In Texas, earthquakes of magnitude 3.0 or over have increased eightfold since 2017. The Railroad Commission of Texas, the regulator, has increased its monitoring and oversight of injection with the goal of ‘no more magnitude 3.5 plus earthquakes after 18 months from the date of an RRC response action’ (est. date 2024). For the affected areas, operator-led response plans have been developed with expanded data collection efforts, contingency responses for future seismicity, and scheduled checkpoint updates with RRC staff.
Because the lag time between reductions in injection volume and reductions in seismicity is uncertain, the RRC has been studying changes to injection operations and subsurface pressure changes within the affected areas. New RRC rules call for more frequent reporting of injection volume and pressure data in areas of seismicity. RRC has teamed with the Texas Comptroller on the TexNet Injection Volume Reporting Tool to facilitate injection report filing and data accessibility.
The RRC has also turned to ‘artificial intelligence’ to speed the process of conducting its reviews of seismicity prior to the award of a permit for an injection/disposal well. The reviews are conducted by the RRC’s underground injection control department which has deployed a machine learning algorithm to process the large amount of information that needs to be digested. The UIC has developed a Python/scikit learn model to assess multiple factors related to the number, severity and proximity of earthquakes and uses a decision tree to assign a grade to the review. The assessed grade determines the allowable amount of fluid that can be injected.
Python scripts automatically collect GIS mapping of historic seismic events within a permit application’s area of interest, data which is used by the machine learning algorithm. The approach has reduced permitting time from over 100 days in November 2018 to about 20 days presently. At the same time decisions are more consistent in how seismic and other factors are considered. More from the RRC.
© Oil IT Journal - all rights reserved.