arising from Cesca et al. Nature Communications https://doi.org/10.1038/s41467-021-24949-1 (2021)
A recent publication by Cesca et al.1 reanalyzes and expands seismic data to identify hypocenters of observed seismicity induced by the Castor Underground Gas Storage (UGS) operations. Their results confirm those of previous studies2,3 that earthquakes occurred below the storage formation on a fault dipping opposite from the Amposta fault, which bounds the reservoir. However, two important sets of disagreements require revising the conclusions by Cesca et al.1: the depth of hypocenters and the processes leading to seismicity. Inaccurate estimates of hypocenters location and partial consideration of the physical mechanisms that induce seismicity may imply endangering future deep underground projects.
Cesca et al.1 located the seismicity at 3 km depth, within sedimentary deposits. We consider this depth highly unlikely for two reasons. First, moderate to large earthquakes nucleate at larger depths, not in shallow formations4,5. Besides, moderate earthquakes rarely occur in sedimentary rocks where failure tends to be ductile because of their moderate strength. That is, the rock is crushed and deformation becomes plastic, i.e., aseismic. In agreement with these two considerations, the observed magnitude of induced earthquakes positively correlates both with depth and rock strength (see compilation of induced events in sedimentary and crystalline rock by Evans et al.6 and McGarr7). The magnitude of the largest induced earthquakes at Castor (M4.1) is far too high for a depth of 3 km in sedimentary rock as suggested by Cesca et al.1.
Beyond the above “qualitative” reasoning, our second set of reasons against a shallow seismicity at the Castor UGS refers to the adopted velocity model and the importance of the seismic velocities of the region to accurately determine focal depths. As also illustrated by Cesca et al.1 in their Fig. 4d, estimated depths can vary several kilometers depending on the adopted velocity model. Of all the analyzed velocity models, Cesca et al.1 have chosen the one that yields the shallowest earthquake locations. Similarly, Juanes et al.8 obtained a detailed 3D velocity model of the region but chose a simple 1D model that provided the shallowest locations, which allowed explaining the induced seismicity by pore pressure buildup. Instead, Villaseñor et al.2 used local seismic data and well log information to derive a velocity model for the region. Using this model and waveform inversion, they obtained focal depths that range from 4 to 10 km. Some of the uncertainty on focal depth results would have been reduced if an adequate monitoring network had been in place. However, the available network lacks stations near the injection site, with the closest station located more than 20 km away. Such a poor array has caused the current debate to be centered on the depth of induced earthquakes, rather than on determining their cause.
Understanding the seismicity triggering mechanisms is important to prevent similar accidents in the future (not to mention to recover the public confidence on scientists and engineers). Cesca et al.1 suggest that the seismicity at Castor was induced by pore pressure buildup. We disagree with this hypothesis for several reasons. First, the high permeability of the storage formation resulted in a low pore pressure buildup at the injection depth, not exceeding 0.8 MPa according to the company. Second, this buildup could not be transferred 1 km below. Cesca et al.1 assume a hydraulic diffusivity of 0.5 m2/s. This value is two to three orders of magnitude larger than what should be expected at the sedimentary formations beneath the Castor reservoir, where nearby wells show alternations of high- (sandstones and limestones) and low-permeability (shales and evaporites) rock layers below the depth of the storage formation9. Hydraulic connection between the storage formation and the depth of the earthquakes requires the existence of some high-permeability conduit or fault. While such faults may occur in fragile crystalline rocks under favorable stress conditions, it is unlikely that they display significant permeability in the soft rock below the site. For a more reasonable, albeit highly uncertain, diffusivity, D, of 10−3 m2/s, it would take some 30 year for a small pressure to reach 1 km below the storage formation (t = r2⁄D, where t is time and r is distance). Finally, the largest earthquakes occurred 20 days after an injection of 15 days. If diffusivity had been sufficiently large for a significant pressure buildup propagation at depth, it should have also caused overpressure to dissipate by the time of the largest earthquakes.
