S. Tessema, E.E. Emojong and M. Maluti
National Dryland Farming Research Station
P.O. Box 340, Machakos, Kenya
P.N. de Leeuw
International Livestock Centre for Africa
P.O. Box 46847, Nairobi, Kenya
Introduction
Research background
Subsystems simulation study
On-farm research
References
Abstract
The traditional methods used in livestock and pasture research usually involve component approaches such as grazing management and stocking rate studies; dry-matter yields and nutritional value, growth rates and milk production measurements. However, because of the complementary nature of the relationship between these components, an integrated research effort is a mandatory requisite if relevant and applicable innovations are to be generated. This is particularly true for small-scale farms under mixed farming situations.
The livestock and pasture research strategy at the National Dryland Farming Research Station, Katumani adopted the farming systems approach with four hierarchical levels: base-line study of the traditional system; experimentation; subsystem simulation study and on-farm research. This paper discusses the details of the latter two levels including methodologies used and preliminary observations made. The technical innovations being tested in the sub-system simulation study and the on-farm research are designed to encompass a range of alternative technologies that could increase output and improve livestock productivity in small farm systems of semi-arid areas.
Research in livestock and pasture production at the National Dryland Farming Research Station (NDFRS) Katumani, is principally aimed at increasing feed resources of smallholder agricultural systems of semi-arid areas and developing technologies and management systems that would raise livestock production from the subsistence level to a more commercial or a semi-specialised enterprise system. The research programme is carried out at four hierarchical levels: 1) base-line studies of the traditional livestock production system, 2) experimentation, 3) subsystem simulation study and 4) on-farm research.
The base-line study of the traditional farming system included comprehensive farm surveys in different parts of the semi-arid zone in Eastern Kenya in order to describe the existing farming systems and identify the critical constraints faced by farmers of the region (Rukandema et al., 1984; Tessema et al., 1985). These studies showed that traditional farmers of the dryland areas in Eastern Province avert risks by mixed cropping and livestock raising with the aim of achieving self sufficiency in food production. The land tenure in these semi-arid areas is a freehold system and farm sizes range from 1 to 20 hectares with a mean of 7.5 ha, out of which 2 to 3 hectares are cropped and 46 ha are left as natural rangelands. In the cropping subsystem, the principal crop grown is maize, usually intercropped with beans, pigeon peas and cowpeas. In the livestock subsystem, cattle, goats and sheep are kept by almost all farms, usually in numbers that exceed the carrying capacities of the farms. This has resulted in poor livestock productivity (e.g. in cattle, 50% calving rate; 18 months calving interval, 65 kg weaning weight and 450 kg milked yield per year per lactation). The large number of livestock kept is related to the threat of recurrent food shortages due to crop failures on account of erratic and unreliable rainfall patterns. Thus, livestock are the only means of hedging against such crop failures. They serve as savings and to maintain cash liquidity. Their functions as sources of draught power, manure, meat and milk are secondary. Given these circumstances, however, these small-scale farmers attempt to operate rationally and optimally. Given appropriate technologies and the means to implement them, these farmers are quick to change and do raise their productivity considerably.
Experimental research has been pursued on a number of areas. Studies of the natural pasture through simple techniques such as selective bush clearing, burning and reseeding have shown that with a bimodal pattern of rainfall, reasonably high dry-matter accumulation of herbage can be achieved allowing higher stocking rates (2-3 ha/LU)1 than were normally accepted (6 ha/LU) for the dryland areas (Tessema and Emojong, 1984a). Improvement of crop residues utilisation has shown good possibilities to solve dry season feeding problems. The quality of crop residues was improved through physical treatment and supplementation with high protein Leucaena and pigeon pea leaves (Tessema and Emojong, 1984b). The economics of milk production was found to be profitable through feeding cultivated fodders such as Napier grass (Pennisetum purpureum) and Bana, (Pennisetum purpureum x Pennisetum typhoides) to crossbred dairy cows (Thairu and Tessema, 1987). Draught Zebu oxen, when properly managed and adequately supplemented, provide sufficient draught power to break the soil and provide good filth for planting crops before the rains thus making it possible to prepare more land for food and feed production (Tessema and Emojong, 1984c). Such findings were used for developing a set of improved practices that can be brought together and tested as a package.
