Predicting the impacts of land use/land cover change on land surface temperature using remote sensing approach in Al Kut, Iraq

Abstract

Changes in land use/land cover (LULC) classes significantly impact land surface temperature (LST). This study predicted LULC change and their impacts on seasonal (summer and winter) LST variations in Al Kut, Iraq. Landsat TM/OLI images for the years 2000, 2010, and 2020 were used to estimate LULC and LST's past status using remote sensing (RS) techniques. Based on the past characteristics of the images future LULC and seasonal LST were predicted for year 2030. Support vector machine (SVM) algorithm was used to classify the LULC classes. Artificial neural network (ANN) algorithm was used to predict the future LST considering the LULC indexes as influential variables. Results suggest a significant increase in urban area by +8.74% and reduction in green cover by −25.87% from 2000 to 2020. Increment in maximum LST took place for both summer and winter season by 3.79 °C and 3.16 °C in last two decades. Maximum LST was recorded in urban area (48 °C) and water bodies (35 °C) exhibit minimum LST. The correlation study demonstrates a strong positive relation of LST with Normalized Difference Built-up Index (NDBI) and negative relation with Normalized Difference Vegetation Index (NDVI) and Normalized Difference Water Index (NDWI). The prediction result also demonstrates increase in urban area (+17.02%) and loss of vegetation cover (−15.57%) for 2030. The maximum LST will likely to increase by 1.62 °C and 2.68 °C for summer and winter seasons in predicted year 2030. This study will provide effective guidelines for urban planners of Kut city for ensuring planned and sustainable urban development in future.