Geospatial statistics elucidate competing geological controls on natural CO₂ seeps in Italy

Site selection for the geological storage of CO ₂ for long timespans requires an understanding of the controls on containment, migration, and surface seepage of subsurface CO ₂ fluids. Evidence of natural CO ₂ migration from depth to the surface is documented at 270 sites from Italy, a prolific CO ₂ province. Previous studies indicate that CO ₂ delivery to and from buried structures that host CO ₂ accumulations is fault controlled but competing controls on the CO ₂ flow pathways affect the location and style of CO ₂ release. Here, we conduct a meta-analysis using a novel geospatial approach to statistically determine the relationship between the geological setting and structures and the CO ₂ seep spatial distribution and characteristics (morphological type, flux, and temperature) in Central Italy. We find that seep distribution differs on two spatial scales corresponding to the geological setting. On large scales (>5 km), seeps are isotropically distributed and align with regional structures such as anticlines, decollements, and extensional faults. On local scales (<5 km), seeps cluster and align with subsidiary geologic structures, including faults and lithological boundaries. The detailed location and flux of seeps within clusters are influenced by the regional structural domain: in the Tyrrhenian, seeps tend to be located along fault traces, whereas seeps are located as springs in the tip and ramp regions of fault scarps in the Apennines. Thus, our geospatial approach evidences, at a regional scale, how macrocrustal fluid flow is governed by deep extensional and compressional features but once CO ₂ reaches shallower structures, it evidences how smaller scale features and hydrogeological factors distribute the CO ₂ fluids in the near surface, dependent on the geological setting. This work not only demonstrates useful application of a novel geospatial approach to characterize competing crustal controls on CO ₂ flow at different scales but also informs the design of appropriate site characterization and surface monitoring programs at engineered carbon stores.