Groundwater level prediction based on GMS and SVR models under climate change conditions: Case Study-Talesh Plain

This study compared the capability of GMS and SVR models for groundwater modeling and evaluated the impact of climate change on future aquifer quantity in Talesh Plain. Groundwater level modeling was performed using GMS and SVR models for the period 2005-2018 (base period). Also, the effects of climate change on temperature and precipitation in the study area were estimated based on the HadGEM2-ES GCM model considering RCP 2.6, RCP 4.5, and RCP 8.5 emission scenarios in 2020-2034 (future period). A correlation of greater than 0.70 was found between the observed and estimated groundwater levels in both models. Moreover, in the base period, the average decline in groundwater level was 0.86 m. SVR model exhibited that the average groundwater level will drop by 0.94, 0.98, and 1.04 m in RCP2.6, RCP4.5, and RCP8.5 emission scenarios, respectively. While in the GMS modeling, under the same emission scenarios, these values were 0.91, 0.95, and 1.06 m, respectively. Moreover, the current trend of groundwater withdrawal may significantly increase the groundwater deficit and aquifer imbalance. It is therefore essential to apply artificial intelligence and mathematical models to accurately predict groundwater level fluctuations in this region to optimize groundwater management. Overall, our results revealed that SVR and GMS models perform almost similarly in simulating groundwater levels in the study area, suggesting that artificial intelligence can serve as a fast decision-making tool in groundwater management in similar aquifers.

Automated summary

This study compared the performance of GMS and SVR models in simulating groundwater levels and assessed the impact of climate change on aquifer quantity in Talesh Plain. Both models accurately predicted groundwater level fluctuations and indicated a continued decline in the future. The current trend of groundwater withdrawal may exacerbate the groundwater deficit and aquifer imbalance, highlighting the need for artificial intelligence and mathematical models in groundwater management.