Artificial Intelligence (AI) Techniques Application for Modelling Expected Trends in Pan Evaporation in Slovak River Basins

The modelling of pan evaporation (Ep) trends in Slovak river basins was performed by utilizing artificial intelligence (AI) techniques algorithms to accurately forecast evaporation rates based on daily climate data spanning from 2010 to 2023 across eight sub-basins within the Slovak Republic. The findings derived from the AI modelling indicate that the river basins of Bodrog, Hornád, Ipeľ, Morava, Slaná, and Váh are experiencing increases in evaporation measurements, whereas the Dunaj and Hron rivers demonstrate declining trends. This divergence may suggest the presence of differing ecological factors that affect the evaporation dynamics associated with each river. In this study, a comprehensive set of 28 machine learning and deep learning models was employed, including: Machine Learning (ML): Linear Regression, Tree-Based, Support Vector Machines (both with and without Kernels), Ensemble, and Gaussian Process methods; Deep Learning (DL): Neural Networks (Narrow, Medium, Wide, Bilayered, and Trilayered). The Stepwise Linear Regression yielded the most optimal fit. The Minimum Redundancy Maximum Relevance (mRMR) method was utilized to assess the efficacy of feature selection by concentrating on both relevance and redundancy. The results suggest that placing greater emphasis on relative humidity (RH) and minimum temperature (tmin) may significantly enhance the predictive accuracy of the model.