Prediction Model for Hydrological Responses of Watershed using SWAT and ANN under Climate Change Scenarios

Abstract

Objectives: Absolutely, climate change is a significant concern impacting our planet in the 21st century. Prediction and understanding how climate change affects the earth planet especially water resources help in planning and managing them effectively. Therefore, in this research an attempt has been made to survey the influence of this issue on climatic and hydrological parameters in Saghez watershed in Iran. Methods: In this research to determine the trend and the slope of changes Sen’s slope estimator and Mann-Kendall tests were applied. Besides, for checking precipitation, temperature, relative humidity, wind speed, and solar radiation the Coupled Model Intercomparison Project phase 6 (CMIP6), and for investigation of runoff the Artificial Neural Network (ANN), and Soil and Water Assessment Tool (SWAT) models under the Shared Socio-economic Pathway scenarios (SSPs) used for the future period (2021-2050) compared to the base period (1985-2014). In addition, the Linear Scaling Bias Correction (LSBC) applied to downscale the CMIP6 data. Results: According to findings of Mann-Kendall and Sen’s slope estimator tests precipitation and relative humidity had the falling trend. This falling trend was significant only in minimum relative humidity at the level of 5%. The temperature trend also showed that minimum, average and maximum temperatures had a rising trend, and these rising trends were not significant in any of the mentioned parameters. The trend of changes in wind speed and solar radiation has also risen, and this rising trend was significant in wind speed with the Mann-Kendall's coefficient equals to 3.4 at the level of 1%. The process of changes in runoff in this area also showed that because of the changes in land use and vegetation, watershed's runoff had an increasing trend, but this increasing trend was not impressive in any of the hydrometric stations. The results of predicting the precipitation will have a decrease of 6.1%, while minimum and maximum temperatures will have an rise of 1.6 Cº and 1.4 Cº, respectively. The findings from SWAT model demonstrated that the surface runoff based on SSP5-8.5, SSP3-7.0, and SSP1-2.6 scenarios will have a decrease of 22.8% on average. In addition, investigating using ANN, precipitation is the most effective parameter in changing the runoff. Conclusion: Based on the rise in temperature, drop in soil moisture and happening the flash floods in the study area due to the climate change, the suitable management for this area is crucial. It is recommended to implement the sufficient watershed management operations such as structural and biological operations in the area in order to enhance the vegetation and mitigate the floods due to the climate change.