Assessing the Updip Spatial Offset of Tremor and Slip during ETS Events in Cascadia

Abstract:

We investigate the updip spatial overlap of tremor and slip during recent episodic tremor and slip (ETS) events in Cascadia using a combination of forward and inverse models constrained by GPS, strainmeter, and tremor observations. Results from major ETS events in northern Cascadia suggest that, although there is significant spatial overlap, slow slip tends to extend further updip than tremor. ETS activity is thought to be dependent on a range of parameters, such as variable fluid pressures, temperature dependent physical properties, and facies changes. A spatial offset would indicant that tremor and slip are reflective of different physical conditions. While a clear offset of tremor and slip has been observed in multiple other subduction zones, a similar offset in Cascadia has remained difficult to constrain. Here we seek to establish whether the updip spatial offset is real in Cascadia and to quantify its extent.

To complement GPS observations in Cascadia, we incorporate high fidelity strainmeter observations into inversions and sensitivity tests of iterative forward models. Tremor distributions are used as a proxy for slip and incorporated into slip models where parameters affecting the distribution and magnitude of slip are allowed to vary. These slip models are used to forward predict surface displacements and strains, which are then compared to the geodetic observations and inferred slip based on geodetic inversions. Results indicate that, while the tremor-derived slip distributions do a good job predicting the broad-scale surface deformation, the best-fit models have slip updip of the peak tremor activity. The fine-scale relationship of tremor and slip appears to vary on an event-by-event basis, where areas of high tremor density do not always correlate with increased surface displacements and vice-versa.