Baringo (Kenya) Pilot Study for Desertification Assessment and Mapping

The purpose of the Kenya Pilot Study was to evaluate the FAO/UNEP
Provisional Methodology for Assessment and Mapping of Desertification,
and to recommend an effective, simple methodology for desertification
assessment within Kenya.

The FAO/UNEP Provisional Methodology (1984) proposes seven processes
for consideration in desertification assessment: degradation of
vegetation, water erosion, wind erosion, salinization, reduction of
organic content, soil crusting and compaction.

In late 1985, a pilot project for the assessment of the FAO/UNEP
Methodology within Kenya was proposed, and in 1987 a memorandum of
understanding between the Government of Kenya and UNEP for the
implementation of that study was signed.

The study areas were:
1) Models can be useful to assist in desertification assessment.
Models can be developed from FAO/UNEP Methodology.

2) Any modeling output requires verification.

3) Ground survey and remote sensing can be important sources of data.

4) An evaluation of data and methodologies necessary to allow
verification of desertification assessment modeling is required.

5) A human use component should be incorporated into desertification
assessment that considers management implications and social, as well
as, economic context.

6) Computer implementation of desertificaiton assessment can be
effective, however, procedures should be well defined.

This study within the Baringo Study Area was designed to address these
concerns.

The Baringo Study Area identified in this study would be typical of
such a training area. The models developed during this study could be
applied to the general region.

The study area lies between 0 15'-1 N and 35 30' -36 30' E.

It is located between the Laikipia escarpment to the East and the
Tugen Hills to the West. Topographic elevations vary from 900m on the
Njemps flats to 2000m in the Puka, Tangulbei and Pokot highlands. The
size of the study area is approximately 15ookm2.

4.0 DATA COLLECTION

A wide variety of data was collected. Detailed data was required to
provide a basis for evaluating more general cost effective data
gathering techniques and to provide a basis for model verification,
particularly the socio/economic data.

Physical Environment
Topographic Data
Topographic contours were digitized directly from 1:250,000 Survey of
Kenya topographic maps. The contour interval was 200 feet. A digital
elevation model was constructed using triangular irregular networks
(TIN).

Soil Data
Soil types were mapped at 1:100,000 scale using existing soil maps,
manual interpretation of SPOT imagery, and field investigations
(Figure 3). During field trips, soil samples were taken from each
soil unit and analyzed by the Kenya National Agricultural Center.

4.2 Climate Data
4.2.1 Rainfall Data
Rainfall data from the Kenya Meteorological Department was analyzed
for 33 stations within and surrounding the study area. A rainfall
erosivity index was calculated based on the Fourier Index (R).
12
RE (p /P)
12
where P = annual rainfall
p = monthly rainfall

A relationship between this erosivity index and the annual rainfall
for each station was calculated using linear regression (Bake, 1988).
A map of rainfall erosivity was generated for the study area by
relating annual rainfall isoheyts to the following:

y = 0.108x - 0.68

This data was coded and digitized.

Wind Erosion Potential
The following required conditions were determined to create high wind
erosion potential (Kinuthia, 1989):

1) Annual rainfall less than 300mm.
2) P/E greater than zero and less than 1, where:
P=mean monthly rainfall (cm).
E=mean monthly PET (cm).
3) Wind velocity greater than 4 m/s at 10m height.

Vegetation Data
A vegetation map for the study area was produced at a scale of
1:100,000 through manual interpretation of a SPOT image and field
investigations (Figure 6). A structural classification system as
adopted by DRSRS was used for naming vegetation types (Grunb).

Systematic Reconnaissance Flight Data Since 1977, DRSRS has been
conducting aerial surveys of Kenyan rangelands. In addition to data
on the number of wildlife and livestock, observations of land use and
environmental condition are also made.

Socio/economic Data
Social Factors

A wide variety of data was collected through literature review and a
field administered questionnaire. Nutritional status was estimated by
measurement of childrens' mid upper arm. Such data is useful for a
Level 1 type assessment.

Permanent Structures Data
For the Level 2 assessment, data on permanent structures was extracted
from DRSRS SRF data. This data was used to indicate presence and
concentration of sedentary populations.

Example
Files: VDS.E00 (Vegetation degradation)
DES.E00 (Plant Species)
Others available on request.