1 1 LU = 250 kg weight
In the past, research efforts to improve the productivity of small-scale farms have had little success because of the failure to consider an integrated approach as a research strategy. An integrated research approach is even more important for the livestock system than it is for crops. Improved technological alternatives in crop production such as new crop varieties, new rates of seeding and different planting dates may each be effectively adopted individually without substantially requiring accompanying technologies or affecting other activities on the farm. Innovations in livestock production, however, may not be considered individually or in isolation without due consideration of their effects on other activities of the farm. For example, the adoption of cultivated pasture and fodder crops depends upon the extent to which the activity is conflicting or competing with food crop farming. Cultivated fodder crops could only be economically utilised by livestock that are highly productive and with good genetic potential for such economic traits as milk production. As the genetic potential of the indigenous zebu cattle for milk production is rather low, the use of cultivated fodder crops for increased milk production would at least require the introduction of crossbred dairy cows in the system. Furthermore, for increased milk production from intensively managed pastures to be fully attained, an adequate water supply, proper animal health care and disease control, appropriate housing and shelter are required. Thus, a single innovation, such as the introduction of cultivated fodder crops in the farming system entails the adoption of a multitude of other activities. Therefore, while a step-wise improvement strategy is possible in crop improvement programmes, the livestock production subsystem must have the 'critical mix' of technological alternatives that need to be implemented and studied simultaneously in order to determine relationships, interactions, competitiveness and/or complementarily of innovations.
At NDFRS in Katumani this is being accomplished at two hierarchical levels in a farming systems framework: subsystem simulation study and on-farm research.
In this study the most promising technological innovations emanating from component research in livestock and pasture production are brought together and tested under simulated farm conditions at the experiment station, using the average resources available to a typical farmer in terms of land labour and capital. The objectives of the study are the following:
1. to identify real constraints of the improved livestock production system when compared with the traditional system;2. to evaluate inputs, outputs and economics of livestock production under specific farming conditions;
3. to formulate new and economically sound technological packages;
4. to determine the competition and complementarily in resource use between the livestock and cropping enterprises.
Design of the Study
An area of land comparable to the average-sized farm (7.5 hectares) was demarcated at the station. Leaving 2.5 hectares for food crop production, the remaining 5 hectares were put under the following improved practices of livestock production:
a) improvement of the natural pasture through selective bush clearing and burning and reseeding with Stylosanthes scabra. The purpose is to improve the quality and quantity of dry-matter production and increase carrying capacity;b) improved utilisation of crop residues for dry season feeding through better collection and storage, treatment and supplementation with Leucaena leucocephala and pigeon peas;
c) planting of cultivated leys (Rhodes grass and Stylosanthes mixture) and fodder crops (Bana/Bajra1/Napier);
d) appropriate feeding of draught oxen during the dry season;
e) the conservation of excess forage as hay and silage;
f) provision of an appropriate stock shed and other facilities for better feeding, water provision and manure collection;
g) maintenance of a systematic disease prevention and control schedule;
h) adjustment of livestock numbers to the available feed resource;
i) introduction of crossbred (Zebu x Friesian) dairy cows with high milk production potential.
1 Pennisetum typhoides
The central theme of the package is to increase feed resources and improve the livestock feeding system by intensifying land use and by integrating feed resources so that continuity of feed supply is assured throughout the year. Livestock production is intensified through the introduction of dairying for commercial milk production based on crossbred cows maintained on improved natural pasture, cultivated fodder crops and supplemented with home grown and conserved foodstuffs.
From this subsystem study, data have been collected over a three-year period on pasture and animal production, including productivity of natural and improved pasture, milk production, calf, lamb and kid growth, weight gains and losses in mature stock, disease-occurrence and mortality and livestock offtake. Along with livestock production operation all input data are also recorded: i.e. labour needed for various activities and cost prices of all items used.
