Leveraging CA-ANN modelling for SDGs alignment: Previse future land use patterns and their influence on Mirik Lake of sub-Himalayan Region

Abstract

The dynamics of land use and land cover are significantly impacted by population growth, mobility, and demand. The objectives of this study are to identify the transition of land-use changes in the vicinity of Mirik Lake between 1990 and 2020. The study uses satellite imagery obtained from the United States Geological Survey (USGS). The datasets have been acquired from the platforms of Landsat 5 Thematic Mapper, 7 Enhanced Thematic Mapper+, and 8 Operational Land Imager. To forecast and determine possible land-use changes for the years 2030 and 2040, the Modules for Land Use Change Evaluation (MOLUSCE) plug-in Multilayer Perceptron-Artificial Neural Network (MLP-ANN) was trained with the factors DEM, aspect, slope, hillshade, proximity to built-up, waterbody, and road. The 2020 projected raster show a high degree of accuracy with a Kappa value of 0.62 (overall), 0.89 (histogram), 0.69 (location) and a correctness percentage of 71.62 %. This study illustrated a decrease in the natural environment and a significant rise in the built environment. The simulation result indicates a 19.12 % to 65.18 % increase in built-up area at the cost of lakes, open space, cropland, and vegetation. If these patterns persist, the future scenario of land use and land cover (LULC) will exhibit the same pattern. The rigorous alternation of land use directly impacts the lake area in terms of aerial extent. The findings of the present study are to provide an effective management strategy to meet the SDGs regarding the preservation and conservation of the lake in the near future.