Identification of heterogeneous aquifer transmissivity using an AE-based method.

Understanding aquifer properties is essential for managing groundwater resources and predicting contaminant movement. Hydraulic head data are widely used in groundwater flow modeling. However, transmissivity, a key aquifer property, often varies unpredictably. Existing studies have shown that transmissivity can be described statistically using moments like variance and correlation scales. Prior work has demonstrated that these statistical properties, combined with limited transmissivity measurements, can help estimate transmissivity at unmeasured locations. Despite this, direct estimation of transmissivity variograms remains challenging due to sparse data. This gap motivated the development of alternative methods that use more abundant head measurements. No prior work had resolved how head data could be used to infer transmissivity statistics. The lack of a reliable method to derive variogram parameters from head measurements remains a significant limitation in aquifer characterization.

The goal of this study is to develop a method for estimating transmissivity variogram parameters using hydraulic head data. Traditional approaches rely on transmissivity measurements, which are often limited. This paper proposes a new approach that uses head data instead. The method combines numerical simulations with theoretical models to infer variogram parameters. The study focuses on aquifers with high log-transmissivity variance, where traditional methods are less effective. The authors aim to demonstrate the feasibility of their approach in a real-world setting. They selected the Eagle Valley basin in Nevada, where extensive transmissivity and head data are available. By applying their method to this region, the researchers hope to validate its effectiveness. The ultimate purpose is to provide a practical tool for aquifer characterization when transmissivity data are scarce.

The methodology involves numerical simulations and theoretical modeling to estimate transmissivity variograms. The authors used the analytic element method for precise groundwater flow simulations. This method allows accurate representation of complex flow patterns in heterogeneous aquifers. The simulations were based on known head data and assumed transmissivity statistics. The researchers then compared simulated head distributions with observed data to infer variogram parameters. The process involved iterative adjustments to match simulated and observed head values. The method accounts for both correlated and uncorrelated transmissivity variations. Variogram parameters, including variance and correlation scale, were estimated using this combined simulation-theory framework.

The method successfully estimated transmissivity variogram parameters using head data. Simulated head distributions matched observed data within acceptable error margins. The estimated correlation scale was on the order of kilometers, consistent with prior studies. The uncorrelated transmissivity variance was also accurately captured. The method demonstrated robustness in aquifers with high log-transmissivity variance. Variogram parameters derived from head data showed good agreement with those from direct transmissivity measurements. The approach was validated in the Eagle Valley basin, where extensive data were available. The results suggest that head data can provide reliable estimates of transmissivity statistics when transmissivity measurements are limited.

The authors conclude that head data can be used to estimate transmissivity variogram parameters in heterogeneous aquifers. Their method combines numerical simulations with theoretical models to infer statistical properties of transmissivity. The approach was successfully applied to the Eagle Valley basin, where abundant head and transmissivity data were available. The estimated variogram parameters matched expected values from prior studies. The method is particularly useful in aquifers with high log-transmissivity variance, where traditional methods are less effective. The results suggest that head measurements can serve as a reliable proxy for transmissivity data. The authors propose that this method can improve aquifer characterization in data-scarce environments. The study highlights the potential of head data to support geostatistical modeling of aquifer properties.