Deformation Models

Ideally, a deformation model would give us crustal stressing and/or straining rates throughout the region. More specifically, we want:

• slip-rate estimates for known major faults
• moment-accumulation rates elsewhere

The first two deformation models below follow the tradition of previous WGCEPs and the USGS National Seismic Hazard Maps. Subsequent versions will be more ambitious as outlined below.

Deformation Model 1

Status: Available

This model is simply Fault Section Database 1 (that is, Fault Model 1 geometries with associated "expert judgement" estimates of average slip-rate and aseismic-slip-factors).

Deformation Model 2

Status: Available

UCERF2 incorporated six different deformation models as described here.

Deformation Models 3.0 (or Greater)

Status: in development

To download current development deformation model files, click here

The slip-rate values assigned in previous models have been based on expert-opinion evaluation of available data (mostly geologic and geodetic), together with summations across various transects to make sure the total plate tectonic rate is matched. One big question is whether more quantitative models, such as Peter Bird’s NeoKinema (e.g., Bird, 2009), can be used in place of expert opinion.  The other big questions here include:

• Can the more sophisticated deformation models give us slip rate estimates for the faults where we currently lack geologic constraints.
• Can these deformation models give us a more refined estimate of the spatial distribution of deformation occurring off the explicitly modeled faults (something better than the polygons defined for the last model)?  This would provide a good alternative to smoothed instrumental seismicity for constraining the rate or maximum magnitude of “background” seismicity (earthquakes off the explicitly modeled faults).
• Related to the above is the overall amount of seismic deformation occurring off the explicitly modeled faults.  Deformation Models 2.x were constructed to explicitly match the total plate rates (within uncertainties), which basically assumed that all seismic deformation occurs on the modeled faults (i.e., assuming rigid blocks in between).  One question is whether we should reduce such fault slip rates by some fraction in order to avoid double counting with respect to off-fault seismic deformation.  In fact, NeoKinema suggests that the off-fault deformation is 30% (which implies we may have over-estimated some fault slip rates).
• Can we get a more refined estimate of slip-rate changes along strike (i.e., to avoid un-physical abrupt changes at the current fault-section boundaries.  An alternative here would be to assign existing slip rate constraints only at section midpoints (unless an abrupt change is expected due to fault branching).  However, exactly how and if slip tapers off between neighboring faults will likely be very influential in solving for the rate of multi-fault ruptures, so we probably need to address such transitions in as much detail as possible.
• Is there a systematic bias in the assignment of slip-rates to some of the slower moving faults (e.g., by implicitly defining a water level)?

Finally, it will be important to have a range of deformation models in order to represent uncertainties with respect to the above issues (versions 3.1, 3.2, …).

Please see the UCERF 3 Project Plan for more information.