Assessing and detecting spatiotemporal land use/cover changes in Uzbekistan using sentinel-2 imageries

¹ National University of Uzbekistan named after Mirzo Ulugbek, Tashkent, 100174, Uzbekistan
² Jizzakh branch of National University of Uzbekistan named after Mirzo Ulugbek, Jizzakh 130100, Uzbekistan
³ Jizzakh State Pedagogical University, Jizzakh, 130100, Uzbekistan
⁴ Urgench State University, Urgench, 220100, Uzbekistan
⁵ Samarkand State University of Architecture and Civil Engineering, Samarkand, 140143, Uzbekistan
⁶ Nukus State Pedagogical Institute named after Ajiniyaz, Nukus 230105, Karakalpakstan, Uzbekistan
⁷ Institute of Agriculture and Agrotechnologies of Karakalpakstan, Nukus, 12000093, Karakalpakstan, Uzbekistan
⁸ University of Geological Sciences, Tashkent, 100041, Uzbekistan

^* Corresponding author: This email address is being protected from spambots. You need JavaScript enabled to view it.

Abstract

Understanding Land use and Land cover (LULC) change is important in environmental modification and natural resource management. This study analyzed the dynamic changes of LULC in the Syrdarya province of Uzbekistan from 2017 to 2024 by applying AI-based classification Sentinel-2 data. The results of the study indicate that certain LULC categories have experienced shifts in extent. While the built-up areas expanded from 40,082 ha to 47,670 ha and the water bodies increased from 13,785 ga to 16,912 ha between 2017 and 2024, other LULC types such as bare land exhibited a substantial decrease from 1,136 ha to 305 ha with the cropland and rangeland experiencing moderate decline to fluctuations. Also, a correlation analysis was performed to better understand the interrelationship between these LULC categories. The flooded vegetation and water bodies (R = 0.62), built-up areas and water bodies (R = 0.69), bare land and cropland (R = 0.56) showed a strong positive relationship. However, the strong negative correlations between cropland and water bodies (R = −0.60), built-up areas and cropland (R = −0.58), bare land and water bodies (R = −0.84), bare land and built-up areas (R = −0.85), and rangeland and cropland (R = −0.88) were detected. As one of the primary driving factors of the LULC types, the province’s population has been considered. The most positive correlation (R=-0.96) was found between population and built-up areas.