11:00 AM - 11:15 AM
[U04-07] Integrated Analysis of Historical and Future Land Use and Land Cover Dynamics in a Tropical River Basin: A Multi-Technique Remote Sensing Approach
Keywords:Cellular Automata-Markov algorithm, Multi-Layer Perceptron modelling, Land use/Land cover (LULC) modelling, Remote sensing/GIS analysis
To effectively manage land resources and plan for sustainable development, it is essential to understand historical and future land use and land cover (LULC) changes. This study examines LULC changes in the Bernam River basin, utilizing a comprehensive approach that integrates multiple techniques, including remote sensing/GIS, Multi-Layer Perceptron (MLP) modelling, and Cellular Automata (CA) -Markov algorithm. Seven LULC categories were identified using multi-temporal 10m resolution Sentinel-2 Landsat images from 2000 to 2019, and 2021 to 2023: water, forest, wetlands, agriculture, urban, barren, and rangeland. Analysis revealed changes from 2010 to 2020, with annual increases in water (0.24%), forest (0.61%), and urban (2.11%) areas from 2010 to 2020, while wetlands (-2.69%), agriculture (-2.47%), barren (-3.51%), and rangeland (-4.58%) areas decreased. The predictive accuracy of the model for 2021 to 2023 was 91.56%. Projections for 2025-2075 show rising trends in water (1.76%), wetlands (29.18%), agriculture (60.08%), urban (96.53%), barren (0.59%), and rangeland (3.57%), with a net decrease in forests (12% ≅ 261.52 km2). The study identified agriculture and urban growth as primary drivers of LULC changes within the river basin. These results can inform evidence-based policies and management strategies for the sustainability of the Bernam River basin and similar river basins. Additionally, predicted LULC patterns can be used as inputs for other models, such as the Soil and Water Assessment Tool (SWAT), for comprehensive evaluations of environmental, agricultural, ecosystem, and water resource impacts.