Influence of Land Surface Temperature and Rainfall on Surface Water Change: An Innovative Machine Learning Approach

• Published in: Water resources management Vol. 37; no. 8; pp. 3013 - 3035
• Main Authors: Jain, Vanita, Dhingra, Aarushi, Gupta, Eeshita, Takkar, Ish, Jain, Rachna, Islam, Sardar M. N.
• Format: Journal Article
• Language: English
• Published: Dordrecht Springer Netherlands 01.06.2023; Springer Nature B.V
• Subjects: Atmospheric Sciences; Change detection; Civil Engineering; climate; Climate change; data collection; Drinking water; Dynamics; Earth and Environmental Science; Earth Sciences; Environment; Freshwater; Geotechnical Engineering & Applied Earth Sciences; Hydrogeology; Hydrology/Water Resources; Independent variables; India; Inland water environment; Internet; Land surface temperature; Learning algorithms; Machine learning; Mathematical analysis; Outflow; people; Precipitation; rain; rain intensity; Rainfall; Rainfall intensity; Rivers; Storage; Surface temperature; Surface water; temporal variation; Thematic mapping; Water level fluctuations; Water levels; Water outflow; India; Rainfall; Climate change; Google earth engine; Surface Water; Remote sensing
• ISSN: 0920-4741; 1573-1650; 1573-1650
• DOI: 10.1007/s11269-023-03476-2

The largest reservoir of drinkable water on Earth is surface water. It is crucial for maintaining ecosystems and enabling people to adapt to diverse climate changes. Despite surface freshwater is essential for life, the current research shows a striking lack of understanding in its spatial and temporal dynamics of variations in outflow and storage across a sizable country: India. Numerous restrictions apply to current research, including the use of insufficient machine learning techniques and limited data series. This work uses cutting-edge and SOTA-method to use the available data and machine learning to accurately understand spatial and temporal dynamics of variations in surface freshwater outflow and storage using extended data series. The authors did the examination of thematic maps produced using ArcMap 10.8 from June’2005 to June’2020 using JRC dataset to track changes in the intensity of surface water. Google Earth Engine in Python API has been devised to detect changes in surface water levels and quantifying shifting map trends. Raster image viewing, editing, and calculation are done with ArcMap. For determining the relationship between declines in Surface water levels, changes in rainfall intensity and land surface temperature, variables were averaged over 13 rivers for 15 years. The change in surface water is reliant on independent variables of change in land surface temperature and rainfall intensity. The authors use the correlation between these parameters to achieve an average R-squared adjusted value of 0.402. The study's findings contribute to a better understanding of the matter and can be used across the world.