Cascadia Rising 2016

The Cascadia Rising excercise is a complex disaster scenario based on a 9.0 magnitude earthquake along the Cascadia Subduction Zone (CSZ) and the resulting tsunami. From June 7-10, 2016 Emergency Operations and Coordination Centers (EOC/ECCs) at all levels of government and the private sector will activate to conduct a simulated field response operation within their jurisdictions and with neighboring communities, state EOCs, FEMA, and major military commands.
E-DECIDER will be partnering with the California Earthquake Clearinghouse and participating in their planned Cascadia Rising exercise activities.
See exercise partners for Cascadia Rising.

Products will be posted as part of the exercise in June!

See our products from the 2014 Cascadia Exercise for example products.

About the ETAS Aftershock Liklihood Forecast

The overlays show short term expected rates of seismicity based on aftershock statistics and scaling behaviors. "Aftershocks" are typically described as sequences of smaller earthquakes that follow large mainshock events.
While this description is reasonably accurate, it is important to understand that 1) all earthquakes have aftershocks -- including aftershocks, 2) aftershocks can have large magnitudes, and 3) like any other earthquake, aftershocks can trigger additional hazards such as building collapse, tsunami, landslides, and more aftershocks.
Aftershock sequences exhibit four important behaviors:
1) The rate of aftershocks is typically highest immediately after the mainshock and then decreases with time.
2) The spatial density of aftershocks decreases with increasing distance from the rupture area.
3) The largest aftershock, in a sequence, is typically approximately one magnitude smaller than the mainshock.
4) Number (3) is not always true. When an aftershock has larger magnitude than its parent event, we reclassify it as the "mainshock" of the sequence and call the original mainshock a "foreshock."
The aftershock forecasts provided for this exercise (Cascadia Rising 2016) are based on an ETAS algorithm published by Yoder et al. (2015) which capitalizes on these relatively well understood behaviors to calculate the initial rate and spatial density of an aftershock sequence based on its parent earthquake's magnitude. A rate map is then compiled by dividing a study region into a grid (typically 0.1 x 0.1 degrees lat/lon) and aggregating the aftershock activity from all earthquakes in the region, over some extended time window -- typically 5 to 10 years for detailed forecasts, at each location (lattice site).
This simulated forecast is computed by relocating the 2011 m=9 Tohoku-oki sequence from off the east coast of Japan to off the west coast of the United States. Specifically, all events in the Tohoku region for some 10 years before the mainshock through the duration of the exercise (between approximately March 2001 and April 2011) are relocated to the Cascadia region, rotated to align with the expected (simulated) rupture area, and the times of the events in the catalog are updated to correspond with the dates of the exercise. The Yoder et al. (2015) ETAS algorithm is then applied to this catalog for a set of dates before and after the simulated mainshock, with the intention of simulating "routine" forecasts, for example as issued on a 24 hour basis) before the mainshock, a forecast issued in response an m=7.6 event -- which will later be classified as a foreshock, and then at regular intervals (in this case, approximately 8 hours) following the mainshock.

Example Data Availability Timeline

Earthquake +0-1 hours

Modeled results based on location, magnitude, and likely faults

  • Modeled deformation maps (tilt maps, displacement maps, synthetic interferograms)
  • Critical infrastructure located near epicenter through KML on demand generator
  • PESH inputs for HAZUS-MH model (HAZUS gadget)
  • HAZUS total loss maps (user generated HAZUS models)
  • Aftershock forecasts


Earthquake +3 hours - 1 day

Optical data, more accurate GPS data and better estimates of fault rupture become available

  • Surface displacement maps with slope change based on improved models from data
  • Improved PESH inputs using new fault rupture estimates
  • Iterative forecast models based on aftershock occurrence


Earthquake +1 - 7 days

InSAR and additional remote sensing data become available to compute further improved fault models and surface displacement, as well as change detection

  • InSAR based damage assessment
  • Remote sensing based change detection