m~o.o_ ..o..~.¢. o.oco~mm. 9......N. ..n.m_.. m.a._Nmo ..ucmma~ ....._.. ..p...¢c. p..~m.¢m ..3..... 0......” m.m.mom. ..Nm....._ ........~ ..~_omom ..~_.m_o ~.o..coo¢- a..¢-o.o. ..emmo¢. e.g..... ..ommomo. r..¢o.ooo. ......... ......... ..o....~. ..o..... ..~._... ........ ..o..... 9.....me. ........ 9....... .....~.. ..o..... ~...m.o_o ..~.ocow ..~_mm.¢ c..mo~o.n. ...oo.m-~ o.e¢e_nm o.occ¢~m o.cmocma_ _..moo.co_ ......... ....~.... ..o...._. ..o..... ..o.._.. ~.m..... ..a..... .....o... ..m..... ..o..... ..omaoo. 0....ccw ....mo.omm c.oweo~mm. ........- e.a¢e_cm e.=¢m¢~c 9.9moemo. o.~=-m‘o. ....~.... ....mmmom o.....o.. o.o~.omo 3.....3. e.g—coem ..emmccm o.m_--c. ...o.mow 9....... o....... m.mo.... .......... ..mm~_.~o. o.oemoooo. e.oe°_om o.oce.~o c.9moemo_ o.ommo_mo. a.....mom ..o....m~ ~....~... ..o..... ....-m. m....... e.g...mm a.m.~¢... c.0ocooo m..ocomm ~.3..... o..m~..m .......... ~.m..m~.n_ c..ooo~.c_ °.ooo.om 9.6.x... c.ommomo_ o..n_oomm_ o....oo.. e...~...~ ~........ o...~._. ........ o.~..... ........ ..3...~.. ..3..... ..N..... ........ ..m..... _.ooomw_on ....eommm- m.~o.¢cam_ 9.9mm... c.oa¢o~o e.g.momo_ m.-m..o~_ o.oo_.._m o.¢°oro~_ °.~_mow._ ~.~o¢.~o ..oo.~om o.o.o.- ...e~.o. m.mmoo'¢ ..mmo.m. o.~mm.m. o...n~m“ o.-o~o. ~.ooe..-_ 3...... o.~_ooom o.~_nm.. ..om_m¢wm- n.-~meom_ ..cnmco_ ..e..... ..emooma_ n.-.¢._m_ ......... ..o..~.o. ..~...... ....~... ..~.__.. ..e..... ..m..... ..m...... .._~.... ..c..... ..o..... ..o..... ~.oooo.e~. ..-.—m¢o~ ..~.=o.w ..~_nm_. 9.9moeme.- ~.cm¢o~.o. 6.9mmea. ..o..... e.g.ocmo. ~.om_oomm. ........m ..a....- ..._..... ~....~.. ..~..... ........ e...~... ...~._~.. ..a..... 3....... ..o..... ........ ~.cooe.o~_ ..-_.moo~ o.~_coom .22.... o.ne.oo¢m. o..a.wneo. ..enmoo. ..o..... ..OMoemo. ...oo.o.nm ..-.—m- ..~m_oo¢~ ..¢—_.~_. o.~owon~ ..uoo.ee ..mamom. o..m¢o¢. ..~m~o.c. ..~noonm ...oeoo. ..m~_m.m ..mom¢.m ~.ouoo~o~. o.~..m¢¢~ o.~.ocom ..~_nm.o u...e.mo.- o.mn__m.o. 9.9mm... ........ 9.9noomo. o.mmna.... ..o~.ooo. .._~....~ .._._.o_. ~..ma.m. ..~...~. .._..... ..~.~... ..omeooo. ....~... ..e...o. ...~_... .......3 ~.ooeo.o~_ €2.38 ..~—eoom ..~_nm.o ..ammcoe~- n..~oen.¢. ..emmo¢. e.g..n_. o.o~oom¢. n..~om.mn. ...en..o. ..o..~.c~ ..ono..~. o..o.... ~.....~. ........ .......m ..a~..... ........ ..o..... ...~.... ....~... ~.ooo.—o~. ..~_.m.¢~ ..~_ooem ..~_nm.o .o.. a... ..- .... c......=...o. .o.. ...c.. _~vr mon.ox 8.3.4:..- .... ..o.. ..u. .c.-...... .... .. .... o. ..u. .. .... .. .... .. .... ~. ..u. .. .... .- .o.. .- .ou. .. .... N. ..u. .- ...... a....c a... ..u. a... ...o. ..u. .o-aoau ..u. 09—3 .3. ..m ....=uo.o-.o o.ucou .< x.ocmau< 137 9...... .... ... ...... 3...... .... ....8: .....2. a... ... ...... ......3. .... .....2. ....2... 3.. ... ....3... ......n... ... .... ...... -"-H .... 9.3.... .... ... ....32 ...-.22 ... .... ...... .PUPL ....8... ... ... ...... 9.3.... 3.. ... ....I: .. 2...... ... .... ......~ .42.... ... ...... ......2' .... ... ....S: 92...... ... .... ...... «.3 . ...-.... ... ... ...... um.o CO_OUm_DO.O .63.... ... ... ...... .gmmnnm .62.... ... ... ...... ...: .mc0.p.LPDZ ...—3... ... ...—v.8. ... ... ...... .... a .8. I ...! .E‘ a. ...... '33:. ... .3 I .15 En. an 1.3.. ..Ilitst .3- I ...t I.‘ .- 1.8.. ..I.....I.9.n .3- 3 all 3.3 u. 1...... ...ls...:... ..l I .15 a... u. 18... . . ....13. ... ..U . all 3.3 u- , 3.. ..I ...‘3. ..3 ‘ a... [I .- 1 .... ..B. i .08 si- lit: ...... '. ... .28.... .m X.UcmuQ< 138 . ......_- ......a..- .......... .....N.... ........._ ......_.._ .....N. ........ ........- .....~.- ........- ........- .... a... ... .......... .. N. . . . . . . . . . ..... . ......._ . ........ N ......c. . ....... m ......o ......~.. ......... ......... ......e.. .......m. .... ............. ....._.. . . . . . ...... . ...... . ...... . ...... .......~ ....~_.. ........ ....._.. ....~_.. ........ ....~... .... ...... ... . . . . . ..... . ...... . ...... . ...... . ...... .......~ .......~ ....N... e.g..... ....N... ....N... ......._ .... ....ox 2.3.4.2. ...~N... ....~... ....N... ......m. ...-... ....N... ...-... ....N... ......m. ......n. ...-... ......n. .... .3: ......... ......... .......... .......... ......... ......... ....._... ......._. ......... ......... ......... ......... .... .s!3: . ......~ .H....~.~ .H....... ......~.m ......._. ........~ ........~ ........~ ........~ ........~ ........~ ........~ .... .. ......_.. ......~.~ ......._. ........m .......N. ........_ .....N.. ......e.. .....~... ......... ...m._... .......__ .... .. ........ 9....... ......N. ......o. ......c. ....N... ........ .......m ........ ......o. ........ .....Nm. .... .. ........ ~....... ...._... ........ ........ 9...»... ......o. ........ ........ ........ ........ ......_. .... .. ........ .....~.. ........ ....._.n ......~. .....~.. ...._.~. ........ ........ .......n ......mm ......mm .... .. ........ .......N .....c.. .......~ .......N ....o... .......~ .......~ ........ .......~ .....~.~ ......o. .... .. .....w.~ ......o. ......o. .....~.. .......~ .......~ .......~ .......~ .......~ .......~ .......~ .......~ .... .. ......m. ......w. ......m. ......m. ......o. .......o ........ ......m. ......m. ......e. ....~... ........ .... .- ........ ......c. ......N. . ...... ...._... ........ ........ ........ .......~ ........ ....3... ........ .... .. ...._... ........ ........ .......~ ....~... ........ .......~ ........ .......~ ........ ....~... ........ .... .. M.._.N_. ..._3... ~......w .......~ .......~ ........ ........ ........ ........ .......~ ....~... ........ .... ~. ...... . ...... . ...... . ...... ........ .......~ .......~ .......~ ..._~... ~......~ ....._.~ ...—.... .... .- ...... ...... a... ......... ......... .....~..- .....N.... .........N .......... ......... ......... ......... ......... ......... ......... .... .... ...9. ....._... ......... ....n_... ......... .......m. ...n._... .......m. .... ...:o.. ....N... ....N... ....~...- ....~... ....~... ....~... ......o. ....N... .... 8: ......o. ......o. ........ ......o. ........ ......o. ....n... ......u. ..w. x . .....oo.o-.. U.uCOU .m x.Ucm0Q< 139 .-. ...: ......5- . . 3...... .....s... 23...... 2:5... ...: z... 33:... 3.2...- 3.2...- 23...... 22.....- .... a... :- _......... 2.5.... ......z... 25:... 28...... 2:32.. 23...... 22...... 23...... 2.5.... 2.2:... 22...... .8. 3‘....B. 2...... 2...... 2...... 2...... ......l .....i 2...... 2...... 2...... 2...... 2...... 2...... .3. ...I. 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... .8. . €33 ...-8...... 2.2.... 3...... 2.3.... 3...... 3...... 3...... 3...... 2.3.... ....I... ....8... 3...... 3...... .8. .8: 22...... 2.8:... .....2... 2.5.... 28...... 2.28... 22...... .....8... 2.5.... 2.3:... 22...... 2.5.... .3. '1'“: 2.2.... 2.3.... 2:3... 2.3.... 2.3.... 23...... 2.2.... 2.2.... 2.2.... 2.2.... 22.8.. 2...... .3. : 2...... 3...... 2.2.... 2...... 2:8... 2.2.... 3...... 23...: 3...... 22.... 2.2.... 2.3.... ...: .. 22:... 3...... 3...... 25.... 3...... 28...... ......8. 2...... 2.3.... 3...... 2:3... 3...... .... .. . . ; . ...... 2...... 2...... 2...... 2...... 2...... .... : . . ...: . ...... . ...... . ...... . ...... .. . ; . . . ......mx ”9.... 2...... :3... .52.. .....z. . 3.... . ...... . ...... . ...... . ...... «.22.. ..8 .. ... .... .... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... . ...... .... a ......... ....u... . ...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... .... .. ”xv“. 2.3.... 3...... 22:... 3...... 3...... 2...... ....c... 3...... 2.2.... 3...... 3...... .... ... ...M .. 2...... 2...... 2...... 2...... 3c... 2...... 2...... 2...... 2...... 2...... . ...... .... .. "3w." . 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... .....8 .... .. 1...". 2...... 3...... 2...... 2...... 2...... 2...... 2...... 2...... 2...... 2:...” . ..."... .un .- . . . . . . . 2...... 2:... .. . .. . .. 2...... 2...... . ...... . ...... . ...... . ...... . . ..U . ...... . ...... ...... 838 .... .. 5...... 23...... 23...... 23...... 23...... ........8 ......2... 22...... 22...... .....E... 22...... 22...... ..n. .... ...... . 2...... 2...... 2...... 2...... 2...... 2...... ..v. 2...... ...-8.. _ - . 2...... 2...... 2...... 2...... 2...... 2...... .8. . ...... . ...... .... . 2...... 2...... 2...... 2...... 2...... 2...... .3. 9...... . ...... in .5888... ... 2...... 2...... 2...... 2...... ... ... ... ... ... ... .. .uu ... ... ... ... a... ... ... ... ... .H. . ...... ...... ...... . .... ...! .l... .u...m.o 3.6u 0so> mc.umz ..mooum .mco...yu32 .u x.Ucmocq 140 5...... ......mm.. .....at. ........ ...»...r. ....p..... ...o.~.n. ......mpu ..oorr.~_ ....»... ”......o .,.o.... ........ ......... ........ ........ ....m... ....n..v .......... ....o..m ........ ........ ...'_~o.m. a..-.~o.r. ..onmor. 9......- ....oonr. 13:32 ...—no..- ....woo-c o.—orr.~— ........ ...._... ........ ........ .....o... ........ ........ ........ ........ “..mmo.—n_ ....wo.n ..~.nqcn ~.m.—o.om~ ...oo..-~ o...u¢~. ...n-omr— ...n-nsoo. ......... 3...... 2......— ........ ........ ~.mm¢..w ........ ......... ........ ........ ........ ......u. «...o.__uq ~..u.-—~ ... ... ...-o. p..p..amm~ ....~__o- ...on.~. ......mo. ...-.33. o.~.-.~n .....owon ...