A hybrid vine copula-fuzzy model for groundwater level simulation under uncertainty

Area of Science:

• Hydrology and Water Resource Management
• Environmental Modeling
• Statistical Hydrology

Background:

• Groundwater level simulation is vital for sustainable water resource management.
• Uncertainties in input data, model parameters, and physical processes challenge accurate groundwater modeling.
• Variable dependencies within hydrological systems complicate simulation efforts.

Purpose of the Study:

• To develop a novel hybrid model integrating copula theory and fuzzy logic for monthly groundwater level simulation.
• To account for both dependency structures among input variables and model uncertainty.
• To enhance groundwater level simulation in the Najafabad aquifer's Nekouabad regions.

Main Methods:

• Utilized C-vine copula structures to model dependencies between precipitation, temperature, surface water, and discharge.
• Employed conditional simulation for deterministic groundwater level simulation.
• Incorporated fuzzy logic by defining α-cuts for copula parameters to handle parameter uncertainty.

Main Results:

• The deterministic model achieved high accuracy, with R-squared values of 0.88 (Nekouabad-Right) and 0.83 (Nekouabad-Left).
• The copula-fuzzy model generated a range of groundwater levels with varying confidence intervals, reflecting parameter uncertainties.
• The model successfully simulated groundwater levels by considering input variable dependencies and possibilistic uncertainty.

Conclusions:

• The developed vine copula-fuzzy model offers a flexible approach to uncertainty assessment in groundwater level simulation.
• Decision-makers can select confidence intervals based on membership degrees, aiding tailored water resource planning.
• This hybrid model provides a robust framework for managing water resources under uncertainty.