Work

Urban Heat Island Dynamics and 2030 Thermal Scenario in Pokhara, Nepal Using Landsat and CA– Markov Modeling

Year: 2026
Type: preprint
Abstract: Urban heat islands are an increasing concern in rapidly urbanizing cities where vegetation and cultivated land are being converted into impervious and exposed surfaces. This study assessed land use/land cover change, vegetation condition, land surface temperature, and urban heat island intensity in the 37.42 km² core urban area of Pokhara Metropolitan City, Nepal. Landsat imagery from 2014, 2019, and 2024 was processed in Google Earth Engine. Support Vector Machine classification was used for land-cover mapping, while NDVI and EVI were used to examine vegetation conditions. Landsat-derived land surface temperature was evaluated using MODIS LST, and DHM air-temperature records were used as supplementary thermal-consistency evidence. A CA–Markov model was applied to generate a 2030 land-cover scenario and assess its likely thermal implications. Results showed that settlement increased from 20.01% in 2014 to 37.45% in 2024, while cultivation declined from 78.04% to 55.53%. Mean land surface temperature increased from 29.87°C to 37.46°C, and maximum temperature increased from 36.19°C to 44.58°C. Landsat LST showed reasonable agreement with MODIS LST, with a correlation coefficient of 0.736 and RMSE of 2.79°C. The 2030 scenario suggests that settlement and airport-related built-up surfaces may occupy 50.43% of the study area, while High and Extremely High UHI classes may together cover 50.63%. The findings highlight the need for urban greening, exposed-surface management, and heat-sensitive planning in Pokhara.
Language: English
Open Access status: green
OpenAlex ID: W7204100106
Cites: 18
Cited by:
Related to: 10