-....“ ..r~.-.o ......»o u...mooa ..wnaoea ......«u- ......rm ....m... ......uq 9.2.8 f~::_9 ~.mo—-n~ ... ..- ....o. . ......o. o..o—onoo. ......mo- v.88...— o..~..mna o...p.3~ ~.on-nu. ......mu ........ ......3 ........ ......... ......o. ....oonn ........ ........ m.~.o.._v~ ...»...- ... ... ...... ......e.¢~ ...-o¢~.o. o..-¢~. ...n-omr— ...».nomn- ...owOOqe o._.~uwn~ ~.o.~o-u_ ........ ........ ........ ........ ......... ........ ........ ........ ........ s....o..w~ ........ ... ... ...... ....__.-. n.~u.ooom. .63.: ..-o_.o~. 28:... 2......— ......oo. 2...... ......ua 2...... 2.....- 2...... 2...... 2...... ...—.... 2...... ....»u..._ ......u. ......qn ..~_n.on ...—..oé.- n..-nooo. ....»uo— .632- ......e.. n..-...n. o.-—-.n ......m.. ......s.. 93:: ..-..-a o...o.- ...-...v ....»u... ....rrr- ....po-c m..~._un ...-.... ~.n..¢.... o...o..~ .....nyn o.~_n..n ..u~.--- ~.o~¢o~.e. ......o. p.233. ~.-n_n-a-— ......... ...._..- ......... ......cp ......u. 2...... 2...... 3...... 2...... ........ ........ ........ .4323. .63! o.~.~.¢o o..ro~n.~- ...o-oooo. ..onnoaw ..o-u_q e.g-82 .._'o_-nm o........ ........~ ......... o.~..o.. ......ou ..wuaunu o..orrrc -.~a~ocom ..~—.our o._prooc ...~_w.~ o..¢mo.a ~.~u.-—... ...—.... o..~a-o o.~_-or ...-2.8. ...»...po. ...—n..— ..nnnu~_~m 2:... 3...... 3...... ........ 2...... 2...... ........ .......o. 2...... ...-.... 2...... 2...... ~.~¢w-.o.. ...-.... o..~np.n ..~_n'¢o .6332- ~oon_o. ...-n_~ asgt. .......- 2.3.... 2.8.... ...-...— 2...... 2...... 2...... 28...: 2...... ......o. 2...... 2...... ~.~r.-ou. ...~..-n o.~.—.¢o .3 2.. ... ...- S‘dfisp ...- ...... ...- .....- 8.32:... ..n- an... ..a- .cn.uo.~.u .... .. ..l .. ..8 .. .... .. .... .. ..l 2 .... .. .... .- ...u .- ...a .- ...c .- ...n .- ...... 838 .... .... a... ...a. ...- u. an... .3... on ...-9 ...-cc fl.+cou .u ..Dcmooq 141 Appendix 01. Democraghic ModeI Outputs: RunZ. (Oxen ' PoouIatIons. Number of male Sold, and Oxnn Balance.) I 2 I I 6 6 I 0 9 I0 HoIdinq * EI‘I at: (years? One» I I9 60 S6 66 II 79 0! 0I 0I 07 Popqu~ IIon I II 60 00 II 90 I00 I06 I07 II2 III 6 66 63 76 00 I06 III I26 I30 I33 I30 6 60 66 00 90 III III IIO I60 I66 I60 I 60 60 03 I00 IIO III I6I I66 III I02 0 60 70 05 ' IOI I22 II2 III III III III Iole solo I I9 36 SI 70 00 IIO IJI I62 I90 2I0 I II 20 II 6I I0 06 I03 III I60 I00 I II II II II 66 69 06 III, III I66 6 I0 20 29 30 II 60 09 I0 I06 I20 I I II 26 33 II II 66 69 06 906 0 0 I6 23 30 II I2 II 66 69 II 0190 Balance 3 ~22 ~20 ~27 ~2I ~23 ~2I ~20 ~32 ~30 ~I2 I ~II ~20 ~I6 ~I ~7 ~3 ~6 ~9 ~I0 ~I6 6 ~I6 ~I6 ~9 0 9 II I6 I6 II I 6 ~I3 ~II -I 6 I6 20 II II 33 32 7 ~I2 ~9 ~I I0 2I II II II 62 63 0 ~II ~I I I3 26 39 I9 62 67 II (—) Oxen Deficit (+) Male Surplus AFrpr flvpn quuIrements are met 142 Aopendix DZ. Demograohic NodeI Cutouts: . Run 3.(0xen ooou'ationc Number or maIes Sold, and Oxen 9aIance.) -' SImuIatIon tIme (years) I 2 ‘3 4. S 6 7 l I N IIcIIcIIru; perIod (years) Oren I II II II 70 II II II II II II Population I II II I2 II II III III III III III I II II II II III III III III III III I II II II III III III ' III III III III I II II II III III III III - III III III I 62 II II III III III III III III II INIIMI ' . I II II II II II III III III III III I II II II II II II III III III III ' I I2 II II II II II II III III III I II II II II II II II II III III I I II II II II II II II II III I I II II II II II II II II II Oven IIIIIII I ~II ~29 -II ~20 ~20 ~20 ~26 ~II ~II ~II I ~II -II -II -I ~I I I ~I -I ~6 5 ~II ~II . ~I I II 1!. IS 25 23 23 I ~I2 -0 I II II II II II II II 7 ~II ~6 I II II II II II II II I ~I -I I II II II II II II II 5-3 Oxen DeFIcIts \+, HaIe Surplus after Oxen RequIrements are satisfied 143 03. Demngraphic MOdPI Outpgts. Run 4( .(Dx. I>< . Append Population. Number Of Male Sold. and fixpn RaIaanI SimuIation time (years) I 2 3 I 5 6 7 8 9 I0 Holding Deriod (vears) Oven 3 SO 53 b3 77 9I I00 IOI IIO .338 87 pIOIIIIIIIII 9 55 II 75 93 IIO 127 I35 144 352 161 5 58 II 8? I03 I2I III III III I86 I9h 6 59 70 87 I09 13‘ I6I I79 I99 215 227 7 II 72 92 III I39 I68 I95 2I7 239 258 8 I? 75 9S II7 III I73 203 232 256 28I Hole soIn 9 l9 37 55 79 93 120 I53 I88 224 261 9 I5 79 43 Sb 7‘ 93 I?! 352 I86 229 5 I? 2‘ 35 46 57 73 9‘ I2I I52 I85 I I0 20 30 IO I9 97 75 9S ' I2I ISI 7 9 I0 26 35 I3 50 59 76 95 Ill 8 8 l6 2‘ 3| ‘0 44 SI 60 77 97 Oxen BIIInce 3 ~20 ~25 ~2I ~II ~5 ~3 ~6 ~5 ~I ~2 I ~l6 ~II ~8 3 I3 2‘ 26 29 29 32 S ~I3 ~II ~I II 29 42 $2 58 63 57 6 ~I2 ~8 3 I9 37 $8 70 93 93 9B 7 ~I0 ~S 8 2‘ 43 65 85 IOI Il7 I29 8 ~9 ~2 II 27 II 70 93 III I34 I52 + é-g Oxen Deficit Hale SUFPIUS After Oxen reouiremenis are