The overall methodology is schematically presented in Figure 2.1, and comprises the following activities:
formulation and selection of crop, livestock and fuelwood land utilization types (LUTs);
determination of ecological requirements of crop, livestock and fuelwood land utilization types;
compilation of climatic resources inventory;
compilation of soil and landform resources inventory;
compilation of land use inventory (including socio-economic aspects);
compilation of a 1:1 million scale computerized land resources data base (agro-ecological cells) of each district in a Geographic Information System (GIS);
determination of land under other uses;
determination of land available for productivity assessment of crop, livestock and fuelwood;
formulation of crop productivity model and assessment of land productivity potential for crop production;
formulation of livestock productivity model and assessment of land productivity potential for pasture and livestock production;
formulation of fuelwood productivity model and assessment of land productivity potential for fuelwood production;
assessments of reference land productivity potentials (from crop livestock and fuelwood) for each agro-ecological cell;
Schematic presentation of methodology
assessment of potential population supporting capacity, taking into account nutritional requirements;
estimation of inputs requirements, using an input technology matrix including soil conservation; and
addressing a range of policy issues for development planning, based on a set of scenarios embodying present and future populations, food and agricultural demands, and socio-economic development needs, opportunities and constraints.
The above 15 activities represent four groups of compound activities namely:
formulation of land utilization types and their ecological requirements;
compilation of national land resources and land use data base;
assessments of land productivity potentials; and
development planning involving assessments of potential population supporting capacities and input requirements to address policy issues.
Activities related to the formulation of land utilization types and their ecological requirements overlapped with those activities concerned with the compilation of the land resources data base. This was necessary to ensure that those land qualities that determine productivity characterized in the land inventory and that land use requirements can be formulated in terms of such land qualities.
Subsequently, the productivity models of crops, livestock and fuelwood were applied to the land resources data base to estimate land productivity potentials of alternative kinds of land uses (land utilization types). These land productivity potentials, in turn, form a basis for quantifying potential population supporting capacities and input requirements at several levels of geographical and administrative aggregation (e.g. sub-district, district, province, nation). When set against present and projected population distribution, food and agricultural demands, socio-economic development needs, opportunities and constraints, such assessments of land and population potentials provide a sound and coherent basis for national development planning.
In the crop productivity model (Chapter 5), a total of 25 crop species are considered. They are differentiated into 64 crop types to account for differences in ecotype adaptation, crop phenology and growth cycles within each species. The crops comprise:
|7||cereal food grain crops (wheat, barley, oat, rice, maize, pearl millet, sorghum)|
|6||leguminous food crops (cowpea, green gram, groundnut, phaseolus bean, pigeonpea, soybean)|
|3||root and tuber crops (white potato, sweet potato, cassava)|
|9||‘cash’ crops (banana, oil palm, sugarcane, coffee, cotton, pineapple, pyrethrum, sisal, tea).|
Coffee, cotton, pineapple, pyrethrum, sisal and tea are considered in the model to take account of the reported land area occupied by these crops as inventoried in the land use data base. The remaining 19 crops (58 crop types) are considered at three levels of inputs circumstances. In all, therefore, the crop productivity model considers 180 crop land utilization types (LUT).
In the livestock productivity model (Chapter 6), a total of 30 pasture and fodder species of grasses and legumes, and six livestock types have been considered. The grass and fodder species comprise:
|18||pasture grass species|
|4||fodder grass species|
|8||pasture and fodder legumes.|
The above grass and legume species are considered at three levels of inputs circumstances, and the livestock model has a provision for considering 90 pasture and fodder LUTs. However, at this stage of model development and application, productivity assessments have been made for groups of pasture and fodder grasses and legumes at three levels of inputs.
The six livestock types considered in the livestock productivity model are: cattle, goat, sheep, camel, poultry and pig, of which the first four are considered under non-pastoral as well as pastoral systems. Poultry and pig are considered without explicitly defining the production systems. The non-pastoral and pastoral systems comprise:
|Non-pastoral:||3||Dairy and meat cattle systems|
|3||Dairy and meat goat systems|
|3||Meat and wool sheep systems.|
|Pastoral:||3||Cattle herds (nomadic distant, nomadic with market access, semi-nomadic)|
|2||Goat herds (nomadic distant, semi-nomadic)|
|2||Sheep herds (nomadic distant, semi-nomadic)|
|1||Camel herd (nomadic).|
The livestock model makes a provision for a total of 27 livestock LUTs so that the above livestock systems, including those of poultry and pig, can be assessed at three levels of inputs circumstances.
In the fuelwood productivity model (Chapter 7), 31 species of fuelwood are considered. They comprise:
15 species with nitrogen fixing ability
16 species without nitrogen fixing ability.
The model considers the 31 fuelwood species at three levels of inputs circumstances, so that the fuelwood model has a provision for assessing 93 fuelwood LUTs.
