Prediction of Lake Van Water Level Using Artificial Neural Network Model With Meteorological Parameters and Multiple Linear Regression Analysis: a Comparative Study

Abstract

The water level of Lake Van has shown changes over time. This study encompasses a statistical investigation conducted to understand the reasons behind the variation in the lake's water level. In this study, an attempt has been made to establish a predictive model by determining the effects of meteorological factors on the lake's water level. Artificial neural networks have been utilized to predict the water level of Lake Van using meteorological parameters such as precipitation, temperature, evaporation, wind speed, relative humidity, and atmospheric pressure. Furthermore, a model equation has been formulated by examining the relationship between independent variables and the changes in the water level of Lake Van through multiple linear regression analysis. The two models have been compared, and the results have been evaluated. The obtained results indicate that the artificial neural network model can provide more realistic predictions for the water level of Lake Van compared to the multiple regression analysis method, demonstrating that artificial neural networks serve as a tool for both temporal and spatial predictions.