Remote sensing based assessment of small wetlands in East Africa

Abstract

Small wetlands in East Africa have in the past few decades become focal points of a broad spectrum of agricultural production and other land-uses. Climate change and population growth are the major factors attributing to increasing use and change of the wetlands. This study aimed at detecting the distribution and extent of small wetlands in Tanzania and Kenya, classifying them into different types, identifying their use patterns and quantifying changes that have taken place from 1976 to 2003.<br /> Field and aerial surveys were conducted; microwave (ALOS-PALSAR, ENVISAT-ASAR, and TerraSAR-X) and optical (LANDSAT and aerial photographs) data, were used to detect spatial distribution of the wetlands using automated and semi automated techniques. Time series LANDSAT images were applied in classification and change detection by post classification comparison (PCC), change vector analysis (CVA) and land use change mapper (LCM). Maps and socio-economic data were also gathered. Driving forces of change were determined qualitatively using group discussions with key informants.<br /> Two types of small wetlands were mainly identified, inland valleys located in the humid highlands and covering 87% of the total surveyed area as well as floodplains in sub-humid lowlands and semi-arid highlands covering the remaining 13%. Eight major land cover and uses were identified with accuracies between 82.76 and 95.17%. Cropland was a dominant land use occupying 57% of the inland valleys and 35% in the flood plains; others included open water, floating vegetation, permanent papyrus swamps, semi-natural vegetation, grazing, shrubs, settlements and bare land. The cover and uses are unevenly distributed between the types and sites.<br /> The major change detected was expansion of cropped land at the expense of natural vegetation. This accounted for 56% of the change in the highland flood plain and 52% in the lowland floodplain. Shrubs proliferated in all wetlands, which is indicated by more than 50% compared to their area coverage in 1976. Climate change, population increase, unemployment, market access, wetland physical access and insufficient knowledge on the use are among the proximate causes of the wetland changes. Underlying factors like poor enforcement of wetland law and policy in Kenya and lack of the same in Tanzania have accelerated these changes.<br /> Combinations of remote sensing data and image processing methods played an important role in achieving the objectives of the study. Optical data proved to be very useful in delineation of small wetlands while microwave data delineated larger areas. The spatial resolution of the images has also proved to be a key factor in studies of small wetlands. To ameliorate the wetlands, it is recommended that a balance is attained between the use and conservation. Policy formulation and law enactment in Tanzania and enforcement of the existing policy and law in Kenya is seen to support wise use. Awareness creation is also important to lessen the over and inappropriate utilization of the wetlands.