Abstract
Tourism activities are changing the global landscape pattern. This study attempted to estimate changes in Land Use Land Cover (LULC) and Land Surface Temperature (LST) in District Buner and Shangla, Khyber Pakhtunkhwa (KPK), Pakistan, and specifically its tourist spots. Using remote sensing data from satellites (1990-2020) and future projections (2035-2050), we applied Artificial Neural Network (ANN) and Cellular Automata Markov (CA-Markov) models to examine past and future LULC and LST dynamics across two districts including four major tourist spots (Shangla Top as tourist spot one (TS1), Bar Puran (TS2), Shahida Sar (TS3), and Daggar (TS4). The LULC classification for the whole study area (1990-2020) indicates that built-up and agricultural areas increased with a net change of +0.8% and +3.2% for the Shangla and Buner districts, respectively. The highest mean LST was found in the built-up areas. The simulation results indicate an expansion of 4.5% and 5.8% of the total built-up areas, and the LST above 31 °C will cover 76% and 88% of the total areas in 2035 and 2050, respectively. This conversion is driven by tourism activities, causing urban heat island effects (UHIs), and environmental degradation. The analysis of tourist spots (1990-2020) shows the highest change in built-up areas at Shangla Top (TS1), while the highest LST (28 °C) for the Daggar (TS4). The future simulation (2035-2050) results for tourist spots show that TS4 would have the highest LULC change in built-up areas (5.67%), and TS4 would have the highest LST (31 °C) from 65.23 to 82.20%. These findings provide an essential understandings for developing long-term tourism policies meant to moderate the environmental impact of tourism in the region.