Chapter 3
Land resources

The land resources inventory brings together two layers of information on physical environmental resources (climate and soil) and allows the creation of unique ecological land units (agro-ecological cells) within which soil, landform and climatic conditions are quantified. This information, compiled at the national level by province and district, constitutes the inventory of the physical land resources.

To create a computerized inventory of land resources, the individual climate and soil inventories have been compiled in map form at 1:1 million scale, and digitized.

The climatic resources inventory consists of three seperate thematic layers: thermal zones, length of growing period zones, and pattern of number of length of growing period zones. The Exploratory Soil Map of Kenya (KSS 1982a), forms the soil base, providing information on soils, landform and geology/parent material.

Six additional layers (1:1 million scale) of information have also been digitized and overlaid on the land resources inventory. These layers provide information on cash crop zones, forest zones, parkland areas, irrigation schemes, tse-tse infestation areas, and province and district boundaries.

The individual map layers have been digitized using the Comprehensive Resource Inventory and Evaluation System (CRIES 1983) a GIS developed at Michigan State University. The digitized information derived from the individual map layers has been converted to a data base of 576,072 grid cells. Each cell (one millimeter square) corresponds to 100 ha.

Subsequent to digitizing the individual layers of the land resources inventory, the soil mapping unit composition of each mapping unit and the associated ecological conditions have been incorporated.

The make-up of the national land resources data base is schematically presented in Figure 3.1, and is described in Technical Annex 1. A map of provinces and districts is presented in Figure 3.2, and their extents are given in Table 3.1.

3.1 Climate Resources

Temperature and water are the major climatic factors that govern crop distribution (both in space and time). In combination with solar radiation, these climatic factors condition the net photosynthesis and allow plants to accumulate dry matter (and to accomplish the successive development stages), according to the rates and patterns which are specific to cultivated plants.

FIGURE 3.1
Make-up of tend resources data base

FIGURE 3.2
District map of Kenya

TABLE 3.1
Extents of Districts and Provinces of Kenya

The growing period has been used as a framework for the assessment of climatic resources (FAO 1978–81). It is defined as the period in which temperature and moisture permit crop growth. Prevailing temperature regimes have been inventoried by identification of thermal zones in order to take into account temperature requirements of crops (including pasture and fuelwood species).

The inventory of climatic resources allows:

1. A differentiation of the country into reference thermal zones, reflecting the geographical and ‘seasonal’ distribution of the prevailing temperature regimes.

2. A differentiation of the country into reference length and pattern of growing period zones, reflecting the prevailing moisture regimes including the year-to-year variations.

3. A quantification of potential yields (of crops, livestock and fuelwood) that can be attained under constraint-free conditions.

4. An assessment of various agro-climatic constraints to take into account yield losses likely to occur.

The climatic data bank, growing period analysis and the creation of the climatic resources inventory are described in Technical Annex 1.

3.1.1 Climatic Data Bank

The climatic data¹ bank compiled for the assessment consists of three data sets. Data set 1 (Historical Data, from Jaetzhold and Kutsch 1980) consists of the following information for 437 stations:

-   Average decadal (10-day total) potential evapotranspiration (mm)

-   Historical decadal rainfall (mm) for individual years.

¹ The primary source of the climatic data is the Kenya Meteorological Department.

Data set 2 (Average climatic data, from FAO/AGPC data bank, FAO 1977) consists of 45 stations with mean monthly values for the following 11 climatic parameters:

-   Precipitation (mm)

-   Mean daily temperature (°C)

-   Maximum temperature (°C)

-   Minimum temperature (°C)

-   Day-time temperature (°C)

-   Night-time temperature (°C)

-   Mean water vapour pressure (mbar)

-   Mean wind velocity (m sec^-1)

-   Hours of bright sunshine as a percentage of maximum possible sunshine hours (%)

-   Solar radiation (cal cm^-2 day^-1)

-   Potential evapotranspiration (mm).

Data set 3 (Average climatic data, from Kenya Soil Survey) provides average data on the following for 1489 stations:

-   Annual daily temperature (°C)

-   Annual potential evaporation (mm)

-   Annual potential evapotranspiration (mm)

-   Annual Rainfall (mm)

-   Monthly rainfall (mm)

-   Type of rainfall pattern; monomodal (M), bimodal (B) or trimodal (T).

Extracts of the three data sets are presented in Tables 3.2, 3.3 and 3.4 respectively. The complete climatic data bank is available on diskettes (ASCII).