met. 144 Appendix D4. DemographIc Model Outputs. Run 4( Ox. PopulatIon, Number 0F Male Sold. and Oxpn Balance) SimuIafion timp (years) I 2 .3 I 5 6 I 9 9 I9 HoIdino Period (years) oxen 3 56 65 77 99 I00 I97 |I2 III I26 I36 POpulation I 6| 73 9| |I2 I25 I36 III I53 I63 I73 9 6| 70 98 III III III III I95 I97 29 I 65 II III I29 |55 III I99 2I5 229 299 7 66 93 I07 I33 I69 I99 2I9 239 259 275 9 67 95 III I37 I65 I94 22I5 259 299 393 Halo solo I I9 II 59 79 III I35 III 209 III 296 I I5 29 II 57 99 III I36 ' I79 m M. I I2 29 35 II II 79 III I36 I79 III I II II II II II 57 99 I95 I35 I69 I 9 I9 26 35 I3 59 59 9| I95 I35 9 9 I6 2I 3| 39 IS 52 50 9| I95 Oren laIInce 3 -I5 -I3 -6 -I I I 3 9 3 7 I °I0 -I 9 22 29 33 35 37 II II 5 -I 9 I5 32 59 59 55 69 II 79 6 -6 I 20 39 59 79 99 99 I97 IIS 7 -5 6 23 I3 II 96 I99 I22 I35 II6 9 -I 9 26 I7 69 9| II6 II2 I59 I75 (-) Oxen DEFIcIt (+7 Male Surqus After Dyan requirements are met. 1u45 FALL p~092ooS2.uc199.ncz. DISP SE.--.A. FTNS. LGO. PROGRAM THESIS REAL DELI.DEL2.DEL3.DELI.DEL5 DELE DEL7 PLRI an2.PLna.ana. . ans.PLR6,PLRT,PDPCU.PDP§H,prFc.PDpnc bopvn P x - POPBLS V!NTI§900).VINT3£I200£ VINTIIDDD .V;NTS(900) . VINT7é§OOI R (to R5(IO an. 9 pop 99 ,OXROT BRT, . anu.a F.LNOO.LND.C.SF.SMB.SMO.TAS.OUR.D .ADOR,POPAM INTEGER N1,N3.N4.N6.N7.NIPP OPEN(10.FILE-'OUTPUT') Ctt.OQ......3.0.0.0000.........OOOOOOOOOOO......‘O......OOOOCOOQOOOOOCOO VARIABLE DIRECTORY I C CtCO......‘O....OO‘0.0.0.0000....COI........ICOOCCOOOOOOOOOOO0.3.0.0.... DELI.DEL2.0EL3 DEL4.DEL5.DEL6.DEL7 - LENGTH OF THE DELAY pnocéss FOR DELAY I.2.3.4.5.6.7 RESPECTIVELY. PLR1.PLR2.PLR3 anI ans.ans.anv . PROPORTIONAL LOSS RATES DEATH LDSSESI rofi DELAY 1.2.3.4.5.e.7 RESPEC IVELYL POPCV : POPULATION OF CONS POPFC : POPULATION OF FEMALE CALVES POPMC : PDULATION OF MALE CALVES POPYM : POPULATION OF YOUNG MALES POPBLS: POPULATION OF BULLS POPOX : POPULATION OF OXEN POAM : POPULATION OF MALE ANIMALS VINT9,VINT3.VINT4 VINT6.VINT7 ; INTERMEDIATE RATES x~ . THE DISCRETE DELAYS 1.3.4.6.? RESPECTIVELY . R2.R5 : INTERMEDIATE RATES OF DISTRIBUTED DELAYS 2.5. * OXRQT : NUMBER OF OXEN RE IRED BMRQT : NUMBER OF BREED! MALE 00.... O REQUIRED ADOR . gQEQNCE BETWEEN ACTUAL AND REQUIRED NUMBER OF ' BMB : DIFFERENCE BETUEEN ACTUAL AND REQUIRED NUMBER OF ‘ BREEDING MALES BRT : NUMBER OF CALVES BRM : NUMBER OF MALE CAL BORN BRF : NUMBER OF FEMALE CALVES BORN SR : SEX RATIO BR : CALVING RATE VIN6 : INPUT RATE TO DELAY 6 VIN7 : INPUT RATE TO DELAY 7 LNDO : INITIAL HECTARAGE CULTIVATED LND : CULTIVATED AREAS 5F : NUMBER OF FEMALE MARKETED SMA : NUMBER OF MALE MARKETEO C...‘.....‘0............OOOOOOOOOQC......lfititt I O I O Q o o I o o ,0 o o D o s o o D o O o O O O O 000000000000000000000000000000000000000 SET -MOOEL PARAMETERS DT - .05 DATA DEL;.O§L3.DEL3.DEL4.0EL5.DEL6.DEL7/9..2..10..9..2.. DATA K2.KS7233/ NI-DELI/DT Na-DELa/DT NI-DELI/DT NB-OELG/OT N7-DEL7/OT DATA PL31.P3§§.PLR3.PLR4.PLR5.PLR6.PLR7/ .25..10..06..25..10. DATA 8R1.SR.C.SMB.SF.SMO.SM/.46..5.6..4t0./ c SET INITIAL COHORT POPULATIONS PORCH 8 I75. 1446 TPOP11)' 3B2 C INITIALIZE INTERMEDIATE RATES Do 1 - 1 N1 100 VINT??II - POPFC/DEL1 Do 11? Is 1 N3 110 VINT3 I)- POPCU/DELa 120 VINT4?I) - PDPnc/DELA 0013? I.‘ 130 VINTG I -PDPBLS/DEL5 140 VINT7?II - PDPDx/DELT 150 92(ITO- PDPPHIDELz 16 I - 150 RS(I)O - POPMC/OELS C SET INITIAL FLOU RATES VOUTA' VI N6 '0. O VIN7 I ROUTS HPITE 19 170) 170 1x 'T' 4x POPFC' X’POPMC' ax. 'POPRH' 3x 'PDPvn ax’PoPcv1'4x;'PDPDLS’,2.1PDPDx’3x,'SHA 6x. no 5x 'SMB'. 6x. 'TSA .sL 'eua .6L 'LND'. 5L ADOR'. 'sx. 'TOTPOP I C SET RUN SPECIFICATIONS DUR ' 10. NIPP- 20 C SET TIME EQUAL ZERO T '0.0 C EXECUTION PHASE DO 200 I 0 1. DUR DO 210 d ' I. NIPP C UPDATE TIME TI T+DT COMPUTE MATURE FEMALES POPULATION SIZE AND OUTPUT RATE FROM DELAY THREE CALL DELAYI(ROUT2.VOUT3.VINT3.POPCH.N3.DT.PLR3) C CDMPUTE HEIFERS POPULATION AND OUTPUT RATE FROM DELAY THO CALL DELAY2(VOUT1.ROUT2.R2.PDRH.PLR2.DEL2.DT.K2) SSEPUTE FEMALE CALVES POPULATION AND OUTPUT RATE FROMOELAY O. 00 no CALL DELAY1(BRF,VOUT1.VINT1.POPFC.N1.DT.PLR1) 0 900 0 0000000000000 000 0 1J47 COMPUTE HECTARAGE CULTIVATED LND - LNDO~(1.+.10-T) COMPUTE NUMBER OF OXEN REQUIRED OXRQT - (LND/C)‘2 COMPUTE BULL POPULATION AND OUPUT RATE FROM DELAY SIx CALL DELAY1(VIN6.VOUT6.VINT6.POPBLS.N6.DT.PLRG) COMPUTE OXEN POPULATION NAO OUTPUT RATE FROM DELAY SEVEN CALL DELAY1(VIN7.VOUTT.VINT7,POPOX.N7.DT.PLR7) COMPUTE YOUNG MALE POPULATION AND OUTPUT RATE FROM DELAYS CALL DELAY2(VOUT4.ROUT5.R5.POPYM.PLR5.DEL5.DT,K5) COMPUTE NALE CALVES POPULATION AND OUTPUT RATE FRON DELAY4 CALL DELAY1(BRN.VOUT4.VINT4.POPNC.N4.DT.PLRA) COMPUTE NUMBER OF MARKETED ANIMALS MATURE FEMALES SF ' SF+DT¢VDUT3 BULLS SMB I SMB+DTEVOUT6 OXEN SMO ' SMO+DTOVOUT7 TOTAL MALE SMA'SMO‘SMB TOTAL ANIMAL TAS'SMA+SF COMPUTE TOTAL POPULATION TOTPOPIPOPFC+POPMC+POPCH+POPRH+POPYM+POPBLS+POPOX 0.0....OOOOOOOOOOOOOOOOOOOOO...