Application of Shannon Entropy Theory in Predicting Potential Evapotranspiration (Case study : Urmia Synoptic Station)

Subject Areas : hydrology

Navid Zeinalpour ¹ , alireza parvishi ^{2 *}

1 - Department of Civil Engineering, Urmia Branch, Islamic Azad University, Urmia, Iran.
2 - Department of Civil Engineering - Islamic Azad University - Urmia Branch- Iran

Received: 2022-01-27 Accepted : 2024-05-20 Published : 2024-12-02

Keywords: Shannon Entropy, Basic Climate Patterns, Basic Delay Patterns, Penman-Monteith FAO Method,

Abstract :

Evapotranspiration is one of the most important components of the hydrological cycle and its accurate estimation is used in many studies such as water balance, water resources management and irrigation planning. The use of intelligent models can be a good tool for estimating nonlinear variables such as evaporation and transpiration. The present study used GEP gene expression programming methods and ANFIS fuzzy-adaptive neural inference system to predict monthly reference evapotranspiration. For this purpose, two different modeling modes were developed. The first case of baseline climate patterns and the second case of memory role in predicting monthly reference evapotranspiration, Shannon entropy method was used to select the most optimal inputs. According to the results of GEP model in Ent-CBM8 model with KGE = 0.91, WI = 0.87 and RMSE = 0.495 had the best performance in predicting the monthly reference evapotranspiration of the synoptic station. The results of the implementation of models with basic delay patterns and Shannon's entopy method were able to correctly identify the optimal delay.