3.1.2 Growing Period Model

The definitions and model used to quantify the reference length of growing period have been described in Technical Annex 1. The growing period is the time period when moisture supply exceeds half potential evapotranspiration; it includes the time required to evapotranspire up to 100 mm of soil moisture storage². The calculation of the reference growing period is based on a water balance model, comparing rainfall with potential evapotranspiration. The length of growing period (and the number of growing periods and dry periods per year) from a climatic viewpoint alone, and independent of crop, soil and landform, is therefore quantified in a reference manner (Kowal and Kassam 1978; Doorenbos and Kassam 1979; Kassam, van Velthuizen, Higgins, Christoforides, Voortman and Spiers 1982; Brammer, Antoine, Kassam and van Velthuizen 1988).

2 The computer program is able to handle a storage term in the range 0–250 mm.

Two types of growing periods are schematically shown in Figure 3.3. The distinction between ‘normal’ and ‘intermediate’ is useful because in the latter it is unlikely that full water requirements can be met during the rainy season without moisture conservation or a supply from groundwater or irrigation.

Two more growing period types have been identified (Figure 3.3). These are: (a) all year-round humid with rainfall exceeding full potential evapotranspiration throughout the year, and (b) all year-round dry with rainfall not exceeding half potential evapotranspiration throughout the year.

3.1.2.1 Length of Growing Period (LGP)

Mean length of growing period and frequency distribution for each individual group of years have been computed. Where there are 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 distribution are calculated (Figure 3.4).

TABLE 3.2
Extract agroclimatic data bank - Data set 1 - Historical data

TABLE 3.3
Extract agroclimatic data bank - Data set 2 - Average climatic data

TABLE 3.4
Extract agroclimatic data bank - data set 3 - Average data

Stations 8535201 to 8940003 latitude North and stations 9034001 to 9439057 latitude South.

Temp: Mean daily temperature (celcius)

Eo: Potential evaporation (mm)

PET: potential evapotranspiration (mm)

P: Rainfall (mm)

RP type: Rainfall pattern type

M = Monomodal,

B = Bimodal,

T = Trimodal

FIGURE 3.3
Schematic presentation of types of growing periods

FIGURE 3.4
Number of growing periods and dry periods per year

TABLE 3.5
Patterns of growing periods (LGP-patterns) - Historical profiles of occurrence of number of length of growing periods per year

H = 365+ days (i.e. year-round humid)
D = zero days (i.e. year-round dry)

TABLE 3.6
Relationships between mean total dominant and mean total associated lengths of growing period

L1 = Total length of one growing period per year
L2 = Total length of two growing periods per year
L3 = Total length of three growing periods per year
L4 = Total length of four growing periods per year

For a group of years with one length of growing period, the length is designated the code L1, and the dry period is coded Dl (Figure 3.4a). For a group of years with two lengths of growing periods per year, the lengths are coded L21 and L22, and the first length(L21) is followed by the first dry period (D22) and the second length (L22) by the second dry period (D22) (Figure 3.4b) The sum of lengths L2 and L22 is coded L2. For a group of years with three lengths of growing periods per year, the lengths are coded L31, L32 and L33, and there are dry periods in between (D31, D32, D33) (Figure 3.4c). The sum of lengths L31, L32, L33 is coded L3.

3.1.2.2 Pattern of Length of Growing Period (LGP-Pattern)

To inventory the year-to-year variation in the number of lengths of growing periods per year, a historical profile is 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 is computed.

This information represents the pattern of growing period. Twenty two patterns are recognized in the climatic resources inventory. The patterns of number of length of growing period and their composition are presented in Table 3.5.

The pattern of growing period code represents the number of growing periods per year in order of frequency of occurrence, e.g. in the pattern coded as 2-1-3, the numeral 2 represents the number of lengths of growing periods per year (i.e. two)that occur in the majority of the years (i.e. 55 percent) - the dominant length number; the numeral 1 represents number of lengths of growing periods per year (i.e. one) that has the next most commonly occurring frequency (i.e. 25 percent) - the first associated length number; and the numeral 3 represents number of lengths of growing periods per year (i.e. three)that has the smallest occurrence (i.e. 20 percent) - the second associated length number.

For each pattern of growing periods, the mean total length of the dominant number is correlated with the mean total length of the associated numbers. Also, when the mean total length is a summation of more than one mean length, the latter are again correlated with the former. These relationships are presented in Tables 3.6 and 3.7.

TABLE 3.7
Relationship between individual component mean length and mean total length of growing period

L2₁ = First length of two growing periods per year
L3₁ = First length of three growing periods per year
L3₂ = Second length of three growing periods per year
L4₁ = First length of four growing periods per year
L4₂ = Second length of four growing periods per year
L4₃ = Third length of four growing periods per year

In the climatic inventory map of Kenya, only the mean total dominant length has been inventoried on the map. The relationships in Table 3.6 are further presented in terms of length of growing period zones in Technical Annex 1; giving the mean total dominant (mapped) and the corresponding mean total associated (unmapped) lengths of growing periods. Similarly the relationships in Table 3.7 are further presented in terms of length of growing period zones in Technical Annex 1; giving the mean total length of growing period zones and the corresponding individual component lengths of growing periods.

