Accelerating agricultural expansion in the greater Mau Forest Complex, Kenya

Abstract

Land use land cover (LULC) change can modify local, regional, and global socio-environmental systems. Globally, much LULC change study is aimed at deforestation and urbanization with less focus on other LULC categories like grasslands and savannas. Here, we focus our LULC change study on southwestern Kenya, a topographically varying region of forest, grassland, and cropland. This study leveraged existing remotely sensed Landsat landcover data to evaluate uncertainty of LULC classification, quantify the nature and magnitude of changes, and illustrate the role of topography (slope and elevation) in determining the spatiotemporal characteristics of the landscape between 1990 and 2018. We also analyzed landscape metrics of fragmentation and dominance at the class-level and employed Kruskal-Wallis tests to examine the characteristics and statistical significance of the changes. We obtained an overall classification accuracy of 86% and establish that over the 28-year period, cropland increased by 22.5% and became less fragmented, forest decreased by 6.6% and became less fragmented, while grassland decreased by 16% and became more fragmented. Results showed statistical difference (p < 0.05) in LULC change among different topographic classes. Our results indicate that deforestation in the region has slowed in the past decades, likely due to conservation efforts, but conversion of grassland to cropland has accelerated. Grasslands, including pastures, provide essential ecosystem services both to people and the environment and should not be overlooked in conservation initiatives.