Recharge site assessment through the integration of surface geophysics and cone penetrometer testing

Abstract

The ability to identify, at potential managed aquifer recharge sites, the presence of connected pathways of hydraulically conductive sediments from the ground surface to the water table could help minimize costs and risks associated with recharge operations. A spatially dense dataset had previously been acquired in an almond [Prunus dulcis (Mill.) D.A. Webb] grove in Tulare, CA, using tTEM, a towed transient electromagnetic (tTEM) geophysical method. In order to interpret reliable information about sediment type from the tTEM data, a transform from the tTEM-derived property, electrical resistivity, to sediment type is required. The uncertainty associated with derived models of sediment type can be significantly reduced if a site-and datasetspecific transform is used. Cone penetrometer testing (CPT) was conducted at five locations, strategically selected based on a review of the tTEM data. Co-located measurements of sediment type, derived from the CPT, and electrical resistivity, derived from the tTEM data, were used to create a resistivity-to-sediment-type transform, with sediment type classified as either coarse-grain-dominated (sand and gravel) or fine-grain-dominated (silt and clay) material. The transform captured the uncertainty associated with variable water salinity and content, the resolution of the tTEM data, and other components of the tTEM measurement workflow. Using this transform, models of sediment type were generated for the unsaturated zone at the site. Within these models are features, which we interpret as potential recharge pathways, corresponding to high fractions of coarse-grain-dominated material amongst regions of fine-grain-dominated material. The workflow developed at this site can provide a framework for using tTEM and CPT for recharge site assessment.

Managed aquifer recharge site assessment with electromagnetic imaging: Identification of recharge flow paths

“…Vertical discretization of the resistivity models begins at ∼1 m at the surface and gradually increases to 6-12 m by the depth of investigation, which was between 50 and 70 m; the depth of investigation is considered to be the depth to which the estimates of electrical resistivity are reliable and depends on the subsurface electrical resistivity. Using measurements of sediment behavior, acquired using cone-penetrometer testing, the resistivity data in the vadose zone from one of the sites were transformed to the fraction of coarse-grain-dominated material, referred to as coarse-dominated material (i.e., material consisting mainly of sands and gravels), by empirically determining a resistivity-to-sediment-type transform (Goebel & Knight, 2021).…”

“…The preferential flow path depicted by the dashed white line was manually selected using the spatial distribution of units with a high fraction of coarse-dominated material. Figure modified from Goebel and Knight (2021) In the study by Goebel and Knight (2021), the sparse 3D model displaying the spatial variation in the fraction of coarse-dominated material was qualitatively interpreted to determine where there appeared to be preferential flow paths from the surface to the water table at a site. An example is shown in Figure 1 (modified from Goebel and Knight [2021]), where the authors identified a potential connected pathway of coarse-dominated material through the subsurface.…”

“…Figure modified from Goebel and Knight (2021) In the study by Goebel and Knight (2021), the sparse 3D model displaying the spatial variation in the fraction of coarse-dominated material was qualitatively interpreted to determine where there appeared to be preferential flow paths from the surface to the water table at a site. An example is shown in Figure 1 (modified from Goebel and Knight [2021]), where the authors identified a potential connected pathway of coarse-dominated material through the subsurface. This qualitative interpretation required manually scrolling through the data to identify these pathways and relied on visual interpolation of sediment type in vertical sections lacking tTEM data.…”

“…The preferential flow path depicted by the dashed white line was manually selected using the spatial distribution of units with a high fraction of coarse‐dominated material. Figure modified from Goebel and Knight (2021)…”

Exploring the Model Space of Airborne Electromagnetic Data to Delineate Large‐Scale Structure and Heterogeneity Within an Aquifer System

“…There is an alternate way to transform the resistivity model to obtain information about the composition of the subsurface, and that is to capture information about the percentage of each sediment type that is present. Such an approach was recently taken in the interpretation of ground‐based transient EM data (Goebel & Knight, 2021) and is directly applicable to AEM data. In this approach, it is assumed that there is finer‐scale layering of sediment type, below the scale of the AEM resistivity cell.…”