While pressure build-up is the traditional explanation for mechanical destabilization10, numerous other mechanisms can intervene (Fig. 1), including thermal stresses11, deformation-induced pressure changes12, and shear slip stress transfer13. Vilarrasa et al.3 suggested that the delayed seismicity at Castor UGS was driven by the combination of three mechanisms: buoyancy, stress transfer driven by aseismic slip of the Amposta fault, and fluid flow to regions destabilized by deformation but maintained stable by deformation-induced pressure drop. An important issue about the triggering mechanism, which ultimately motivates this comment, is that we strongly disagree with two of the statements made in Cesca et al.1 The authors state “Another reason for the low seismogenic potential of gas storage operations could be that gas injection schedules are typically designed and engineered not to exceed the stress conditions existing prior to or during the original reservoir production”. We contend that in extensional stress settings, like the Valencia Trough, the injection schedule or the stress computed with conventional models are not key factors. Instead, seismicity is activated by the vertical stress changes caused by buoyancy. Archimedes’ principle states that the buoyancy force equals volume multiplied by the density of the displaced fluid (actually density contrast, when concerned with changes in stress). Therefore, the key destabilizing factor in extensional settings is the volume of the storage facility, which is not accounted for in models such as the one by Cesca et al.1 and, we fear, in the design of most deep-injection operations. The need for full poroelastic models and for proper integration of monitoring data has been argued for long (e.g., Juanes et al.14). In this sense, the second statement “Using… methods… allows us to reconstruct… proving that this is possible despite the lack of a dense local network” is questionable at best. Comparing the relocation and focal mechanisms presented now with those calculated by Cesca et al.15, it is clear that it has taken a lot of time and effort to reach a conclusion we do not share. Worse, this conclusion might be construed to interpret that sophisticated inversion might overcome the limitations of an inadequate monitoring network. We contend that one of the lessons learnt from Castor is just the opposite.
Physics-based modeling is a useful tool not only for understanding failures such as that at Castor UGS, but also for forecasting and eventually managing induced seismicity. The goal of doing it in real time can only be reached if a thorough monitoring system and decision-making protocol work properly. This is especially important at this stage in which several low-carbon geo-energy technologies compete for deep pore space: geothermal energy, geologic carbon storage, and subsurface hydrogen storage, all identified as essential to meet net-zero carbon emissions.
Data availability
The open access fully coupled thermo-hydro-mechanical code that we have used for our simulations can be downloaded at https://deca.upc.edu/en/projects/code_bright. The dataset used for our interpretation of the triggering mechanisms at Castor can be accessed at https://digital.csic.es/handle/10261/216863, with https://doi.org/10.20350/digitalCSIC/12553.
References
Cesca, S. et al. Seismicity at the Castor gas reservoir driven by pore pressure diffusion and asperities loading. Nat. Commun. 12, 4783 (2021).
Villaseñor, A., Herrmann, R. B., Gaite, B. & Ugalde, A. Fault reactivation by gas injection at an underground gas storage off the east coast of Spain. Solid Earth 11, 63–74 (2020).
Vilarrasa, V., De Simone, S., Carrera, J. & Villaseñor, A. Unraveling the Causes of the Seismicity Induced by Underground Gas Storage at Castor, Spain. Geophys. Res. Lett. 48, e2020GL092038 (2021).
Das, S. & Scholz, C. H. Why large earthquakes do not nucleate at shallow depths. Nature 305, 621–623 (1983).
Verdon, J. P., & Bommer, J. J. Comment on “Activation rate of seismicity for hydraulic fracture wells in the Western Canadian Sedimentary Basin” by Ghofrani and Atkinson. Bulletin of the Seismological Society of America, https://doi.org/10.1785/0120200350 (2021).
