Michael, Andrew
Dear Colleagues,
We hope you will consider submitting an abstract to our session, "Beyond Poisson: Seismic Hazards and Risk Assessment for the Real Earth," at the SSA virtual meeting in 2021. Our goal is to examine the utility of fundamental changes in how we assess hazards for a wide range of societally important applications.
Beyond Poisson: Seismic Hazards and Risk Assessment for the Real Earth
Traditional probabilistic seismic hazard assessments (PSHA) assume that it is adequate to model earthquakes as temporally random, independent events modeled as a Poisson process. This approach removes the obvious clustering due to aftershocks and swarms; averages or carves out rate variations on short time scales due to natural processes such as volcanoes, as well as short-lived induced seismicity in order to focus on long-term so-called tectonic rates; and may not even use time-dependent mainshock probability variations due to elastic rebound. This may be adequate for long-term (e.g. 50-year) models aimed at low probabilities of exceedance for engineering purposes. The risk industry, typically managing risk transfer contracts from 1 to 5 years, is interested in assessing risk due to processes that act at or affect shorter time scales, from induced seismicity to aftershocks to fault interaction and elastic rebound. Over shorter timescales, when considering the impacts of multiple events and/or at higher probabilities of exceedance, non-Poissonian behavior becomes more important. Recently our ability to describe non-Poissonian behavior has been improved through short-term aftershock and swarm models, models of fluid injection and long-term physics-based simulators. As a result, some national hazard models have incorporated these processes. Many challenges remain. For instance, understanding the impact of incomplete data and non-stationarity on long-term empirical rate estimates is a particularly difficult issue in regions with low seismicity rates and is critical even for traditional PSHA. We encourage contributions that explore how we can better model the broad range of real Earth behavior in different time scales that goes beyond the Poisson process; how we can test those models and include them in hazard and risk assessments; and the societal utility of doing so for a range of users including engineers, the insurance industry, emergency planning and mitigation.
Conveners
Andrew J. Michael, U.S. Geological Survey (ajmichael<at>usgs.gov)
Edward H. Field, U.S. Geological Survey (field<at>usgs.gov)
Delphine D. Fitzenz, RMS (delphine.fitzenz<at>rms.com)
Matthew C. Gerstenberger, GNS Science (m.gerstenberger<at>gns.cri.nz)
Andrea L. Llenos, U.S. Geological Survey (allenos<at>usgs.gov)
Warner Marzocchi, University of Naples Federico (warner.marzocchi<at>unina.it)
Email ajmichael<at>usgs.gov<ajmichael<at>usgs.gov> (michael<at>usgs.gov<michael<at>usgs.gov>)
Due to the pandemic, please contact me via email and ask before mailing or shipping me anything.
Phone: 650-439-2777
Mailing address: USGS P.O. Box 158, Moffett Field, CA 94035
Boxes, Fedex, UPS...: USGS MS9777, 345 Middlefield Road, Menlo Park, CA 94025
We hope you will consider submitting an abstract to our session, "Beyond Poisson: Seismic Hazards and Risk Assessment for the Real Earth," at the SSA virtual meeting in 2021. Our goal is to examine the utility of fundamental changes in how we assess hazards for a wide range of societally important applications.
Beyond Poisson: Seismic Hazards and Risk Assessment for the Real Earth
Traditional probabilistic seismic hazard assessments (PSHA) assume that it is adequate to model earthquakes as temporally random, independent events modeled as a Poisson process. This approach removes the obvious clustering due to aftershocks and swarms; averages or carves out rate variations on short time scales due to natural processes such as volcanoes, as well as short-lived induced seismicity in order to focus on long-term so-called tectonic rates; and may not even use time-dependent mainshock probability variations due to elastic rebound. This may be adequate for long-term (e.g. 50-year) models aimed at low probabilities of exceedance for engineering purposes. The risk industry, typically managing risk transfer contracts from 1 to 5 years, is interested in assessing risk due to processes that act at or affect shorter time scales, from induced seismicity to aftershocks to fault interaction and elastic rebound. Over shorter timescales, when considering the impacts of multiple events and/or at higher probabilities of exceedance, non-Poissonian behavior becomes more important. Recently our ability to describe non-Poissonian behavior has been improved through short-term aftershock and swarm models, models of fluid injection and long-term physics-based simulators. As a result, some national hazard models have incorporated these processes. Many challenges remain. For instance, understanding the impact of incomplete data and non-stationarity on long-term empirical rate estimates is a particularly difficult issue in regions with low seismicity rates and is critical even for traditional PSHA. We encourage contributions that explore how we can better model the broad range of real Earth behavior in different time scales that goes beyond the Poisson process; how we can test those models and include them in hazard and risk assessments; and the societal utility of doing so for a range of users including engineers, the insurance industry, emergency planning and mitigation.
Conveners
Andrew J. Michael, U.S. Geological Survey (ajmichael<at>usgs.gov)
Edward H. Field, U.S. Geological Survey (field<at>usgs.gov)
Delphine D. Fitzenz, RMS (delphine.fitzenz<at>rms.com)
Matthew C. Gerstenberger, GNS Science (m.gerstenberger<at>gns.cri.nz)
Andrea L. Llenos, U.S. Geological Survey (allenos<at>usgs.gov)
Warner Marzocchi, University of Naples Federico (warner.marzocchi<at>unina.it)
Email ajmichael<at>usgs.gov<ajmichael<at>usgs.gov> (michael<at>usgs.gov<michael<at>usgs.gov>)
Due to the pandemic, please contact me via email and ask before mailing or shipping me anything.
Phone: 650-439-2777
Mailing address: USGS P.O. Box 158, Moffett Field, CA 94035
Boxes, Fedex, UPS...: USGS MS9777, 345 Middlefield Road, Menlo Park, CA 94025
