USGS visual identity mark and link to main Web site

USGS Mendenhall Postdoctoral  Research Fellowship Program

14-30.  Operational Earthquake Forecasting.

We seek a postdoctoral fellow to contribute to the multidisciplinary goal of Operational Earthquake Forecasting (OEF), which seeks to provide authoritative information on the time dependence of seismic hazards, including the effects of spatiotemporal earthquake clustering (including aftershocks).  The 2010 Canterbury, New Zealand earthquake sequence was one dramatic reminder that aftershocks can be more damaging than main shocks, and the need for proper information dissemination was exemplified by the indictment of six Italian scientists following the 2009 L’Aquila, Italy earthquake.  It is for these reasons that OEF is listed as a strategic-action priority by the USGS.  Specifically, the goal is to provide “effective situational awareness” support for effective decision making during hazardous events, with core responsibilities including issuing warnings and providing timely information to emergency managers, the media, and the public (; page 31).

As discussed in an article by Jordan and Jones (2010, Seismological Research letters, volume 81, number 4, pages 571-574), OEF has effectively been practiced in California for years via the California Earthquake Prediction Evaluation Council (CEPEC).  However, what has been lacking is a well tested, automated, and real-time OEF system, as well as a model that gives reliable estimates of forecast uncertainties. Also lacking are formal and official protocols for disseminating results to the public, especially in terms of risk metrics, such as estimated damage and casualties, rather than just earthquake probabilities.

The USGS, Southern California Earthquake Center (SCEC), and California Geological Survey (CGS) have set up a Working Group on California Earthquake Probabilities (  to incorporate short-term forecasting into the next version of the fault-based Uniform California Earthquake Rupture Forecast (UCERF3).  This infrastructure will provide a comprehensive framework for implementing a variety of spatiotemporal clustering models, which can then be formally tested in the Collaboratory for the Study of Earthquake Predictability (CSEP,

OEF also forms a bridge between long-term seismic hazard analysis and another operational system under development in California: earthquake early warning.  Real-time forecasts can serve as a priori information for an earthquake early warning system, because a signal is less likely to be a false alarm in time periods of higher hazard.  Particularly in the wake of a large earthquake, the public expects up-to-date hazard information from the USGS, much as they currently receive real-time earthquake information.  Operational forecasting serves the USGS goals of expanding real-time systems and providing improved information to reduce earthquake losses.

We invite research proposals that focus on one or more of the many outstanding issues associated with OEF, all of which will require strong scientific originality.  A partial list of possible topics includes:

  1. The compatibility of elastic rebound and aftershocks statistics. Traditional models used in short-term earthquake forecasting, such as Epidemic-Type Aftershock Sequences (ETAS) and the Short-Term Earthquake Probabilities (STEP). model, have ignored the effects of elastic rebound.  However, preliminary UCERF3 modeling indicates that leaving out elastic rebound causes large earthquakes to re-rupture the same fault surface much more frequently than is seen in nature, implying that some relaxation mechanism must be operating.  We do not currently understand how to best model these effects together.
  2. The compatibility of characteristic magnitude frequency distributions on faults with aftershocks statistics.ETAS and STEP models also assume a Gutenberg-Richter distribution of earthquakes, which is incompatible with fault-based models such as UCERF3, where faults can be strongly characteristic (i.e., incompatible with Gutenberg-Richter).  Therefore, an outstanding issue is whether faults should be more Gutenberg-Richter or whether our models of aftershock statistics need to be revised.
  3. Assessment of the variability of aftershock sequences.  Aftershock productivity from one sequence to another apparently varies by over an order of magnitude, which has dramatic implications for aftershock probabilities.  Early in a sequence there is little data with which to compute sequence-specific aftershock statistics, necessitating the use of generic parameters that may not be a good fit to a particular region.   Research into the variability of sequence-specific parameters and the factors such as tectonic regime that affect the variability, especially considering the fact that many aftershocks go undetected, could greatly improve OEF.
  4. How to compute loss estimates from a real-time OEF? The USGS PAGER system ( gives real-time information on expected losses from an earthquake that has just occurred.  An obvious extension would be to also forecast possible future losses from any triggered events.
  5. How to test OEF forecasts which are made with preliminary data that is changing rapidly?  Should the forecasts be tested against the preliminary data or the final results?  How does the preliminary nature of the data affect the forecasts through increased uncertainty? How can these forecasts be effectively tested in CSEP?
  6. How to best communicate results to the public? This is especially challenging when the probability increase may be quite high (compared to long-term probabilities), but the overall probability of a damaging event remains low.

We are also open to research proposals that address other challenges associated with OEF.  Although the scientific challenges are significant, the Mendenhall Fellow will be joining a multidisciplinary, multi-institutional project (UCERF3; that has been making progress toward OEF for over three years.  The Fellow will also be able to utilize and contribute to a modern computational infrastructure that includes object-oriented programming and high-performance computing.  In short, the postdoctoral researcher will be well placed to not only pick some of the ripest fruit scientifically, but to also contribute to a very societally relevant forecast system.

Proposed Duty Station: Golden, CO; Menlo Park, CA; Pasadena CA

Areas of Ph.D.: Geophysics, seismology, geology, geodesy, engineering, mathematics, computer science, or other fields related to the deployment of OEF such as sociology related to natural disasters (candidates holding a Ph.D. in other disciplines, but with extensive knowledge and skills relevant to the Research Opportunity may be considered).

Qualifications: Applicants must meet one of the following qualifications - Research Geophysicist, Research Geologist, Research Geodesist, Computer Scientist, Research Physicist, Research Mathematician, Research Engineer, Research Economist, Research Sociologist.

(This type of research is performed by those who have backgrounds for the occupations stated above. However, other titles may be applicable depending on the applicant's background, education, and research proposal. The final classification of theposition will be made by the Human Resources specialist).

Research Advisor(s): Edward (Ned) Field, (303) 273-8469,;  Morgan Page, (626) 583-6804,; Karen Felzer, (626) 583-7822,; Andrew Michael, (650) 329-4777,; Lucile Jones, (626) 583-7817,

Human Resources Office Contact: Robert Hosinski, (916) 278-9397,

Go back to Summary of Opportunities

U.S. Department of the Interior, U.S. Geological Survey
URL: Field.htm
Direct inquiries to Rama K. Kotra at
Maintained by: Mendenhall Research Fellowship Program Web Team
Last modified: 17:58:05 Tue 23 Jul 2013
Privacy statement | General disclaimer | Accessibility