Land Use/Land Cover Prediction and Transformation for 2035 Utilizing MLP Neural Network and Markov Chain Model: A Case of Hisar City, Haryana, India

Abstract

The integrated different geospatial approaches such as GIS and multi-spectrum remote sensing-based analysis are significant in comprehending land use changes and provide invaluable information for the management of land and sustainable development. This study aims to analyse the LULC change between 2011 and 2021 and the simulation of future land use for the year 2035 in Hisar City and its per-urban area. The LULC maps of 2011 and 2021 are prepared by using Landsat images after applying the supervised classification and classified into six different categories. The respective maps are used for the predation of LULC of the year 2035 by using the Multi-Layer Perceptron Neural Network and Markov Chain model built-in Land Change Modeler tool of TerrSet software. The model has been validated by applying Validate and Crosstab modules. The examination of LULC change revealed that the area under built-up approximately doubled between 2011 and 2021 at the cost of open spaces, agricultural land and vegetation cover. The LULC model for 2035 predicted that the built-up will increase by 47 percent between 2021 and 2035. The outcomes of the study can be utilised for the sustainable urban development and conservation of fertile agricultural land.