The impact of land use and land cover changes on land surface temperature in the first ring of Suceava Metropolitan Area

Abstract

Urbanization and agriculture intensification accelerate the land use and cover conversions unbalancing the surface energy budget. Land Surface Temperature (LST) represents the land radiative skin temperature which is derived from solar radiation and is one of the most important indicators of local climate variability. The present work aims to analyze the effects of land use land cover changes (LULCC) on spatial pattern distribution of Land Surface Temperature in the first ring of Suceava Metropolitan Area over the last 35 years. In order to meet our demand we have conducted a spatio-temporal analysis using Geographical Information Systems (GIS) and Remote Sensing (RS) techniques. We have used two satellite images from Landsat 5 TM (23 August 1985) and 8 OLI/TIRS (23 August 2020) in order to create land cover maps by applying a supervised classification with spectral angle algorithm and to estimate Land Surface Temperature through the Plank Equation. Given that we have applied the supervised classification to define the four major land cover classes (bare soil, built-up area, vegetation and water bodies) we have used the classification-based method to determine the surface emissivity. The overall accuracies of the land cover maps of 1985 and 2020 were found to be 93.45%, and 96%, while the Kappa coefficients were found to be 0.90 and 0.94 for the years 1985 and 2020, respectively. The land cover change matrix showed that during the study period, 140.67 km2 representing 34.60% of the total study area faced mutual conversion among four land cover types while 265.90 km2 representing 65.40%  of the total study area remained unaltered. More exactly, built-up area and vegetation surfaces increased by 78.31% and 3.78%, respectively, while bare soil and water bodies decreased by 38.71% and 10.21%, respectively. LSTs found in the study area ranged from 18.27 to 33.91°C and 21.67 to 40.48°C for the years 1985 and 2020. The increases of spatially distributed maximum, mean and minimum LST were found 6.57°C, 3.84°C and 3.40°C, respectively. This means a LST increase by around 0.11°C per year for the study period of 35 years. Moreover, the results showed that covers without vegetation and artificial surfaces have recorded the highest temperatures: 26.18 to 30.11°C and 25.49 to 29.66°C for bare soil and built-up area, respectively. The increases of mean LSTs were 4.16°C, 3.96°C, 3.92°C and 3.53°C for the bare soil, vegetation, built-up area and water bodies, respectively, during the study period. On the other hand, in 1985 the highest maximum LST was 33.91°C in the built-up area followed by 31.26°C, 28.11°C and 24.90°C by bare soil, vegetation and water bodies, respectively, while in 2020 the highest maximum LST was 40.48°C in the built-up area followed by 35.28°C, 33.54°C and 28.98°C by bare soil, vegetation and water bodies, respectively. Based on the above findings, policymakers and urban planners should be concerned about future urban expansion and agriculture management in order to reduce the LST-related urban heat island or drought intensification problem.