Cellular Automata and Markov Chain Model (CA-Markov), Based Forecasts of Future Land Use and Land Cover Scenarios During (2002-2043) in the Batha District Southern Iraq using GIS

Abstract

This study focused on forecasting future changes in land cover and land use in the southwest of Thi Qar Governorate for the years 2033 and 2043. The research utilized Landsat-7 and Landsat-9 satellite imagery, following the necessary digital processing of these images using ArcGIS 10.5 and ENVI software. Classification was performed using supervised classification techniques in Erdas Imagine 2014 and ArcGIS 10.5, adhering to Anderson's land use and land cover classification system.
The accuracy of the classification was evaluated using an error matrix based on field verification. The study aimed to predict land cover and land use categories for 2033 and 2043 using the hybrid Markov model and to quantify spatial changes in the area of different land cover types. Additionally, it sought to identify the categories experiencing the most significant changes, whether increasing or decreasing.
The research employed the IDRISI TerrSet software to simulate future changes in land cover and land use within the study area. This software enables the application of the CA-Markov model, which integrates cellular automata and Markov chain analysis. The model produced high-quality predictive maps illustrating differences in land cover areas across past, present, and future timeframes.