Evaluating the impact of supplementary irrigation on the WRSI index of rainfed wheat under climate change scenarios in Tabriz, Iran using FLF-LSTM and LARS-WG models | ||
| Egyptian Journal of Agronomy | ||
| Volume 47, Issue 4, December 2025, Pages 1023-1037 PDF (1.54 M) | ||
| Document Type: Original Article | ||
| DOI: 10.21608/agro.2025.384725.1695 | ||
| Authors | ||
| Vahid Mouneskhah; Saeed Samadianfard* ; Abolfazl Majnooni-heris | ||
| Water Engineering Department, Faculty of Agriculture, University of Tabriz, Tabriz, Iran | ||
| Abstract | ||
| The aim of this study is to investigate the effect of climate change on the Water Requirement Satisfaction Index (WRSI) for rainfed wheat in Tabriz during the future time period up to 2100. The LARS-WG statistical model was used to generate future climate data, which was calibrated and run with CNRM-CM6-1 and MPI-ESM1-2-LR, under three SSP126, SSP245, and SSP585 scenarios. These models were selected due to their high ability to reconstruct the region's climate data and are able to simulate changes in temperature and precipitation. Subsequently, the Long short-term memory (LSTM) and Forex Loss Function-LSTM (FLF-LSTM), as deep learning models, were used to predict reference evapotranspiration (ET₀) values. The LSTM model, as a recurrent neural network-based model, has the ability to identify complex patterns in climate time series. Moreover, the FLF-LSTM model, using a composite loss function, provided more accurate performance compared to the classic LSTM. The error metrics reveal that the FLF-LSTM model outperformed the standalone LSTM in terms of accuracy and reliability, with a root mean square error of 0.71 and a mean bias error of 0.06 during the test period. Additionally, examination of the WRSI in the future time periods and considering climate change scenarios showed that WRSI will have a downward trend under the influence of increased temperature and decreased precipitation. This means an increase in water stress in different growth stages of rainfed wheat. However, the use of supplementary irrigation in sensitive growth stages reduce the negative impact of climate change. | ||
| Keywords | ||
| Climate change; Deep learning models; Rainfed wheat; Supplementary irrigation; WRSI | ||
|
Statistics Article View: 5 PDF Download: 4 |
||
