Exploring the capabilities of electrical resistivity tomography to study subsea permafrost

• Published in: The Cryosphere
• Main Authors: M. Arboleda-Zapata, M. Angelopoulos, P. P. Overduin, G. Grosse, B. M. Jones, J. Tronicke
• Format: Article in Journal/Newspaper
• Language: English
• Published: Copernicus Publications 2022
• Subjects: Environmental sciences; GE1-350; Geology; QE1-996.5; Arctic; Ice; permafrost; The Cryosphere
• Online Access: https://doi.org/10.5194/tc-16-4423-2022; https://doaj.org/article/6416180a7ade404a8dde4136e7ebb6ea

Sea level rise and coastal erosion have inundated large areas of Arctic permafrost. Submergence by warm and saline waters increases the rate of inundated permafrost thaw compared to sub-aerial thawing on land. Studying the contact between the unfrozen and frozen sediments below the seabed, also known as the ice-bearing permafrost table (IBPT), provides valuable information to understand the evolution of sub-aquatic permafrost, which is key to improving and understanding coastal erosion prediction models and potential greenhouse gas emissions. In this study, we use data from 2D electrical resistivity tomography (ERT) collected in the nearshore coastal zone of two Arctic regions that differ in their environmental conditions (e.g., seawater depth and resistivity) to image and study the subsea permafrost. The inversion of 2D ERT data sets is commonly performed using deterministic approaches that favor smoothed solutions, which are typically interpreted using a user-specified resistivity threshold to identify the IBPT position. In contrast, to target the IBPT position directly during inversion, we use a layer-based model parameterization and a global optimization approach to invert our ERT data. This approach results in ensembles of layered 2D model solutions, which we use to identify the IBPT and estimate the resistivity of the unfrozen and frozen sediments, including estimates of uncertainties. Additionally, we globally invert 1D synthetic resistivity data and perform sensitivity analyses to study, in a simpler way, the correlations and influences of our model parameters. The set of methods provided in this study may help to further exploit ERT data collected in such permafrost environments as well as for the design of future field experiments.