Integration of artificial intelligence–based LULC mapping and prediction for estimating ecosystem services for urban sustainability: past to future perspective

The present study aims to investigate the dynamics of land use and land cover (LULC) and their relationship with ecosystem services value (ESV) from the past to the future (1990–2028) in Abha-Khamis twin city of Saudi Arabia. The support vector machine (SVM) was used to classify LULC maps over the years 1990–2018, and their dynamics were examined using delta change and a Markovian transitional probability matrix (TPM). From 1990 to 2018, the ESV was calculated for each LULC type using a coefficient. An artificial neural network-cellular automata model was used to predict the future LULC map for 2028. Sensitivity analysis has been performed using the probability distribution function, Pearson’s correlation methods, random forest, and classification and regression tree. Future LULC was used to derive future ESV from different ecosystems. The results of LULC maps showed that urban areas rose by 334.4% between 1990 and 2018. Delta change rate showed that urban areas have increased by 16.34% since 1990, while the TPM for the period of 1990–2018 reported that the built-up area was the largest stable LULC with a TPM value of 83.6%, while agricultural land, scrubland, exposed rocks, and water bodies were converted by 17.9%, 21.8%, 12.4%, and 10.5%, respectively, into built-up areas. Due to the accelerated and continuous urbanization process, all-natural resources and ecosystem services have been diminished considerably except for dense vegetation. The future LULC map of 2028 showed that the built-up area would be 343.72 km2, followed by scrubland (342.98 km2). The new urban area in 2028 would be 169 km2. The sensitivity analysis showed that proximity to the urban area, vegetation, and scrubland are highly sensitive to simulating and predicting the LULC maps of 2018 and 2028. The authorities and planners should focus more on the sustainable development of the urban areas; otherwise, it would harm both the natural and urban environments.