Quantification of groundwater storage heterogeneity in weathered/fractured basement rock aquifers using electrical resistivity tomography: Sensitivity and uncertainty associated with petrophysical modelling

摘要

Quantifying groundwater storage in weathered/fractured basement rock aquifers can be challenging owing to both their high degree of heterogeneity and their overall low storage capacity. Therefore, in these aquifers, the use of direct borehole hydraulic data is usually insufficient. Here we assessed the popular method of electrical resistivity tomography (ERT), combined with borehole data and including associated uncertainties, to resolve the spatial variability of groundwater storage properties at high resolution within a fractured mica schist aquifer in Ireland. Porosity distributions across both the saturated and unsaturated zones were calculated from two-dimensional (2D) ERT resistivities using two standard petrophysical models, Archie and Waxman & Smits (WS), the latter accounting for the influence of clay minerals on resistivity data. Our results demonstrated the importance of the hydrogeological conceptual constraints provided by ERT when parametrizing the 2D petrophysical models from borehole point data. They also confirmed the importance of accounting for clay minerals (the products of bedrock weathering processes) in the WS model, whereas predictions from Archie's model produced unrealistically high porosity values of over an order of magnitude higher than the WS model. The WS model predicted porosities decreasing exponentially with depth, with values ranging from a few % in the shallowest, most-weathered part of the bedrock (upper 5 m on average) and deep fractured zones (to about 20 m deep), to <1% in the underlying fissured aquifer, and possibly down another order of magnitude in the deep massive bedrock. WS-derived porosities were in agreement with independent vertical water content profiles derived from magnetic resonance sounding (MRS), as well as point storativity values estimated from borehole hydraulic testing at the study site, with particularly good matches in the upper weathered/fractured bedrock and deeply weathered/fractured zones associated with regional faults. Detailed comparison suggested that WS provides an upper-bound estimate of groundwater storage in this environment. In the deep massive, un-weathered, and poorly fractured bedrock, however, discrepancies between groundwater storage estimate obtained from the three methods (ERT, MRS, and hydraulic) prevented reliable storage quantification, owing to the methods' inherent technical limitations in such low porosity rocks. Our results demonstrated the suitability of resistivity tomography to quantify groundwater storage heterogeneity in weathered/fractured basement rock aquifers at high resolution and with reasonable overall uncertainty given the relative high uncertainties in petrophysical parameters at the kilometric scale. The results are promising for better characterization of groundwater storage variations in these hydrogeological systems, which are crucial to predict their response to climate variability and human exploitation.