Thread: SSA Session on CSEP and Increasing Testability

Started: 2018-01-12 19:04:31
Last activity: 2018-01-12 19:04:31
Topics: SSA Meetings
Andrew Michael
2018-01-12 19:04:31
Hundreds of forecast models are currently being tested by the Collaboratory
for the Study of Earthquake Predictability (CSEP). We will have a special
session at the 2018 LACSC/SSA joint meeting in May to consider future
improvements in the CSEP tests and ways to expand from earthquake forecasts
to forecasts of hazard and risk. We invite abstracts on these ideas,
regardless of past involvement with CSEP itself. The full session
description is below.

Abstract Deadline is January 24, via the meeting website,
and the meeting itself is May 14-17, 2018 in Miami.

Andy Michael, Danijel Schorlemmer, Maximilian J. Werner, and Warner

Increasing Testability – Expanding Possibilities and Future Developments of
the Collaboratory for the Study of Earthquake Predictability

The Collaboratory for the Study of Earthquake Predictability (CSEP) has
expanded over the years to many different testing areas hosted at multiple
testing centers. Hundreds of earthquake forecast models have been submitted
to CSEP and are being tested. New testing metrics were developed and
implemented and a lot of progress was made to establish CSEP as an
institution that cannot be ignored when issuing earthquake forecasts. Its
rigor and independence became the standard in evaluating earthquake
forecasts and in reporting on the results.

Although the tests CSEP has conducted have been successful and
well-received, they have also shown the limitations of the CSEP approach.
What is a sufficient testing period for models? Are time-invarying models
really describing the long-term seismic activity? Are long-term models
testable at all? Do short-term models provide significant information for
the forecasting problem or do they only model aftershock sequences? What
other signals should be included in forecasting models to improve them? Do
improvements in forecasting models translate into improvements of hazard
models? How can the tests be improved? Many aspects of seismic hazard or
earthquake forecasting remain inherently untestable if only the model
forecasts are tested and not the model ingredients. We propose to create
new areas of activity for CSEP, namely targeted experiments that cannot be
conducted with the current CSEP software system.

We solicit contributions addressing forecasting models, forecast testing
problems, new ideas for CSEP experiments, possibilities of further CSEP
developments, ways of expanding CSEP into the hazard and risk domain and
more general views on the forecasting problem.

Session Conveners
Andrew J. Michael, U.S. Geological Survey, <michael<at>>
Danijel Schorlemmer, GFZ Potsdam, <ds<at>>
Maximilian J. Werner, University of Bristol, <max.werner<at>>
Warner Marzocchi, Istituto Nazionale Geofisica e Vulcanologia, <

07:55:27 v.ad6b513c