Interrogating the model space of airborne electromagnetic inversion to answer hydrogeologic questions
Abstract
New approaches are needed for the management of groundwater resources in the Central Valley of California in order to achieve the legislative requirement of groundwater sustainability. The airborne electromagnetic (AEM) method can play an important role in mapping out the large-scale hydrogeological structure of the subsurface. The acquired AEM data are first inverted to obtain an electrical resistivity distribution of the subsurface. With the use of a rock physics transform, information about the lithologic distribution can be obtained. Through integration with well data, a hydrogeologic conceptual model can be built that can be used to address critical questions related to groundwater management. However, uncertainties in inverting and interpreting AEM data can be significant due to the non-uniqueness in AEM inverse problems. Therefore, exploring the range of possible resistivity models that can explain the AEM data is an essential step in capturing the uncertainty that can impact the derived lithology models.
In this study, we focus on three important hydrogeologic questions posed by the local water agency in Butte and Glenn counties, CA: (1) What is the connectivity between the upper aquifer and surface water? (2) What is the connectivity between the upper and lower aquifers? (3) What is the spatial variation in the integrated clay fraction, to be used as an indicator of subsidence risk? To address these questions, we develop a workflow that can extract an ensemble of lithology models from both AEM data and existing well data including electrical logs (E-logs) and lithology logs. First, we invert SkyTEM data obtained in Butte and Glenn counties in 2018 to obtain resistivity models. Here, we use various regularization functions (e.g. reference model) to obtain an ensemble of resistivity models with the incorporation of high-quality E-logs available at the region; ~152 within 870 m of AEM soundings. By using co-located E-logs and lithology logs (~21), we then construct relationships between resistivity and lithology. This allows us to obtain ensemble of lithology models displaying the probability of each lithologic unit: clay/silt, sand, and gravel. Using the obtained lithology models, we demonstrate the capability and limitation of using the AEM-derived lithology information to answer the three questions.- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2019
- Bibcode:
- 2019AGUFMNS44A..02K
- Keywords:
-
- 0933 Remote sensing;
- EXPLORATION GEOPHYSICS;
- 1829 Groundwater hydrology;
- HYDROLOGY;
- 1835 Hydrogeophysics;
- HYDROLOGY;
- 1880 Water management;
- HYDROLOGY