Analysis of land-cover changes in the Transboundary Sio-Malaba-Malakisi River Basin of East Africa: Towards identifying potential land-use transition regimes

Abstract

ABSTRACT This study evaluated historical land-cover states in order to identify potential land-use transition regimes leading to land degradation. Landsat satellite datasets were used to characterize land-cover states for 1986–2017 period. The multinomial probability distribution was used to establish sample size for training and accuracy assessment. Using a hybrid image classification approach, individual satellite images were initially clustered using the ISODATA technique, and spectral classes later transformed posteriori into respective thematic classes. Maximum Likelihood Function was subsequently used to assign pixels into classes with highest probability. Approximately 12% of mixed forest declined, while cropland increased by 30% between 1995–2008.