Open AccessExploring the potential linkages between oil‐field brine reinjection, crystalline basement permeability, and triggered seismicity for the Dagger Draw Oil field, southeastern New Mexico, USA , using hydrologic modeling
Author(s) -
Zhang Y.,
Edel S. S.,
Pepin J.,
Person M.,
Broadhead R.,
Ortiz J. P.,
Bilek S. L.,
Mozley P.,
Evans J. P.
Publication year - 2016
Publication title -
geofluids
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.44
H-Index - 56
eISSN - 1468-8123
pISSN - 1468-8115
DOI - 10.1111/gfl.12199
Subject(s) - geology , induced seismicity , hydraulic fracturing , permeability (electromagnetism) , brine , wellhead , oil field , caprock , petrology , thermal diffusivity , subsea , structural basin , geomorphology , geochemistry , seismology , petroleum engineering , geotechnical engineering , quantum mechanics , physics , thermodynamics , membrane , biology , genetics
We used hydrologic models to explore the potential linkages between oil‐field brine reinjection and increases in earthquake frequency (up to M d 3.26) in southeastern New Mexico and to assess different injection management scenarios aimed at reducing the risk of triggered seismicity. Our analysis focuses on saline water reinjection into the basal Ellenburger Group beneath the Dagger Draw Oil field, Permian Basin. Increased seismic frequency (>M d 2) began in 2001, 5 years after peak injection, at an average depth of 11 km within the basement 15 km to the west of the reinjection wells. We considered several scenarios including assigning an effective or bulk permeability value to the crystalline basement, including a conductive fault zone surrounded by tighter crystalline basement rocks, and allowing permeability to decay with depth. We initially adopted a 7 m (0.07 MP a) head increase as the threshold for triggered seismicity. Only two scenarios produced excess heads of 7m five years after peak injection. In the first, a hydraulic diffusivity of 0.1 m 2 s −1 was assigned to the crystalline basement. In the second, a hydraulic diffusivity of 0.3 m 2 s −1 was assigned to a conductive fault zone. If we had considered a wider range of threshold excess heads to be between 1 and 60 m, then the range of acceptable hydraulic diffusivities would have increased (between 0.1–0.01 m 2 s −1 and 1–0.1 m 2 s −1 for the bulk and fault zone scenarios, respectively). A permeability–depth decay model would have also satisfied the 5‐year time lag criterion. We also tested several injection management scenarios including redistributing injection volumes between various wells and lowering the total volume of injected fluids. Scenarios that reduced computed excess heads by over 50% within the crystalline basement resulted from reducing the total volume of reinjected fluids by a factor of 2 or more.