New Mexico Geological Society Annual Spring Meeting — Abstracts

×

Airborne electromagnetic (AEM) surveys provide subsurface resistivity imaging over large areas at relatively low cost compared to similarly scaled ground based geophysical or geological surveying, making them attractive tools for mapping groundwater basins in New Mexico. To utilize AEM surveys for hydrogeologic purposes, a scalable transform is needed to convert resistivity-depth models into lithologic and hydrogeologic interpretations. To construct this transform, we developed a quantitative resistivity–lithology calibration workflow that integrates AEM data-derived resistivity with co-located lithology information from a large database of legacy driller’s logs from the New Mexico Office of the State Engineer Water Rights Reporting System. Preliminary results from southwestern New Mexico, specifically in Hidalgo, Grant, and Luna County, show promising correlations between resistivity data and lithologic characteristics.

Legacy borehole logs (PDFs) from thousands of wells were compiled and processed using optical character recognition coupled with artificial-intelligence text-extraction and classification workflows. Lithologic descriptions were standardized into eleven classes. Across southwestern New Mexico, this workflow produced >20,000 classified lithologic intervals from 3,442 wells. Due to the impact of saturation on resistivity, separate transforms were developed for unsaturated and saturated conditions. Estimation of the water table in the region was done using two methods: (1) inverse-distance-weighted interpolation of trend-adjusted measured water levels; and (2) a constrained gradient-search method applied to AEM resistivity profiles guided by known water-level control points (Dewar and Knight, 2020). These surfaces were used to partition well-log intervals and AEM pixels into unsaturated and saturated populations.

The resistivity-to-lithology transforms were developed by treating lithologic intervals intersecting individual AEM pixels as resistors in parallel, such that pixel-scale bulk resistivity is represented by the thickness-weighted harmonic mean of component facies resistivities (Knight et al, 2018). Co-located well-log intervals and AEM pixels were assembled into systems of harmonic-mean mixing equations, with the resistivities of each distinct lithologic interval as the unknowns, and bootstrap resampling was used to estimate resistivity distributions for each lithologic class. Sensitivity to the definition of spatial co-location was evaluated using maximum well-to-AEM distances ranging from 50 to 1,000 m in 50 m increments (Kang et al., 2025).

Using a 350 m co-location threshold and interpolated water table, 1,565 well–AEM pairs yielded 2,260 mixing equations above the water table in southwestern New Mexico. Bootstrap analysis (n = 1,000) yielded distributions with unsaturated rock as the most resistive of the facies categories (geometric mean = 97 Ωm), whereas clay/fine and shale/fine are the least resistive (16 and 15 Ωm, respectively). Coarse-grained sediment exhibits an intermediate but distinct geometric mean of 30 Ωm. Below the water table, all facies shift to lower resistivity while preserving their relative separation (Figure 1). The geometric mean of distributions of resistivity for selected lithologic classes include rock (41 Ωm), coarse-grained (18 Ωm), clay/mixed (13 Ωm), and clay/fine and shale/fine (9 Ωm). At the 350-m threshold, distributions for the principal hydrogeologic classes of interest were well separated.

These results highlight the effectiveness of combining legacy well logs with robust AEM calibration to provide a scalable framework for the transformation of AEM resistivity sections into lithologic probability models and provides a foundation for regional groundwater mapping and aquifer characterization across New Mexico.

References:

1. Dewar, N., & Knight, R. (2020). Estimation of the top of the saturated zone from airborne electromagnetic data. Geophysics, 85(5), EN63-EN76. https://doi.org/10.1190/geo2019-0539.1
2. Kang, S., Goebel, M., & Knight, R. (2025). Harnessing the power of geophysical imaging to recharge California's groundwater. Earth and Space Science, 12(4), e2024EA003958. https://doi.org/10.1029/2024EA003958
3. Knight, R., Smith, R., Asch, T., Abraham, J., Cannia, J., Viezzoli, A., & Fogg, G. (2018). Mapping aquifer systems with airborne electromagnetics in the Central Valley of California. Groundwater, 56(6), 893-908. https://doi.org/10.1111/gwat.12656