Predicting priority management areas for land use/cover change in the transboundary Okavango basin based on machine learning

• DOI: 10.1016/j.heliyon.2023.e22762
• Publication Date: 2023-11-23T18:46:53Z
• Authors: Blessing Kavhu; Zama Eric Mashimbye; Linda Luvuno
• Keywords: Land Cover; Shrubland

Remote sensing and modelling of land use/land cover (LULC) change is useful to reveal the extent spatial patterns landscape changes at various environments scales. Predicting susceptibility LULC crucial for policy formulation management. However, use machine learning (ML) limited. This study modelled in Okavango basin using ML techniques. Areas with high are termed priority management areas (PMAs) this study. Trajectories between 1996 2020 derived from existing maps basin. Overlay analysis then used detect patches transitions. Three transitional categories adopted PMAs, namely 1) natural anthropogenic classes (Category A); 2) B); 3) another class C). An ensemble algorithms calibrated social-ecological drivers produce showing Thereafter, thresholding done on probability based maximum sum sensitivity specificity (max SSS) delineate PMAs. Results trajectories indicate that activities (croplands, built-up areas, barelands) generally expanded, displacing (wetlands, woodlands, water, shrubland) 2020. Regarding anthropogenic-related PMAs A ∼34 560 km2) covered a larger area compared ones (Categories B∼33 407 C∼15 040 km2). The findings emphasize value identifying guiding transboundary planning. Overall, highlights role driving Transboundary Drainage Basins (TDBs) suggests need promote sustainable practices predicted through comprehensive planning ensure water availability