3.1.2.3 Variability of Length of Growing Period

In addition to the frequency distribution mentioned in Section 3.1.2.1, coefficient of variation was calculated to allow a comparison of the variability in the mean length of growing period, and to take into account the likely losses in production. An aggregate relationship is given as follows.

3.1.2.4 Intermediate Lengths of Growing Periods

From the frequency distribution information (Section 3.1.2.1) occurrence of intermediate lengths of growing periods was quantified by relating P/ET ratio and moisture excess values with length of growing period.

The P/ET ratio for the intermediate lengths of growing periods of less than 150 days corresponds to values in the range 0.70 – 0.75.

The relationship which exists between the individual length of growing period and occurrence of intermediate periods is shown here.

3.1.3 Thermal Zones

To identify thermal zones, temperature criteria corresponding to the requirements of crops (including pasture and fuelwood), were taken into account (Technical Annexes 3, 4, 5, and 6)¹.

¹ Field crops, pasture and fodder grasses and legumes, and fuelwood tree species have been classified into temperature-photosynthesis adaptability groups. Four temperature adaptability groups are recognized for field crops, four for pasture and fodder grasses and legumes and two for fuelwood species (each with three productivity classes).

To cater for differences in temperature requirements of crops in the compilation of the country inventory, commensurate with the scale of the assessment (1:1 million), thermal regimes have been defined based on 2.5 °C intervals. A thermal difference of 2.5 °C corresponds to an altitudinal change of some 385 m, thus allowing a sufficiently fine matching of crop thermal requirements to prevailing thermal conditions as inventoried².

² Data from meteorological stations above 150 m altitude conform closely to the following relationship between average daily temperature in degree Celcius (T) and altitude in metres (A): T = 30.2 – 6.496 ( A /1000 ). Temperature seasonality effects of latitude are minor due to the equatorial position of Kenya.

For Kenya nine reference thermal zones have been recognized as shown here.

3.1.4 Area Inventory of Climatic Resources

The area inventory of thermal zones, pattern of growing period zones and length of growing period zones, by district was prepared at 1 : 1 million scale.

The mapped inventory was compiled by:

1. plotting the individual station data of temperature, pattern of length of growing period and mean total dominant length of growing period¹; and

2. constructing boundaries of thermal zones, pattern of number of length of growing period zones, growing period zones and isolines of mean total dominant length of growing period with the values 0, 30, 60, 90, 120, 150, 180, 210, 240, 270, 300, 330, 365 and 365⁺ days respectively, delineating the mean total dominant length of growing period zones of 0, 1–29, 30–59, 60–89, 90–119 120–149, 150–179, 180–209, 210–239, 240–269, 270–299, 300–329, 330–364, 365^- and 365⁺ days.

¹ The individual component lengths of both the dominant (mapped) and associated length of growing period zones are presented in Technical Annex 1.

In addition to normal extrapolation techniques, extensive use was made of landsat images, climatic maps, vegetation maps, land use maps, topographic maps and soil maps to guide the delineation of boundaries and isolines.

The three climatic inventories in map form were digitized, and the digitized information from the maps was converted to a grid cell data base.

A generalized map thermal zones is presented in Figure 3.5, and their extents are presented in Table 3.8. A generalized map of mean total dominant length of growing period zones is presented in Figure 3.6, and their extents are presented in Table 3.9. A generalized map of pattern of length of growing period zones is presented in Figure 3.7, and their extents are presented in Table 3.10.

Extents of mean total dominant length of growing period zones by pattern of length of growing period zones for each of the thermal zones, and for all thermal zones combined, are presented in Technical Annex 1.

3.2 Soil Resources

3.2.1 Exploratory Soil Map of Kenya

The Exploratory Soil Map of Kenya (Siderius and van der Pouw 1980; Sombroek, Braun and van der Pouw 1982) at 1:1 million scale was used to compile the soil resources inventory for this assessment.

FIGURE 3.5
Generalized map of thermal zones

FIGURE 3.6
Generalized map of mean total dominant length of growing period zones

FIGURE 3.7
Generalized map of pattern of length of growing period zones

TABLE 3.8
Extents of thermal zones

TABLE 3.9
Extents of mean total dominant length of growing period zones

This soil map, published by Kenya Soil Survey in 1980, provides the latest country-wide soil data base and includes information on distribution and characteristics of soils, landform and geology/parent material.

