Document Type : Original Article
Authors
1 PhD student of Irrigation and Drainage, University of Birjand, Birjand, Iran.
2 Master of Science in Hydraulic Structures, University of Zabol, Zabol, Iran.
3 Associate Professor, Department of Water Science and Engineering, University of Birjand, Birjand, Iran.
Abstract
The decline in precipitation and excessive groundwater extraction in recent decades, particularly in arid and semi-arid regions such as Birjand County in South Khorasan Province, has resulted in a significant decrease in groundwater levels and a reduction in qanat flow rates. Given that Birjand County possesses over 1875 qanats, contributing to a total discharge of 23 million cubic meters annually, and relies on qanats for more than 90% of its water consumption, accurate prediction of their flow rates is of paramount importance. In this research, an adaptive neuro-fuzzy inference system (ANFIS) has been employed as a powerful tool for modeling intricate and nonlinear systems. This model is capable of identifying the complex relationships between input variables such as rainfall, evaporation, and groundwater level, and the output variable, which is the qanat flow rate. The model's capability enables it to deliver precise predictions of future flow rates.
The findings of this study demonstrate that the ANFIS model outperformed other models, achieving a correlation coefficient of 0.98, a Nash-Sutcliffe efficiency of 0.97, and a mean squared error of 0.049. This exceptional accuracy underscores the model's potential for predicting qanat flow rates. Consequently, this model can be effectively utilized in decision-making processes related to the sustainable management of groundwater resources within the study area.
Keywords
Main Subjects
)