-............‘OOOOOOOOOOOOO00...... THIS PART OF THE PROGRAM IS USED FOR INTERNAL CONSISTENCY USING THEORETICAL AND SIMULATED DEATHS SIFC'DT'VINTI N1 $1MC:DTtVINT4 N4 TPOP IF1 (1)- +52 éPOP FC+POPMC -PLR1+ POPYM+POPRH)‘PLR2 cw- PLR3+ OBLS+PO mOX&PPLR6 RITE 0.900) T. SIND H THDTHD FORMA ' '.r4.2. 3L 2(Eé.a .ax). 2L F7. 2 ‘ HMO-0005M“ O......COCCOOUOO..........O......‘fiitttt0.0.0.........OOOOOOOCOOOO... COMPUTE BIRTH RATES BRT'BR1tPOPCU BRF-BR T‘SR BRM-BRF ALLOCATE OUTPUT FROM DELAY 5 TO DELAY 6 AND 7 POPAM ' POPB S * POPOX BMR QT IPOPCU 3O ADOR IOXRQT-POPOX BM? -POPBLS BMRQT IF BMB. GT. ogm IF(ADOR. G O. O)THEN 1J48 N6‘ROUT5'81.-R1) N7-RDUT5‘ 1 .1. 8)THEN U 5 ' OUTS‘RRZ ' OUT5‘(1.-RR2) n1: 0 0 END IF 210 CONTINUE HRITE (10. 270) T. POPFC. POPMC POPRH. POPYM POPCU POPBLS. P0P0x. . SN A. 5M0 TA§.M m0 OTPDP 270 FORMAT( O'.F4. 1.2L 1165"8 1. 3x F5.1.3L rs. 1.2L F6. 1. 3L F0.1) 200 CONTINUE STOP END 00 -1 N STRG=STRG+VINT(I)'OT SONTIW BRDUTINE RDESAY2(RIN. .ROUT. R. STRG. PLR, DEL. OT. K) MEN SIO ON R“ I i 1 KM I IRéII+A‘(R(I+1)-BPR(I)) &. 2 IR K)+A'(RIN-B‘R(K)) 00-5T:¢+R( I )‘fiDEL/EK) ROUT-R 1) R URN Appendix E. Mathematical Procedures of the Demographic Model. The theory and mathematical developments used in the model are from Manetsch(1982). Biological processes such as gestation and maturation can be conveniently represented by time delays. There are two types of commonly encoutered time delays: discrete delays and continuous or distributed delays. The discrete delay is defined by the following difference equations: Q(t) = O (t-DT) (1) l O (t)= O (t-DT) (2) l 2 O = I(t-DT) (3) n—l or Q(t) = I(t-T ) (4) where: 0(t)= lagged variable I(t)= unlagged variable O.= intermediate rate of the ith stage DT = time increment The delayed variable is simply the unlagged variable shifted in time by T time units. When the time delay is distibuted rather than dicrete , 149 150 this means that, for aggregate flows individual entities in the aggregate have different lag times; so that while entities may enter the delay process at the same time, the output flow will be ditributed over time(Abkin and Wolf,l976) .The following differential equation define a distributed delay: where: k k- d y(t) d y(t) ak + ah4_ . + ... + a‘ y(t) = X(t) —| c311;k at:K (5) X(t) = the unlagged variable y(t) = the lagged varible k = the order of the delay and represents k sub-population in the cohort. The intermediate rates in the delay process can be easily manipulated to compute the actual population size in the cogort at any time. The total number of individuals in the t i stage of the distributed delay, is a function of the flow rate and the lenght of the delay: Q (t)= D/K * R (t) (6) i i where: th Q t = storage in the i stage of the delay i D = lenght of the delay K = order of the delay R = flow rate of the ith stage i th The net change in storage in the i stage of the delay is computed from the following differential equation: _ A .4, l _____._ __.__4l 151 d Qi(t) = R - R (t) - L (t) (7) i+l i i dt where th R (t) = Input rate in the i stage i+l th R (t) = Output rate from the i stage 1 th L = Loss rate of the i stage of the delay at time t Losses occur at each stage of the delay. Propotional loss rates are applied toeach stage of the delay. L (t) = PLR(t) * Q (t) (8) i i Where: PLR(t) = proportional loss rate Q (t)= storage in stage i i Storage in the delay process for the distributed delay and the discrete delay respectively as follow: K. D Q(t) = -——{2: Ri(t) (9) K f=l N Q(t) = DT*Z R.(t) (10) where: i=1}. D = delay lenght K = order of the distributed delay R (t)= intermediary rate of the ith stage 1 DT = incremental time in the simulation process 152 Q(t) storage in the delay process N = number of stages in the discrete delay BIBLIOGRAPHY Abkin,M.H. Wolf,C. 1976. Distributed Delay Routines: DEL, DELS, DBLF, DELLF, DELVF, DELLVF. Computer Library for agricultural Systems Simulation. Agricultural Analysis and Simulation Projects. Department of Agricultural Economics. MSU. Class Document no. 8. 1976. Achterstraat,A.N. 1983. Agriculture d'Autosubsistance on Agriculture de Rente? La Prise de Decision des Paysans entre Differentes Cultures et 1' Impact de l'Intervention de la SODEFITEX et de la Culture Cotonniere sur les Societes Peul en Haute Casamance. l'Example du Village de Lenguewal. Institute of Cultural Anthropology/Non western Sociology Free University. Adegbola,A.A. Smith, 0.8. 