Inputs and Investments
The initial and annual costs are given in Table 1 and included the following activities:
Bush clearing
The five hectares of bushy grazing land was selectively cleared. Reseeding with Stylo and Rhodes grass using minimum cultivation methods is still under study.
Table 1. Estimated cost of inputs.
|
Activities |
Labour min man days |
Initial cost |
Annual |
|
|
5.0 ha of bush clearing |
60 |
1500 |
? |
|
|
1.0 ha of fodder crops |
||||
|
|
- planting |
25 |
625 |
- |
|
|
- weeding (2X) |
50 |
1250 |
625 |
|
0.5 ha of fodder legumes |
||||
|
|
- planting |
10 |
300 |
- |
|
|
- weeding |
10 |
300 |
300 |
|
0.5 ha of grass/legume ley |
||||
|
|
- seed |
- |
100 |
- |
|
|
- planting |
10 |
300 |
- |
|
|
- weeding |
10 |
300 |
300 |
|
2.0 ha crop residues collection, transport and storage |
20 |
600 |
600 |
|
|
Ensilage of 5 tonnes |
||||
|
|
- pit digging |
30 |
900 |
- |
|
|
- filling |
140 |
4200 |
4200 |
|
Hay making |
|
|
|
|
|
|
- baling per tonne |
20 |
600 |
|
Establishment of grass/legume leys
The grass/legume ley is composed of Rhodes grass and stylo and requires one thorough weeding each season. This forage is grazed or cut and conserved as hay for dry season feeding.
Establishment of fodder grasses (Bane Napier)
The established plot is expected to have a life span of four to five years and would require one weeding each season (short and long rains). The fodder is zero-grazed during the growing season and any excess is ensiled.
Establishment of fodder legume (Leucaena)
The established plot is expected to have a long life span requiring one thorough weeding each season. The fodder is cut and fed to zero-grazed animals both during the growing season and the dry season.
Utilisation of crop residues
Crop residues from maize, pigeon pea and sorghum are collected properly each season and used during the dry season. Various techniques are used to improve the efficiency of their utilisation, i.e. chopping; treating with urea and supplementation with legumes such as Leucaena or pigeon pea leaves and stems.
Forage conservation
Silage-making
The forage material used to make silage in this study was obtained from a one-hectare Bana Napier field which is not part of the five hectares assigned to the subsystem unit. To adopt this silage-making technology, it would be necessary to increase the fodder area of the unit (Table 1) from one to two hectares in order to get the bulk required.
Circular pit silos, 3 deep and 2 in diameter containing 5 tonnes of chopped forage material packed properly by continuous manual filling and trampling have been made. The inside of the silo is lined with a polythene sheeting to prevent contamination with soil and help in excluding air.
Although this method of silage-making was part of the package tested in the subsystem study, silage-making at small-scale farms is recommended only as a communal effort as it may not be possible for a farm to have the necessary bulk of forage crops or the necessary family labour force to do the work, nor would it be economical to do it by hired labour. If made on a communal basis, it is assumed that it will also be shared equally among those who contributed in the effort.
Hay-making
This was done using a simple wooden box hand-baler developed at the station and costing KShs. 260/=1. Three men can cut and make 10 bales of hay per day each weighing 15 kg on the average.
1 One US dollar = 16 KShs.
The following are the purchased investments included in the package:
Ox-cartA locally made wooden body ox-cart made from an old car axle and wheels and tyres filled with saw dust cost KShs. 2000/-. These carts are made by a small firm called Ndume in Gilgil near Nakuru town.
Hand chopper - Livestock shed
A hand-operated chopper for cutting fodder crops and crop residues cost Kshs. 4000/-. This machine is available in Nairobi. To keep animals in reasonable comfort, to eliminate unhygienic situations and to feed and manage them properly and separately, a shed with partitions for dairy cows, oxen, calves and small stock was constructed. Locally available materials were used in the construction of the shed. If corrugated iron sheet roofing is used, it can also serve to collect rain water for use by animals. The bulk of water requirement is, however, hauled from a river source using ox-cart and drums. The livestock shed is estimated to cost KShs. 4000/-.
