African Rain Forest Project

The current rates of global land cover changes are most dramatic in tropics (Houghton 1994) especially in the African rainforests where deforestation and forest fragmentation (and the resulting loss of Biodiversity) are increasingly becoming important. This has lead to alarming increases in the release of carbon dioxide (CO2) and other greenhouse gases (Andreae 1991, Botkin et al. 1989, Enting and Mansbridge 1991, Houghton et al. 1987, Houghton 1991), serious loss of Biodiversity (Terborgh 1992), and degradation of watersheds (Salati and Vose 1984). Deforestation and forest fragmentation as a result of slash-and-burn agriculture (also referred as shifting cultivation) is a major issue in the African rainforests. This has been shown in a preliminary study of the Alternative to slash-and-burn benchmark research area (Figure 1 and Figure 2). The forest fragmentation as a result of slash-and-burn along the "corridors" of road-networks and settlements (see Figure 3) has been most dramatic in recent years as a result of decreasing fallow periods. This phenomenon is seen not only in regions of significant populations and settlements such as, for example, around M'balmayo (Figure 4) but also in relatively dense rainforest areas like Ambam in Southern Cameroon (Figure 5). These fragmentation's lead to loss of Biodiversity, provide access to logging, encourage non-sustainable agriculture and decreasing length of fallow periods. Indeed the slash-and-burn agriculture is the cause for about sixty- percent of world’s deforestation (World Bank 1991), involving 300-500 million people mostly in tropics (FAO 1985). Also, studying fallows in the tropical forest systems is extremely important for agricultural researchers. Fallows are the storehouses of nutrients in their plant Biomass, which can be tapped by future crops by slash-and-burn. Thereby, a precise estimation of fallows is crucial for estimating the nutrients available in plant Biomass. Also, the lesser the time available for fallows the lesser is the fertility and productivity of the soils.

In tropics region, satellite remote sensing may be the only feasible technique to study forest clearing, shifting cultivation, and land use conversion trends (Sader 1995) with difficulties inherent in traditional methods (Behrens 1990). Further, studying tropical regrowth with high-resolution imagery such as Landsat TM could substantially improve estimates of carbon sequestration subsequent to tropical deforestation. Use of satellite data in combination with multidisciplinary research has been strongly advocated to find solutions to deforestation and in identifying solutions to environmental problems of global import (Moran et al. 1994) although regional scale satellite data alone might form the basis for empirical models with limited predictive capability (e.g., spatial trend or diffusion modeling) (Skole et al. 1994). A wide range of satellite imagery has been used to characterize and map tropical forests (Baltaxe 1980, FAO 1981, Green and Sussman 1990, Sader et al. 1990, Malingreau 1991, TREES 1991, Justice et al. 1992, Lawrence 1992, Lawrence and Chowmentowski 1992, FAO 1993, ENRIC 1994, Tucker 1995). For example, the widely varying estimates of deforestation (Monastersky 1993) are more reliably answered by systematic methodologies adopted using remote sensing (see for example, Skole and Tucker 1993). The distinct vegetation characteristics of uplands versus lowlands are critical to a proper understanding of the distribution of biomes (see Figure 6 showing lowland rainforest characteristics of Southern Cameroon). There is also now a greater possibilities of mapping wider range of land cover and vegetation classes of forests using newer high resolution satellites (see preliminary studies by Rao et al. 1996, and Roy et al. 1996). For example, simple vegetation index maps derived from remote sensing (see Figure 7) provide a powerful means of biomass assessment and land degradation. Further, vegetation mapping (e.g., forests, fallows, grasses, crops, and weeds) using remote sensing is one of the most common operations using remote sensing, is generally quick, and provides useful information to agricultural researchers with regard to a number of issues like presence and spread of weeds, spread and age of fallows (which in term help in assessing the Biomass and their nutrient potentials), and a general understanding of the stratification of agroecological zones.