1982. Use of Pasture and agro- industrial by-products for Ruminant production in West Africa. FAO Animal Production and health Paper no 32, p. 115-126 Alexander, 6.1. 1978. Non Protein Nitrogen Supplements for Grazing Animals in Australia. FAO Animal Production and Health Paper no 12 p. 93-96 Ange, A. 1984. Les Contraintes de la Culture Cotonniere dans les Systemes Agraires de la Haute Casamance au Senega1.Th. Doct. Ing., Geol. Appl., INA, Paris Grignon. Ariza—Nino,E. Shapiro, K.H. 1984. Cattle as Capital, Consumables and Cash: Modelling Age-of-Sale Decisions in African Pastoral Productions in Simpson and Evangelou ed., Livestock DevelOpment in Subsahan Africa.1984. Balch,C.C. 1976. The Potential of Poor-Quality Roughages for Animal Feeding. FAO Animal Production and Health paper no 4. p.1-15 Banque Mondiale .1983. Rapport D'Evaluation. Senegal. Projet de Developpement Rural du Senegal Oriental. Banque Mondiale . Rapport No 4297-SE. 1983. Bersten, R.H.; Fitzhugh,H.A.; Knipscheer,H.C. 1983. Livestock in Farming Systems Research. Paper Presented at the third Animal Farming Systems Symposium, Kansas state University, Manhattan, Oct. 31—Nov.12, 1983. 153 154 BOAD.1984. Etude Prealable du Projet de Developpement Rural integre de Bonconto ( Casamance ) en Republique du Senegal. Zone test de Mballacounda. Agro progress- Kienbaum International GmbH. Boudet, G.l970. Paturages Naturels d3 Haute gt moyenne Casamance L Republique g3 Senegal 1. IEMVT. Etude Agrostologique No 27 IEMVT Calvet, H. 1979. Les Sous produits AgroIndustriels Disponibles au Senegal et leur Utilization en Embouche Intensive. 9ieme Journees medicale de Dakar Casse, Dumas, and Garin. 1965. Bilan des EXperiences de la Culture Attelee en Afrique Occidentale d'Expression Francaise, Guinee Exeptee. Tome II . IEMVT. CEEMAT. 1975. Manuel d3 Culture avec Traction Animal. Techniques Rural en Afrique. Coulomb,J. Serres, H. Tacher, G. 1980. L'Elevage en Pays Saheliens. Presses Universitaires de France. CRZ/KOLDA. 1882. Rapport d'Activitees Annee 1981. ISRA. Senegal. CRK/KOLDA. 1983. Rapport d' Activitees Annee 1982. ISRA, Senegal De Reviers,B. 1979. Considerations sur les Possibilites des Productions Animales en Casamance. SOMIVAC-Senegal. Diao, I. 1985. VII-plan Quadrienal de Developpement Econonique et Social. Inspection Regional de la production Agricole de Kolda. Donefer, E. 1976. Physical Treatment of Poor Quality Roughages at Commercial and Farm Levels. FAO Animal and Health Paper No. 34. p. 17-23 Eicher, C.K. Baker, D.C. 1982. Research on Agricultural Development in Sub-Saharan Africa: A critical Survey. MSU . International Development paper No 1. E1 Naga, M.A. 1984. Potential for the Better Utilization of Crop Residues and Agro Indrustrial By-products in Animal Feeding in North Africa with Special Reference to Methodology, Equipment, Facilities and Personnel Involved, as well as an Outline of Research Priorities of the Region in Guidelines for Research Egg the Better Utiliuzation of EEBp Re31dues and Agro Industrial By-Products ii Animal Feeding 12 Developing Countries. Proceedings 155 FOA/ILCA EXpert consultation. 5—9 March 1984. Fonnesbeck, P.V. Kearl, L.C. Lorn, E. Harris,E. 1975. Feed Grade Biuret as a Protein Replacement for Ruminants. A Review. Journal of Animal Science, vol. 40, No.6, 1975, p. 1150-1179. Gueye, E. Nicolas, A. Gandemer, G. 1979. Comportement Ponderal des Taurins Ndama , au cours des Saisons Seche et Pluvieuses: Influence du Regime Alimentaire. ISRA, CRZ/Kolda. Girardot-Berg, I. 1982. Livestock: Policy Issues in Trade, pricing and marketing. in Fine, C.J. and Lattimore, R.G. ed., Livestock in Asia.. 1982. Hardin, G. 1968. The Tragedy of the Commons. Science, vol.162, 1243-1248. Hart, R. Onim, M. Russo, S. Matuva, M. Otieno, K. Fitzhugh, H. 1984. An Analytical framework for feed Resources Research on Mixed Farms in Western Kenya. Winrock International No. 46 Horowitz, M.1980. Research Priorities in Pastoral Studies: An Agenda for the 19803 in Galaty et al. ed., The Future of Pastoral People. Proceedings of a Conference held in Nairobi, Kenya, IDCRC. ILCA. l979.(a)Trypanotolerent Livestock in West and Central Africa.vol.1 General study. ILCA Mimeograpg 2. 1979. . l979.(b) Small Ruminant Production in the Humid Tropics . ILCA Systems Study No 3. . 