Number and types of livestock
Based on the feed resources developed on the five hectares of land as given in Table 2, the following livestock were kept in productive state throughout the year:
- three crossbred dairy cows (Zebu x Friesian),
- one replacement heifer, and two oxen (indigenous),
- seven sheep and their followers,
- seven goats and their followers.
Table 2. Estimated total fodder resources.
|
Type of resource |
Yield t DM ha1 |
Total t DM |
|
5 ha improved rangelands |
2.3 |
11.5 |
|
0.5 ha grass/legume ley |
6.0 |
3.0 |
|
1.0 ha planted fodder |
7.0 |
7.0 |
|
0.5 ha Leucaena |
4.0 |
2.0 |
|
2.0 ha crop residues |
3.4 |
3.4 |
|
Total resources |
|
26.9 |
1 Total yield per annum from two growing seasons.
An in-calf crossbred heifer is generally estimated at approximately KShs. 4500/- when directly purchased. But the crossbred cows can be obtained through artificial insemination of indigenous cows, thus eliminating this high investment cost.
A stocking rate of approximately one livestock unit (250 kg) per hectare of farm holding (7.5 hectares) is achieved while the mixture of livestock kept will meet the farmer's needs for draught power, milk production (for sale and home consumption) and small stock for meat and periodic sale to meet large payments such as school fees, etc.
Labour required
The labour required to carry out the activities of this livestock subsystem is shown in Table 3. The sum total of man-hours required per day amounts to 11.8 hours. Out of these, seven hours are needed for herding, which under real farm situations can be undertaken by 7 to 10-year-old children who go to school only part of the day. Therefore, these hours need not be considered in terms of adult labour force. In the subsystem unit, grazing areas were fenced, thus the labour requirement for grazing was only imputed. The real requirements of labour hours for the livestock activities in terms of adult equivalents were therefore 4.8 hours.
Table 3. Labour requirement for livestock work.
|
Activities |
hrs/day |
|
Milking: 3 cows at 0.3 hrs/cow |
1.0 |
|
Feeding: cutting, transporting, chopping, feeding |
2.7 |
|
Cleaning barn and transporting manure |
0.6 |
|
Watering, hauling water (1 km) in drums by oxcart |
0.5 |
|
Herding of cattle and smallstock |
7.0 |
|
Total |
11.8 |
Other expenditures
These include:
- drugs for all animals(= KShs. 980.00),
- mineral and protein supplements (= KShs. 544.00),
- dipping costs(= KShs. 760.00).
Outputs
Fodder resources
The total amount of foodstuffs produced is shown in Table 2 and indicated that 27 tonnes of dry matter were potentially available in an average year.
Milk production
The total average lactation yield from three cows over two lactations was 1870 kg per cow without the calf at foot. This was achieved under semi-zero grazing where the cows were grazed on natural pasture during the day and, in the evenings, were offered chopped Napier grass (ad libitum) during the growing season and treated maize stover mixed with green Leucaena (at 30% of DM intake) during the dry season.
When 500 liters of the milk produced is fed to the calf over a six-months period, 1370 kg of milk per cow is left for sale or home consumption. This is three times the amount produced by the average cow under traditional management. At a current value of milk in the local market of KShs. 3/- per liter, the value of milk offtake from three cows amounted to KShs. 12,330 over a 14-month period.
Livestock growth and offtake
Annual average weight changes of all livestock over a three-year period is shown in Table 4 below.
Table 4. Annual liveweight gains of different classes of stock.
|
Class of animals |
ADG (g) |
Class of animal |
ADG (g) |
|
Cows < 3 years old |
264 |
Oxen 5-6 years old |
0 |
|
Heifers < 2 years old |
493 |
Shoats, young 0-5 months |
80 |
|
Calves 0-12 months |
319 |
Shoats, adults > 5 months |
30 |
The annual offtakes consisted of three lambs weighing an average of 28 kg, with a value of KShs. 672/- (at 8 KShs/kg) and four weaner goats weighing 30 kg with a value of KShs. 1080 (at 9 KShs. liveweight).
