--> Using InSAR Surface Deformation Measurements to Study the Potential Link Between Industry Operations and Earthquakes in the Delaware Basin of West Texas
[First Hit]

2019 AAPG Annual Convention and Exhibition:

Datapages, Inc.Print this page

Using InSAR Surface Deformation Measurements to Study the Potential Link Between Industry Operations and Earthquakes in the Delaware Basin of West Texas

Abstract

There has been a noticeable increase in Previous HitseismicityNext Hit in the Delaware Basin in the past few years. Industry-related fluid injection and extraction can induce earthquakes by altering the pore pressure and stress acting on pre-existing faults. When production and injection occur in close proximity as is the case in the Delaware Basin, it can be difficult to unravel the potential contribution of either activity to fault destabilization. Measurable surface deformation surrounding wells can sometimes be a result of industry operations. Thus, one promising method to study the underlying mechanisms is to use surface deformation in combination with production and disposal data to model changes in pressure and stress in the subsurface. Interferometric synthetic aperture radar (InSAR) uses phase measurements between multiple passes of a satellite to estimate mm-scale deformation over wide areas at high spatial resolution (~5-20m) and repeat frequency (6-12 days). Shirzaei et al. (Science, 2016) showed that InSAR displacement measurements can be used to estimate volume strain rates in the subsurface, which can be related to pore pressure and stress changes. Analyzing key patterns linking estimated pore pressure and stress changes to observed Previous HitseismicityNext Hit may shed light on the mechanisms leading to Previous HitinducedNext Hit earthquakes in the Delaware Basin. As a preliminary investigation, we processed 72 descending Sentinel-1A scenes over the basin, spanning from Dec. 2014 through Jul. 2018. A weighted stack of 1395 interferograms with a maximum temporal baseline of 400 days reveals regions undergoing both uplift and subsidence near Pecos, TX. We compared earthquake locations from the TexNet array with the deformation signals, highlighting three categories of deformation/Previous HitseismicityNext Hit relationships: 1) deformation associated with high levels of Previous HitseismicityNext Hit, 2) little to no deformation associated with Previous HitseismicityNext Hit, and 3) deformation associated with little to no Previous HitseismicityNext Hit. At some locations, seismic activity follows spatially linear subsidence patterns, with one interpretation suggesting slowly-accumulating slip along optimally-oriented normal faults. Further work will include integrating production and injection fluid volumes, InSAR time-series, and a local geologic model to model stress changes in sub-regions of the identified categories. Model predictions will then be compared to the observed Previous HitseismicityTop.