1982. Evaluation of the Productivities of Djallonke Sheep and Ndama Cattle at the Centre de Recherche Zootechniques, Kolda, Senegal. ILCA Research Report No.3 Jackson, M.G. 1978. Treating Straw for Animal Feeding. An Assessement of its Technical and Economic Feasibility. FAO Animal Production and Health paper 10. Jahnke, H.E. 1982. Livestock Production Systems and Deve10pment in Tropical Africa. ieler Wissenschaftesverlag Vauk. 1982. Jarvis, S.L. 1982. To Beef or not to Beef? Portfollio Choices of Asian Small Holder Cattle Producer. in Fine, J.F and Lattimore, R.G. ed.,Livestock in Asia. 156 Jaske, M. 1976. System Simulation Modeling of a Beef Cattle Enterprise to Investigate Management Decision Making Strategies. Thesis(Ph.D) Michigan State University. Dept. of Electrical Engineering and Systems Science. 1976. Kategile, J.A.l983. Utlization of Low Quality Roughages With or Without NaOH Treatment. in Kiflewahid, B. ed., By—Products Utilization for Animal Production. Kang, B. T. Wilson, G.F. Sipkens, L. 1981. Alley Cropping Maize (Zea mays L. and Leucena Leucena leucocephala) in Southern Nigeria. Plante and Soil 63, 165- 179 Karbe, E. 1980. Trypanotolerance. Manuscript for the FOA- Workshop, 26.11—5.12. 1980 in lome, TOgO Khajaren, S. Khajaren, J. 1984. Potential for the Better Utilization of Crop Residues and Agro-Industrial By- products in Animal Feeding in Southeast Asia, with Special Reference to Methodology, Equipement, Facilities and Personnel Involved as well as an Outline of Research Priorities of the Region, in Guidelines for Research on the Better Utilization of Crop Residues and Agro- -1ndustrial By- products in Animal Feeding 1n Deve10ping Countries. Proceeding of FAQ/ILCA Expert Consultation 5—9 march 1984 Kearl, L. C. 1982. Nutrient unirement 9f Ruminants in Developing Countries. International Feedstuffs Institute Utah State University Kiangui, E.M.I. Kategile, J.A. 1981. Different Sources of Ammonia for Improving the Nutritive Value of Low Quality Roughages. Animal Feed Science and Technology 6(1981) 377-386 Kiflewahid, B.l983. An Overview of Research Methods Employed in the Evaluation of By-products for Use in Animal Feed. in Kiflewahid, B. ed.,By-Products Utilization for Animal Production. 1983. Kowalczyk, J. 1976. Maximizing NPN Use in Feeding Systems Based on Agro-Industrial By-products. FAO Animal Production and Health Paper No. 4 p. 173-185 Landais, E. 1983. Analyse des Systems D'Elevage Bovin §edentaire du Nord de la Cote d'Ivoire. Tome I; Donnees Zootechn1ques et _Conclusions Generales. IEMVT. These de Doctorat es science. 157 . 1984. Element pour la Preparation du Programme de Reserche sur les Systemes de Production et 1e Transfert des Technologies en Milieu Rural. Haute Casamance et Senegal Oriental.ISRA/GCAS.Senegal. Mack, S.D. Sumberg, J.E. Okali,C. 1984. Small Ruminamnt Production under Pressure. The Example of Goats in Northern Nigeria in Sumberg, J.E. Cassidy, K. ed., Sheep and Goats in Humid West Africa. Manetsch, T. J. Park, G. L. 1977. Systems Analysis and Simulation with Applications £9 Economic and Social Systems. Part II. MSU. Department of Electrical Engeneering and Systems Science. . 1982. Systems Analysis and Simulation with Applications £9 Social and Economic Systems. Part I. Methodology, Modelling and Linear Systems Fundamentals Department of Electrical Engineering and Systems Science. MSU. . 1983. A Collection 9f Readings and Notes for SYS-410. MSU. Department of Electrical Engineering and Systems Science. Mcdowell, R.E. 1977. Steps Necessary in Effective Planning and Evaluation of the Genetic Improvement of Tropical Livestock in Research 92 Cattle Production in Humid Tropical Countr1es. Colloque de Bouake- Republique de Cote d'ivoire. 12-22 avril 1977. Mongondin, B.; Tacher, G. 1979. Les Sous Produits AgroIndustriels Utilisables dans 1'Alimentation animale 22 Senegal. IEMVT, Maison Alfort. 1979. Munzinger, P. 1982.L§ Traction Animale en Afrique . GTZ. Eschborn. Nicholson. M.J.L. 1984. Pastoralism and Milk Production. ILCA Bulletin No 20 october 1984, p. 23-28 Nourrissat,P.1965. La Traction Bovine. CRA/Bambey, Senegal. Panayotou, T.; Tokrisma, R. Microeconomics of Rural Livestock: the Case of Buffalo and Cattle in Thailand in Fine, C.J. and Lattimore, R.G. ed., Livestock in Asia. Issues and Policies. 158 Para, R.; Escobar, A. 1984. Use of Fibrous Agricultural Residues (FAR) in Ruminant Feeding in Latin America. in Guidelines for Research on the Better Utilization 9f Agro-industrial By-Products in Animal Feeding in Developing Countries. FAQ/ILCA Expert Consultation —_ 5- 9 March 1984. Pearce, G.R. 1982. Plant Cell Wall and the Effect of Pretreatment on the Digestibility of Fibrous Residues. FAO Animal Production and Health Paper No. 32 p. 19-27 Peleton, H. L'Enbouche de Bouivillons Ndama en Zone A.V.B. Colloque de Bouake Republique de Cote d'Ivoire, 18-22 Avril 1977. Pratt, D.J. 1984. Ecology and Livestock. in Simpson, J.K. and Evangelou, P. ed., Livestock Development 12 Subsharan Africa. Preston, T.R. 1982. A Strategy for the Efficient Utilization of Crop Residues and Agro-Industrial By-Products in Animal Production Based on their Nutritional Constraints. FAO Animal Production and Health paper No 32 p.29 . 