Draught power output
The two oxen provided adequate traction power to plough two to three hectares of land each season, before rains started. They also pulled an ox-cart whenever required to draw water or to transport fodder. The opportunity cost of renting a team of oxen to plough one hectare of land in the area is KShs. 550/-.
Manure
A total of 30 tonnes of manure (a compost made of 80% dung and 20% soil and other plant material) was collected yearly from the herd. The monetary value of this manure, according to current sale price is KShs. 2400 (at KShs. 80/- per tonne).
The technological improvements tested in this study certainly required high inputs. But the outputs are also correspondingly high. It is, however, felt that more data would be required to make an appropriate assessment of the profitability of the enterprise. Calving, kidding and lambing intervals; death rates; the rate of decline in yield of established fodder fields; value of replacement heifers sold or kept etc. are some of the vital data that are still required to make a comprehensive assessment. It is estimated that a period of five years would be adequate to make such-an assessment.
The existence of strong interactions between crops and livestock in these small-farm systems is well recognised. While the crop subsystem is being studied in a separate simulation programme, the on-farm research programme discussed below is, however, carried out with an integrated multi-disciplinary approach.
The most promising results from the multi-disciplinary on-station research were used to formulate technological improvements that would appear to increase the productivity of livestock on small-scale farms in semi-arid areas. Preliminary results of the profitability of these improvements when tested as a package under simulated farming systems also appeared to be very promising. The on-farm research programme was therefore designed to test the validity of these technological innovations under actual farm conditions.
There are many factors which determine how rapidly a practice will be adopted by small-scale farmers. These include the costs and risks involved, the ignoring of potential returns, the complexity of the practice, the time it takes for better results, the computability of the practice relative to cultural customs and established pattern of agriculture. Farmers are expected to respond differently to a given economic opportunity depending upon their subjective evaluation of the innovation. The smaller the potential increase in income from a particular recommendation, the more varied the response of different farmers is likely to be. For example, an increase of 10% to 20% in milk yield or in growth rate above the common experience can hardly be distinguished from changes in yields resulting from the usual seasonal or annual fluctuations. Incentives which in the initial stages bring about increases of 100% or more have a much better chance of adoption. Thus, the concept of the on-farm research programme has necessitated a somewhat radical approach, namely the introduction of a commercialised livestock enterprise in small-farm systems.
Smallholder dairying based on crossbred cows maintained almost entirely on well managed forages and pasture grasses and supplemented with home grown and conserved foodstuffs would indeed seem to be-a feasible innovation worth extensive on-farm testing. Except for the problems of feed resources and diseases that may limit level of production, the economic aspects of dairying in low potential areas would not be too different from that of the high potential regions when expressed in terms of returns to labour and capital investment. However, a successful dairying enterprise needs to adapt to the cyclical nature of the environment. Therefore a year--round feed supply of sufficient quality needs to be guaranteed.
The feeding of dairy cattle for improved productivity is best achieved by the integration of herbage from natural rangeland, sown pasture, planted fodders and crop residues. Without the combined use of these different feed resources, increased and sustained milk productivity cannot be achieved. In order to attain this objective a change must come about from the present system of keeping livestock as a separate enterprise to that of fully integrating them in the farming system. As soon as farmers accept that grasses and forages need the same care and management of food crops, a transition from livestock keeping for subsistence to that for productive purposes will be effected.
Design of On-Farm Research
The technological innovations are introduced step-by-step into a number of farms representing the main ecological areas, farming systems and size classes. An optimum plan is made to each individual farm, incorporating as many of the recommended technologies as possible. The recommended improvements in pasture and livestock production in particular are those that require the minimum capital input and the maximum use of family labour force. Since the farms selected are being considered as experimental units, a protective mechanism has been built into the on-farm research programme whereby the inputs are provided to the farmer on the basis of loans. The participating farmers are required to repay these loans only after it has been proven that the innovations have contributed substantially to the net income of each farm. Thus, participating farmers are not subjected to unnecessary risks and losses due to the research exercise being carried out on their farms.
