Researcher Nourani from Near East University, in collaboration with other scholars, conducted a study to assess the long-term impacts of climate change on groundwater fluctuations in the Ardabil plain, Iran. Utilizing Global Climate Models (GCMs) data from the Coupled Model Intercomparison Project (CMIP6) and the Shared Socioeconomic Pathway 5-8.5 (SSP5-8.5) scenario, the study projected a potential increase in mean annual temperature by 0.8°C per decade until 2100, alongside an anticipated 8% decrease in mean precipitation compared to the base period. Employing Machine Learning (ML) techniques, including Feedforward Neural Network (FFNN) and ensemble modeling, the research aimed to understand the complex interactions between climate variables and groundwater levels.
Results indicated a decline in groundwater levels across the Ardabil plain, with an average drop of 6.62 meters from 1997 to 2014. Through clustering analysis and model simulations, spatial variations in groundwater responses to climate change were observed. Notably, models incorporating lagged groundwater levels and climate variables demonstrated higher accuracy than those solely reliant on climate parameters. Furthermore, ensemble modeling revealed a 6% increase in predictive accuracy compared to individual ML models, highlighting the effectiveness of collaborative approaches in mitigating uncertainties.
The study emphasized the need for sustainable water management practices in light of declining groundwater levels, attributing the phenomenon primarily to overexploitation alongside notable climate change impacts. As groundwater recharge rates remain limited in the Ardabil plain, the researchers underscored the importance of stricter regulations on groundwater discharge to mitigate further declines. They also advocated for increased attention to climate change considerations in water management strategies.
Looking ahead, the researchers suggested exploring the impact of climate change on the spatiotemporal variability of groundwater quality to enhance understanding and inform long-term management strategies. By bridging the gap between climate change projections and groundwater dynamics, this study offers valuable insights applicable not only to the Ardabil plain but also to similar arid and semi-arid regions worldwide, contributing to informed decision-making in water resource management.
More Information:
https://www.sciencedirect.com/science/article/pii/S0301479723004413
