Assessing past and future prediction of land use land cover changes using CA-Markov chain model
摘要
This study aimed to simulate the past and future Land Use Land Cover Change (LULCC) in the Faisalabad region using Landsat images. The LULCC was calculated using supervised classification and a maximum likelihood algorithm for the years 1983, 2003, and 2023. Some softwares including ArcGIS 10.8, ERDAS Imagine 2015, and IDRISI 17 were used to analyze the past LULCC and future prediction. A hybrid simulation model (CA-Markov model) was used to simulate the future LULCC for the years 2043 and 2063. According to our results, the vegetation area was decreased from 532,497 Ha (90.93%) in 1983 to 511,896 Ha (87.41%) in 2023, marking a decrease of 20,601 Ha (3.52%). Built-up area was increased by 35,315 Ha (6.03%) between 1983 and 2023 and is projected to grow by an additional 27,864 Ha (4.76%) by 2063. The overall accuracy and kappa statistics have declined from 92% to 0.89 in 1983 to 85% and 0.8 in 2023 respectively. The demographic changes from 1981 to 2023 reflect a significant urbanization trend. The urban population has grown more rapidly than the rural population, resulting in a higher urban ratio by 2023. This study provides helpful information about LULCC planning and decision making in improving future LULCC policies and formulating management policies within the framework of sustainable land use planning.