On the basis of rainfall and existing farming systems, the study area is divided into three domains encompassing parts of Machakos and Kitui districts. Domains have been identified:
1. Most favourable, 700-800 mm rainfall;
2. Medium favourable, 600-700 mm rainfall;
3. Least favourable, 500-600 mm rainfall.
Farms are grouped according to farm size in small (5 ha), medium (5 to 10 ha) and large (>10 ha).
Two farms from each domain and each class are selected, making a total of 18 farms. Another group of 18 farms within the same domains are selected to serve as controls where only comparative data are collected.
Each farm is surveyed and appropriate technologies are developed for each domain's classes. Ex ante analyses are made by computer modelling and simulation to assess potential applicability, input requirements and constraints and outputs. The technical coefficients used for these analyses were generated from the subsystem simulation study discussed above.
The technological packages being tested on the farms are the same four enumerated under the subsystem simulation study. Improvement packages vary according to each domain e.g. dairy cattle for most favourable rainfall zones and dairy goats for drier zones. Farm plans are discussed with farmers and modified to reflect their preferences and willingness to test innovations. Budgeting and sensitivity analysis techniques are used to determine optimal farm plans.
To date, the on-farm research programme has been implemented only on six participating farms. For the six farms, data have been collected on pasture and animal production including production of natural and improved pasture, milk production; calf, lamb and kid growth; weight gains and losses in mature stock, disease occurrences and mortality. Along with livestock production operations, selected input/output data on all farm enterprises are monitored continually to quantify labour, traction and material inputs as well as the returns for various enterprises, and to assess the pressure on farm resources and the levels of risk of different enterprise combinations faced by farmers. The data will be analysed by linear programming, regression equations and variance analysis to evaluate results and identify variables. Cost-benefit analyses will be made to test economic feasibility.
Achievements of applicable results from livestock and pasture research are lengthy and difficult processes. While substantial data are available from the subsystem simulation study, collected over the past three years, data from the six farms on the on-farm research programme are scanty. Preliminary observations of the research approach used, however, show that the researcher is more knowledgeable of the real constraints of the production system and therefore can design more relevant and effective programmes. It seems also that the desired end results of the research efforts can be arrived at in a reasonably shorter period of time.
Rukandema, M. 1984. Farming systems of semi-arid Kenya, a comparison. East African Agric. and For. Journal 44:422-35.
Tessema, S. and Emojong, E.E. 1984a. The influence of stocking rates and grazing management on liveweight changes in cattle, sheep and goats grazing natural pastures. East African Agric. and For. Journal. Special Issue. Vol. 44.
Tessema, S. and Emojong, E.E. 1984b. Utilization of maize and sorghum stores by sheep and goats after feed quality improvement by various treatments and supplements. East African Agric. and For. Journal. Special Issue. Vol. 44.
Tessema, S. and Emojong, E.E. 1984c. Feeding of draft oxen for improved and more efficient power. East African Agric. and For. Journal. Special Issue. Vol. 44.
Tessema, S., Emojong, E.E., Wandera, F.P. and Nderito, M. 1985. Features of traditional farming systems as they affect livestock production. A case study of 18 small-scale farms in the dryland areas of the Eastern Province of Kenya (Machakos, Kitui and lower parts of Embu districts). Dryland Farming Research and Development Project, KEN/8/011, Machakos. Document No. 6.
Thairu, D.M. and Tessema, S. 1987. Research on animal feed resources: Medium potential areas. In: J.A. Kategile, A.N. Said and B.H. Dzowela (eds), Animal feed resources for small-scale livestock producers. Proceedings of the Second PANESA Workshop on Animal Feed Resources for Small-scale Livestock Producers held in Nairobi, Kenya, 11-15 November 1985. IDRC-MR 165e. International Development Research Centre, Ottawa, Canada. pp. 125-148.