The three levels of inputs circumstances used in the models are: low inputs, intermediate inputs and high inputs as detailed in Chapters 5, 6 and 7. The low level circumstance assumes low capital input and subsistence management practices, the use of ‘indigenous’ cultivars of crops and breeds of animals, hand labour only, no use of fertilizer or biocides, no large-scale conservation measures, and cultivation in rotation with bush fallow to maintain soil fertility. It can be compared to traditional systems of bush fallow rotations. The intermediate level circumstance assumes medium capital input, partly subsistence and partly commercial management practices, the use of improved cultivars of crops and breeds of animals (including crossbred animals), use of improved hand tools and draught implements, some mechanization, some use of fertilizer and biocides, some soil conservation measures, and cultivation in rotation with sown grass fallow. The high level circumstance assumes capital intensive management practices, full use of most productive adapted cultivars of crops and breeds of animals (including exotic breeds), complete mechanization, optimum use of farm chemicals, and full soil conservation measures.
Determination of the climatic and edaphic (soil) requirements of crop, livestock and fuelwood land utilization types used in the assessment has been a major activity. Previous attempts to quantify climatic requirements of crops (including pasture and fuelwood) have not adequately recognized the importance of the time courses of temperature and soil moisture balance (including seasonal and between year variations) in relation to crop growth(photosynthesis), development (phenology) and production (yield). Adequate emphasis has been placed on these two sets of parameters (temperature and soil moisture regimes) in this assessment.
Of similar significance is the nature of the photosynthesis response to temperature and radiation, which determines crop yield and land productivity when the phenological requirements are met during the period when soil moisture is available for growth. Accordingly, an inventory of crop, pasture and fuelwood species was prepared, based on their climatic requirements for both photosynthesis and phenology. Four main climatic adaptability groups of crops, pasture grasses and legumes, and fuelwood species are recognized in the assessment, as detailed in Chapters 5,6 and 7. This inventory gives, among other information, ranges of temperature requirements for different aspects of growth and development. These are subsequently matched to existing thermal climatic conditions.
Once the photosynthetic and phenological thermal requirements are met, the agronomic (or silvicultural) yield potential of a crop, under constraint-free conditions, is governed by the number of days (or years) to maturity. This, in turn, is determined by the length and quality of growing period (including its year-to-year variation). Constraint-free yields were calculated for all crop, livestock and fuelwood LUTs for each length of growing period. The results were used as the basis of the climatic suitability assessment, as described in Chapters 5, 6 and 7.
Soil requirements of LUTs were assessed as follows: For each crop, pasture and fuelwood species, available data on soil characteristics considered meaningful for production were listed, e.g. soil depth, texture, salinity, stoniness, etc. For each LUT, each property was then quantitatively subdivided into those for optimum conditions and for acceptable range of conditions. When a property fell outside the defined range, the soil was considered as currently not suitable.
The information on optimal range and minimal or maximum values of soil properties for each LUT formed the basis for subsequent suitability rating of the soil units for production of crops, pasture and fuelwood (Chapters 5, 6 and 7).
The information on climatic and soil requirements of crop, pasture and fuelwood land utilization types, was used as a guide in the compilation of the land resources inventory (Chapter 3).
In the case of climate, temperature and soil moisture availability are key factors in determining the distribution (in both space and time) of rainfed crops. In combination with solar radiation, these climatic factors condition photosynthesis and allow plants to accumulate biomass (and accomplish successive development stages) according to their ecophysiological rates and patterns.
The temperature (heat) attributes were quantified by defining thermal zones. To cater for the differences in temperature requirements of crops, pasture and fuelwood species, nine reference thermal zones have been inventoried based on 2.5 °C interval in daily mean temperature.
The moisture attributes were quantified through the concept of the reference length of growing period, defined as the duration (in days) of the period during which the supply of available soil moisture from precipitation, and from storage in the soil profile (set at a reference 100 mm), is greater than half the potential evapotranspiration.
Lengths of growing periods were computed from historical data sets of some 435 stations, and derived from average data sets of some 1500 stations. With the historical data set, length of growing periods were computed for individual years, and frequency distributions for each mean length were computed for the historical series. Where there were more than one length of growing period per year, the total mean length as well as the individual mean lengths (e.g. two, three) and their frequency distributions were calculated. These computations represented the information on the length of growing period (LGP). Fifteen mean length of growing period zones, at 30 day interval, have been delineated in the climatic inventory of Kenya, as explained in Chapter 3.
To inventory the year-to-year variation in the number of lengths of growing periods per year, a historical profile was compiled showing groups of years each with a different number of growing periods per year. The proportional representation of each group in the total historical series was computed. This information represents the pattern of length of growing period (LGP-Pattern). Twenty-two LGP-Pattern zones have been recognized in the climatic inventory of Kenya.