Evans, K. F., Zappone, A., Kraft, T., Deichmann, N. & Moia, F. A survey of the induced seismic responses to fluid injection in geothermal and CO2 reservoirs in Europe. Geothermics 41, 30–54 (2012).
McGarr, A. Maximum magnitude earthquakes induced by fluid injection. J. Geophys. Res.: Solid Earth 119, 1008–1019 (2014).
Juanes, R. et al. Coupled flow and geomechanical modeling, and assessment of induced seismicity, at the Castor Underground Gas Storage Project. Final Report, April 2017. Retrieved from http://www.minetad.gob.es/es-ES/GabinetePrensa/NotasPrensa/2017/Paginas/informe-castor20170503.aspx (2017).
Fernendez, M., Torné, M. & Zeyen, H. Modelling of thermal anomalies in the NW border of the Valencia Trough by groundwater convection. Geophys. Res. Lett. 17, 105–108 (1990).
Ellsworth, W. L. Injection-induced earthquakes. Science 341, 6142 (2013).
Parisio, F., Vilarrasa, V., Wang, W., Kolditz, O. & Nagel, T. The risks of long-term re-injection in supercritical geothermal systems. Nat. Commun. 10, 1–11 (2019).
Cocco, M. & Rice, J. R. Pore pressure and poroelasticity effects in Coulomb stress analysis of earthquake interactions. J. Geophys. Res.: Solid Earth 107, ESE-2 (2002).
De Simone, S., Carrera, J. & Vilarrasa, V. Superposition approach to understand triggering mechanisms of post-injection induced seismicity. Geothermics 70, 85–97 (2017).
Juanes, R. et al. Were the May 2012 Emilia‐Romagna earthquakes induced? A coupled flow‐geomechanics modeling assessment. Geophys. Res. Lett. 43, 6891–6897 (2016).
Cesca, S. et al. The 2013 September–October seismic sequence offshore Spain: a case of seismicity triggered by gas injection? Geophys. J. Int. 198, 941–953 (2014).
Acknowledgements
V.V. acknowledges funding from the European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Programme through the Starting Grant GEoREST (www.georest.eu) under grant agreement No. 801809. A.V. acknowledges funding from Spanish Ministry of Science and Innovation grant CGL2017-88864-R. IDAEA-CSIC and ICM-CSIC are Centres of Excellence Severo Ochoa (Spanish Ministry of Science and Innovation Grants CEX2018-000794-S and CEX2019-000928-S, respectively, funded by MCIN/AEI/ 10.13039/501100011033).
Author information
Authors and Affiliations
Contributions
V.V., S.D.S., J.C. and A.V. contributed to the development of this matters arising and were involved in revising it.
Corresponding author
Ethics declarations
Competing interests
The authors declare no competing interests.
Peer review
Peer review information
Nature Communications thanks the anonymous reviewer(s) for their contribution to the peer review of this work.
Additional information
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Vilarrasa, V., De Simone, S., Carrera, J. et al. Multiple induced seismicity mechanisms at Castor underground gas storage illustrate the need for thorough monitoring. Nat Commun 13, 3447 (2022). https://doi.org/10.1038/s41467-022-30903-6
Received:
Accepted:
Published:
DOI: https://doi.org/10.1038/s41467-022-30903-6
This article is cited by
-
Reservoir impoundment-triggered seismicity in Brazil: the case of M4.0 Nova Ponte earthquake
Scientific Reports (2023)
-
The physical mechanisms of induced earthquakes
Nature Reviews Earth & Environment (2023)
-
In Situ Stress Evolution and Fault-Slip Tendency Assessment of an Underground Gas Storage Reservoir in the Turpan Basin (China): In Situ Stress Measurements and Coupled Simulations
Rock Mechanics and Rock Engineering (2023)
-
Reply to: Multiple induced seismicity mechanisms at Castor underground gas storage illustrate the need for thorough monitoring
Nature Communications (2022)
Comments
By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.