3.2.2 Soil Mapping Units

The soil mapping units are soil associations or soil complexes composed of dominant soils, associated soils and inclusions. The soil mapping units have been registered on the map by a symbol reflecting the landform in which they occur. In Table 3.11 the occurrences of soil mapping units are presented by landform.

The Exploratory Soil Map consists of 390 different soil mapping units. Extents of the individual soil mapping units are presented in Technical Annex 1.

TABLE 3.10
Extents of pattern of length of growing period zones

For each soil mapping unit the following semi-quantified information in terms of description, classes and extents has been transferred to the soil resources data base of this assessment:

-   Landform

-   Geology/Parent material

-   Soil units (with implied characteristics)

-   Slope-gradient classes

-   Soil texture classes

-   Soil phases

3.2.3 Landform

Landform is the first entry in the legend of the Exploratory Soil Map. It provides information on physiography, altitudinal position and slope patterns. A generalized map of landforms in Kenya is presented in Figure 3.8, and a description of landforms and their extents is presented in Table 3.11.

For the Exploratory Soil Map of Kenya six slope classes have been employed in 12 combinations. The slope classes are: A, 0–2%; B, 2–5%; C, 5–8%; D, 8–16%; E 16–30%; and F > 30%.

TABLE 3.11
Occurrence of soil mapping units by landform

A generalized map of slope-gradient classes in Kenya is presented in Figure 3.9. The combinations of slope classes employed and their extents are presented in Table 3.12.

To each of the 12 slope classes, inventoried in the Exploratory Soil Map, associated slope classes have been assigned. These associated slope classes, covering up to 10% of the land area of the 12 slope classes, are used for evaluation purposes and included in the land resources inventory. The inventoried slope classes and associated slope classes are presented in Table 3.13. For the same purposes assumed mean slopes of quartiles of the land area of each of the slope classes have been assigned. These values are presented in Table 3.14.

FIGURE 3.8
Generalized map of landforms

FIGURE 3.9
Generalized map of slope-gradient classes

TABLE 3.12
Extents of slope classes

TABLE 3.13
Associated slope classes

TABLE 3.14
Quartiles of slope classes

3.2.4 Geology

Geology is the second entry in the legend of the Exploratory Soil Map. The geological subdivisions reflect mainly resistance to weathering and richness of parent material in order to provide linkage with soil formation. The first level subdivision comprises three types of rocks: igneous rocks, metamorphic rocks and sedimentary rocks.

Igneous rocks and metamorphic rocks are futher subdivided from basic to acid. The sedimentary rocks are futher subdivided from fine to coarse textured.

Each soil mapping unit is accordingly characterized for its geological setting/parent material. A generalized map of geology/parent material is presented in Figure 3.10 and the descriptions and extents of geological units are presented in Table 3.15.

3.2.5 Soil Units

The individual soil units of the soil associations or soil complexes (soil mapping units of the Exploratory Soil Map) have been defined in accordance with the FAO-Unesco legend of the Soil Map of the World (FAO 1974). The soil units adopted were selected on the basis of present knowledge on the formation, characteristics and distribution of the soils, their importance as resources for agricultural production and their significance as a factor of the environment.

In the legend of the Exploratory Soil Map some adaptations of the FAO-Unesco legend have been introduced. At the first level (great group) the terminology for Lithosols and Nitosols has been modified. At the second level (unit level) new subgroups have been introduced (cambic and orthic Rendzinas) and others modified (vertic Gleysols, mollic Nitisols, chromic Acrisols, chromic Luvisols and chromic Cambisols). In order to reflect the greater amount of detail of the Exploratory Soil Map of Kenya, a third level of terminology (subunit level) has been introduced for subdivision of soil units into subunits. The prefixes used to distinguish subunits are ando, calcaro, chromo, ferralo, luvo, nito, ortho and verto.

In the Exploratory Soil Map, 123 different soil units and five miscellaneous units occur. Table 3.16 presents the soil units and miscellaneous units and their extents.

The soil units have been defined in terms of measurable and observable properties of the soil itself, and specific clusters of such properties are combined into ‘diagnostic horizons’ and ‘diagnostic properties’.

The diagnostic horizons have been used as defined in the FAO-Unesco legend. Diagnostic horizons and properties of the soil units are given in Technical Annex 1.

Complete definitions of soil units are given in Volume 1 (Legend) of FAO-Unesco Soil Map of the World (FAO 1974) and Exploratory Soil Survey Report (KSS 1982a).

FIGURE 3.10
Generalized map of geology/parent material

TABLE 3.15
Geology/parent material

¹ If the source of alluvial sediments and bottomland infills is known (e.g., basalts), then the code for this rock is used, otherwise the code A applies.