Modeling monthly evaporation amounts and related data using AI and reducing evaporation losses

As highlighted by internationally recognized climate change reports, water scarcity is expected to become an imminent global concern. In addition to this, factors such as increasing population density and changing living standards have led to a significant rise in water consumption. Considering that these two issues are also relevant to our country, the efficient utilization and monitoring of water resources has become an urgent necessity. Within the scope of this study, factors causing water loss—specifically evaporation—and the parameters influencing it have been examined in the context of protecting water resources. Due to its inherently complex nature and dependence on numerous variables, evaporation is particularly difficult to estimate accurately. Traditional methods of estimating evaporation often fall short in terms of location specificity, seasonal variability, data type and quantity, as well as the quality of measurements. In recent years, artificial intelligence (AI), which has shown considerable advancements, has proven to outperform traditional estimation methods by producing more reliable and meaningful results.In this study, formulas selected from conventional evaporation estimation methods and models developed using AI were compared against reference evaporation values derived from both raw and processed data. Among the AI-based models, the Ensemble model provided the most meaningful results for both raw and processed data.