of resource
ther semi-
hic techni-
considera-
igh photo-
documen-
TERRAIN ANALYSIS
Terrain is used here to refer to regions that are characterised by particular landforms, geolo-
gic deposits, soils, vegetation and hydrology (Townsend, 1981). Emphasis will be on the devel op-
ment of efficient classification, compilation and mapping of the identified components. Some
elements will be mapped separately for special areas, but the main aim will be to produce compo-
sites for the general area. Mapping will be based primarily on visual internpretation of enhanced
Landsat transparencies, but this will be complemented by the use of radar imagery, aerial and
ground photographs.
LAND USE/COVER MAPPING
This aspect of the project concerns the identification, classification and mapping of land use/
cover in selected areas for both the wet and the dry seasons. Ideally, Landsat data will be acqui-
red in accordance with the crop calendar, and this phase of the mapping will rely on the digital
processing of this multi-temporal Landsat data. Supervised classification will be applied, and en-
hancement and change detection techniques will also be used.
SOCIO-ECONOMIC ASPECTS
Resource studies are often criticised for ignoring human factors. As this project is primarily
aimed at the methodologies of remote sensing, it is not possible to accommodate a comprehen-
sive analysis cf all human factors. However, an attempt will be made to examine local environ-
mental perception as it relates to traditional land systems. Such information may permit the ma-
rriage of scientific and local land classifications, so enhancing scientific interpretation and assis-
ting the development of mutual understanding between ''scientist'" and farmer.
INSTRUMENTATION
Not withstanding the “reference level’ of the personnel involved, the nature, type and accu-
racy of the interpreted data depend to some extent on the instruments used. Digital analysis in
this project will depend mainly on a Dipix ARIES Il image analysis system located in the Faculty
of Environmental Studies at the University of Waterloo. This system will be used for Landsat
image registration, enhancements and classifications. An IBM—PC XT with "MICROPIPS"
software is available for analysis of more limited areas at the University of Lagos, For visual
analysis, a Zoom Transfer Scope and a Stereo Zoom Transfer Scope are available at the Univer-
sity of Waterloo, while a PROCOM-2 is available at the University of Lagos for image registration
and map revision.
Cartographic equipment at Waterloo includes 3 Map-O-Graph enlarger, a cartographic came-
ra and computer facilities, with SURFACE ll, DISSPLA and THEMAP software operated on an
IBM-4341 mainframe. Digitizing and plotting equipment is also available. At Lagos, equipment
includes a process camera, Grant pojector, Rectoplan plate ma king machine and contact printer.
THE SOKOTO-RIMA BASIN
The basin has a semi-arid climate with 2 prolonged dry season (October-May) and a short
wet season (end of May to early October). Mean annual rainfall varies from less than 700mm in
the north to about 1,100mm in the south. With high level of potential evapo-transpiration, the
area has a negative water balance in most months, and its marginality for agriculture is compoun-
ded by the high variability of the rainfall regime (Fig. 3). Although temperature is sufficient for
plant growth throughout the year (Table 1), the precipitation regime imposes severe limitations
on rain-fed agriculture.
SE
SÉRIE
RARE UT ic
i i! iiia