1984. Validity of Feeding Standards and Development of Feeding Systems Based on Crop Residues and Agro- Industrial By-Products in Guidelines for Research on the Better Utilization of Crop Residues and Agr- Industrual By-Products 1n Animal Feeding 1n Developing Countries Proceedings of FAO/ILCA Expert Consultation 5- 9 march 1984. Riviere, R. 1977. Manuel d'Alimentation des Ruminants domestiques 3g milieu Tropical. IEMVT. 1977 Roush, J.L. Ndiaye, A.L. Rasmussen, L. Boyle, P. 1982. Evaluation Report. SODESP Livestock Project (685-0224). Senegal. Sandford, S. 1982. Management of Pastoral Development in the Third World. published in Association with the Overseas Development Institute, London John Wiley et Sons. 1982. Sargant, M.W. Litche, J.A. Halton, P.J. Bloom, R. 1981. An Assessment of Animal Traction in Francophone West Africa. MSU Department of Agricultural Economics. Working Paper No. 21. Shannon, R.E. 1975. Systems Simulation: The Art and Science. Prentice-Ha11,Inc. Englewood C1iffs,N.J. 1975. 159 Simpson, J.R. 1984. Problems and Constraints, Goals and Policy: Conflict Resolution in Deve10pment of Subsaharan Africa's Livestock Industry in Simpson, J.R and Evangelou, P. ed., Livestock Development 1g Subsaharan Africa. SOMIVAC. 1978.P1an Directeur g3 Development Rural Pour la Casamance. Avant-Projet. Tome II livre I. Le Potentiel Agronomique de la Region. Ministere du Development Rural. Senegal. .1980. Rapport sur les Resultats des Enquetes Effectuees au Niveau du Projet Rural du Departement de Sedhiou (Campagne 1978-79 et 1979-80 Republique du Senegal. Ministere du Development Rural. . 1985. Presentation du System Permanent de collecte des Stastistiques Agricoles. Bureau D'Evaluation et d'Etude des Projets. SODEFITEX/SONED. 1980.Projet g3 Development Rural 33 Senegal Oriental 3E en Haute Casamance. Volume 3; Amenagement des Perimetres Irriguees 35 deg EEE Fonds. Republique du Senegal. Starkey, P. 1981. Farming with Work Oxen in Siera Leone. Ministry of Agriculture and Forestry: Njala University College. Stryker, D.J. 1984. Land Use Development in the Pastoral Zone of West Africa. in Simpson and Evangelou , ed., Livestock Development 13 SubSaharan Africa 1984. Sullivan, G.M. 1984. Impact of Government Policies on The Performance of the Livestock-Meat Subsector in Simpson, J.K and Evangelou, P. ed., Livestock Development 13 SubSaharan Africa. Sumberg, J.E. 1983. Leuca-Fences: Living Fence for Sheep Using Leucena Leucocephala. World Animal Review No 47 p.49. 1984. . 1984. La Culture en Alley dans la zone Humide: Une association de 1'Agriculture et de l'elevage. Bulletin du CIPEA no. 18 Avril 1984. . Cassady, K. 1984. Sheep and Goats in Humid West Africa in Sumberg, J.E and Cassady, K. ed., Sheep and Goats £9 Humid west Africa. Proceedings of the Workshop on Small Ruminant Production Systems in the Humid Zones of West Africa Held in Ibadan , Nigeria , 23-26 January 1984 160 Sundstol, F. 1984. Procedures for research into the Treatment of Crop Residues and Agroindustrial By-products in Developing Countries. in Guidelines for Research 92 Egg Better Utilization of Cro residues and AgroIndustrial By-Products in Animal Feeding 12 developing Countries. FAO7iLCA Expert Consultation , 5—9 march 1984. Topps, J.H. 1972. Urea or Biuret Supplements to low protein Grazing in Africa. World Animal Review NO. 3. Toure, S.M. 1977. La Trypanotolerence, Revue de Connaissances. Rev. Elev. Med.vet. pays trop.,1977, 30(2) 1567-174 Trail, J.C.M. Wissocq,Y. 1981. Trypanotolerent Cattle in West and Central Africa. in ILCA Working Document 22. Trverse, S. Tradition et Modernisation des Techniques Rizicoles en asse Casamance. Agronomie Tropical XXX E9 1 (1975) pp 28—34 Upton,M. 1984. Models of Improved Production Systems for Small Ruminants . in Sumberg, J.E. and Cassady, K. ed., Sheep and Goats lg Humid West Africa. Wilson, T. 1984. Goat and sheep in the Traditionnal Livestock Production systems Semi—Arid and Northern Africa: their Importance, productivity and Constraints on production. in Simpson and evangelou ed., Livestock Developmentin Subsaharan africa. 1984. Williamson,G. Payne, W.J.A. 1978.Ag Introduction E2 Animal Husbandry 13 the Tropicspy Third Edition. Trop1ca1 Agricultural series. Longman.