For each LGP zone delineated, average values of major climatic elements (radiation, day and night temperature, humidity, etc.) were inventoried to characterize the climate during the growing period. These together with the information on the year-to-year variation in the number of length of growing periods per year and in each component length of growing period, formed the basis for subsequent matching and productivity estimation. Details of the climatic resources inventory are given in Chapter 3.
The soil inventory was compiled essentially from the 1:1 million scale Exploratory Soil Map of Kenya, which comprises 390 different soil map units. For each map unit, information on landform, geology/parent material, soil unit (with implied characteristics), slope-gradient, soil texture and soil phases, in terms of description, classes and extents was transferred to form the soil resources inventory of this assessment (Chapter 3).
On completion of the climatic inventory, the three layers (thermal zone, LGP zone and LGP-Pattern zone) were superimposed on the Exploratory Soil Map of Kenya. The different layers of climate and soil information were digitized and the information was converted to a data base of about 575 thousand one millimeter square grid cells, each corresponding to 100 ha. The resultant map output created about 91 000 unique agro-ecological cells of the inventory, whose land attributes, defined by climate, soil and landform, are known and quantified. This information, compiled at the national level by province and district, constitutes the physical land resources data base of Kenya.
Additional six layers of information were also digitized and overlaid on the land resources inventory. These layers contain information on cash crop zones, forest zones, parkland areas, irrigation schemes, tse-tse infestation areas and administrative boundaries (provinces and districts).
The climate, soil and land use inventories make-up the computerized data base for the assessment, and allow any desired geographical and administrative aggregation to be made of inventoried parameters and results.
The assessment of land productivity starts by formulating and selecting crop, livestock and fuelwood land utilization types (shown at the head of the flow chart in Figure 2.1), and their ecological (climate, soil and landform) requirements (ii).
Then, from the agro-ecological cells in the land resources inventory (iii, iv, v, vi), district by district, land used or required for irrigation, cash crops and for non-agricultural purposes (vii) is deducted. The remainder is an inventory of land potentially available for rainfed cultivation, and for productivity assessments (viii).
For each of the agro-ecological cells in this inventory, the next stage is to determine the potential rainfed yield or output of crops, livestock and fuelwood at one or more levels of inputs (ix, x, xi) in order to find out which land utilization types (cropping patterns and rotations, livestock systems, fuelwood land uses) are most productive, stable and sustainable in the unique conditions of the cell. The land productivity potentials can then be calculated (xii), either in a reference manner or within the context of a set of planning scenarios.
The crop, livestock and fuelwood models (Chapters 5, 6 and 7) are all designed to operate on the computerized land resources data base. They permit quantitative land suitability assessments to be made based on growth and yield predictions of each LUT and combinations of LUTs in each agro-ecological cell. All three productivity models include a provision for quantifying soil erosion hazard of each LUT in terms of productivity loss. This is achieved through the soil erosion and productivity model described in Chapter 4. The model also estimates ‘tolerable’ soil loss, and costs of alternative conservation measures.
The crop productivity model (Chapter 5) explicitly formulates options in respect of individual crops, annual cropping patterns and crop rotations, and quantifies their production potentials. The model formulates optimum cropping patterns and output therefrom to meet a reference or given food demand, taking into account desired level of production stability.
The livestock productivity model (Chapter 6) quantifies primary productivity potential which is then converted into secondary production (milk, meat, wool, draught power) for pastoral and non-pastoral herds.
The fuelwood model (Chapter 7) quantifies wood biomass productivity potential in terms of mean annual increments over the rotation age of each fuelwood LUT.
The crop, livestock and fuelwood productivity models are interphased with each other. This allows land productivity to be optimized for a given set of development constraints and demands.
Beyond this, the assessment allows for development planning applications, which involve the calculation of the quantities of edible calories and protein that would be produced by the different crops and livestock (and products from other land uses) from information on the nutritional composition of the products. The crops or crop mixes (including grassland) that can produce the largest or desired quantity and quality of calories and protein in each agro-ecological cell are then selected, and the results from each cell in each climatic zone in each district are added to determine the optimal maximum potential production of calories and protein from each climatic zone in each district, from whole districts and groups of districts, and from whole provinces and country.
Dietary and other constraints such as minimum protein requirements are applied to estimate potential population supporting capacity (xiii) at various desired levels of geographical and administrative aggregation. Similarly, by applying the extended FAO technology matrix for Kenya (including conservation inputs), the associated inputs requirements (xiv) are quantified (Bruinsma, Hrabovszky, Alexandratos and Petri, 1983).
The potential population supporting capacity (in xiii) is computed as potential population density (persons per ha) which is compared with the present and projected population densities, and examined against food and agriculture demands, and socio-economic needs, opportunities and constraints, to address a range of policy issues for development planning. These relate, for example, to food and economic self-sufficiency, areas with surplus potential and areas that are critical, domestic and export trade, infrastructure, services, employment, incomes, revenues, industries, investments and human resources development (xv).