In this research, we critically examine the impact of equivalence on the accuracy of unconstrained inverse resistivity models derived from 2D Electrical Resistivity Tomography (ERT) surveys targeting horizontal conductive zones (HCZs), associated with horizontally extending aquifers. Using forward and inverse modelling techniques, we aimed to assess how well these models retrieve the true thicknesses and resistivities of HCZs in various scenarios, including different depths and resistivity contrasts in relation to host formations.
Methods and Findings
We used Res2DMod to construct forward models of horizontal aquifers under varying conditions, such as depths and resistivity contrasts with their surrounding host rocks. These forward models were then inverted using Res2DInv with both coarse and fine grids to produce inverse resistivity models. Our results consistently demonstrated that unconstrained inversion tends to overestimate both the thickness and resistivity of HCZs, particularly when the zones are deeper or have a higher resistivity contrast compared to the host rock.
The issue of equivalence, a well-known problem in geophysics where different combinations of thickness and resistivity yield similar apparent resistivity data, significantly affected the result interpretation. The inverse models often failed to accurately represent the true structure dimensions of the HCZs, highlighting the non-uniqueness of the solutions. Sensitivity analyses indicated that the boundary depths were poorly defined, especially in deeper models, due to the reduced sensitivity caused by near-surface conductive layers.
Implications and Benefits
Our study underscores a critical implication for groundwater investigations: resistivity models obtained from unconstrained 2D ERT surveys should be treated with caution, especially when used to estimate aquifer properties such as thickness and resistivity. Overestimations could lead to incorrect assessments of groundwater resources, potentially impacting decisions on water extraction, resource management, and environmental sustainability.
However, understanding the limitations of 2D ERT surveys also brings significant benefits. By recognizing the effects of equivalence and non-uniqueness, geophysicists and hydrogeologists can better design their surveys and data interpretations. Our findings advocate for the integration of complementary investigation methods, such as drilling and other geophysical techniques, to validate ERT models and reduce the uncertainties associated with inverse modelling. This approach ultimately enhances the reliability of groundwater assessments and supports more informed management of subsurface water resources.
While 2D ERT remains a valuable tool for imaging subsurface conductive zones, the results from unconstrained inversions should be carefully validated. Our research provides essential insights that can guide future survey designs and interpretations, contributing to improved accuracy in hydrogeophysical studies of horizontal aquifers.
Authors: F.D. Fourie and J.W. Haumann, Institute for Groundwater Studies, University of the Free State, South Africa.
