IDENTIFYING SEISMIC RISK IN THE APPALACHIAN BASIN GEOTHERMAL PLAY FAIRWAY ANALYSIS PROJECT USING POTENTIAL FIELDS, SEISMICITY, AND THE WORLD STRESS MAP Frank Horowitz Cornell University 3 November 2015 The information, data, or work presented herein was funded in part by the Office of Energy Efficiency and Renewable Energy (EERE), U.S. Department of Energy, under Award Number DE-EE0006726.

The GPFA-AB Team • Project Partners

•  Cornell University: Terry Jordan (PI), Frank Horowitz, Jery Stedinger, Jefferson Tester, Erin Camp, Calvin Whealton, Jared Smith

•  Southern Methodist University: Maria Richards (Lead), Cathy Chickering Pace, Matt Hornbach, Zachary Frone, Christine Ferguson, Rahmi Bolat, Maria Beatrice Magnani

•  West Virginia University: Brian Anderson (Lead), Kelydra Welcker, Xiaoning He

DISCLAIMER •  The information, data, or work presented herein was funded in

part by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the United States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof.

Appalachian sedimentary

basin

Play Fairway Analysis Area Potential users of low-temperature geothermal resource found widely

Focus area: New York Pennsylvania West Virginia

Why Seismic Hazards Here? •  For most geothermal exploration, finding faults is a good thing

•  Because it tends to locate either slip or dilation-induced permeability tendency, or to identify structural controls on flow systems

•  In our region, we felt a strong issue would be a potentially detrimental effect on geothermal projects’ “social license to operate”

•  Seismicity has adversely affected geothermal projects in at least: •  Basel, Switzerland •  The Geysers, California •  Landau, Germany •  And those are only off the top of my head from the last decade or so…

•  So we set out to identify potential locations at risk of induced seismicity to caution users of our “Geothermal Play Fairways”

Pre-existing “Fault” Maps Do not share the GPFA-AB boundaries or scale. Leads to problems of uneven coverage, varying interpretation of faults vs. lineaments, and different mapping scales.

Poisson Wavelet Multi-Scale Edge Analysis of Potential Fields (“Worms”) in One Slide

Above Ground

Underground

Gravity Worms for the Region

Magnetic Worms for the Region

Earthquakes

Events from the NEIC (green) and Earthscope’s TA (red) occurred between 1/1/1965 and 31/5/2015. An approximate time of day based decontamination procedure was used to remove mine/quarry blasts from the TA events which could have also removed a bit less than half of the real TA EQs.

Two Techniques for Identifying Risky Structures 1.  Worms near recorded events 2.  Worm orientation in regional stress field

All Worms Within 20 km of EQs DE-EE0006726

Cornell University

FY2015, Q4

Page 26 of 73

Figure 21: Map of risk of induced seismicity based on the proximity to a recorded seismic event and/or the co-occurrence of a “worm” and a nearby seismic event. Red indicates highest risk; green is lowest risk.

As the project progressed, after discussion with experts both within and outside of the team, we tried an additional approach. Plotting multiscale edge Poisson wavelet analysis lateral boundaries within World Stress Map ı1 directions (see Catalog of Supporting Files for more information in the Identifying Potentially Activatable Faults Memo) yields an alternative qualitative approach to identifying reactivatable structures. Here, we use “worms” to identify structures with strikes favorably oriented for failure by Byerlee's law (Figure 22, Stress Orientation Hazard). While this might be a necessary criterion for fault activation (under an assumption of the validity of Byerlee’s Law model) it is not a sufficient criterion - because we lack detailed information about stress magnitudes throughout the GPFA-AB region. This approach is termed a slip-tendency estimate.

Ultimately, we judged that the most useful representation of the total seismic risk resulted from the combined risk map formed by averaging the risk factor categories given by the proximity technique and the slip-tendency technique (Figure 23, Combined Seismic Risk).

World Stress Map

From Heidbach et al. 2010.

Orientation in Stress Field

Byerlee’s (1978) Law

Griffith-Coulomb Failure

Assume worms are representing vertically dipping structures and σ3 is horizontal. ⇒ Orientation in σ1 direction and

slip tendency is a function of azimuth only

Brief Aside: Byerlee’s Law

Byerlee (1978)

Worm Segments Oriented for Slip Relative to WSM

DE-EE0006726

Cornell University

FY2015, Q4

Page 27 of 73

Figure 22: Map of the risk of induced seismicity estimated based on slip-tendency, as calculated by the premises of the locations of planes of weakness relative to the regional stress field. For this analysis, the “worms” are treated as if they are all planes of weakness, and Byerlee’s Law criteria used.

Some Slip Orientation Classification Statistics

MISORIENTATION •  0-5° •  5-10° •  10-15° •  >15°

Averaging Proximity and Orientation Risks DE-EE0006726

Cornell University

FY2015, Q4

Page 28 of 73

Figure 23: Preferred map of the spatial distribution of risk of induced seismicity, created by averaging the proximity and slip-tendency techniques.

In addition to the summary of the seismic risk method in the Catalog of Supporting Files, the reader should refer to (Seismic Risk Map Creation Methods Memo and the Identifying Potentially Activatable Faults Memo) for a complete discussion of these matters, including an overview of the wavelet theory, the earthquake catalog cleaning, and the computational techniques developed to estimate risks and their uncertainties at all points along the worms.

Challenges The risk of induced seismic activity is now on many peoples’ mind who live within the shale plays across the county. Those living in New York, Pennsylvania and West Virginia are no different. There is concern that the reinjection of water produced during oil and gas extraction will induce seismicity, and it is easy to imagine that a similar concern will be raised about the recirculation of water in a geothermal energy extraction project.

EQ-Worm Proximity Histogram •  From a sister USGS funded project over a different

(Adirondack) footprint with Cindy Ebinger and Korin Carpenter (U. Rochester)

Worm Segments Oriented for Dilation Relative to WSM

These identify locations where one might find (mode I) fracture permeability for geothermal developments – a good thing!

Conclusions •  Some lateral discontinuities detected by worms are active faults

•  Even in intra-plate settings •  Worms nearby to EQs are candidate active faults •  Orientation-in-principal-stress-directions, while a necessary

condition for induced seismicity under a Byerlee’s Law model, doesn’t actually appear to work very well for natural seismicity •  Possibly due to a lack of stress magnitude information •  Possibly due to that being an incorrect model for active faulting/

induced seismicity •  We’ve found locations where direct use geothermal prospects

appear to be at higher risk of induced seismicity than for other locations

Thanks! • Questions?

•  frank.horowitz@cornell.edu