GPS Analysis of Olympia quake at RPI

Refresh (re-load) often as the contents will change.

We processed several days of continuous GPS data leading up to, and a few days after, the Olympia, Washington, M 6.8 earthquake of Feb. 28, 2001. Data are taken from sites operated by the:
Geological Survey of Canada (Western Canada Deformation Array) (NANO, CHWK, DRAO, ALBH)
University of Washington (NEAH, SEAT, SEDR, KELS)
Central Washington University (REDM, GOBS, PABH, LIND, CHZZ)
US National Geodetic Survey CORS (FTS1, WHD1, RPT1, GWEN)
Cascade Volcano Observatory (JR01, SATS)
RPI and OSU (CORV, NEWP)
Yakima city (COY1)
and one private citizen (DWH1).
We use gamit/globk software to estimate the daily positions in ITRF96. Linear trends were estimated from the data for 22 days prior to the quake and this trend was removed from all days. The average site position after the quake relative to this linear trend was taken as the uncorrected co-seismic offset. Once the North and East offsets were estimated for all sites, the overall net rotation of the displacement field was removed to minimize any common network motions.
This analysis revealed co-seismic displacements of up to 8 mm as shown in the figures below. The displacements can be matched fairly well with simple dislocation models of the quake based on parameters taken from seismological solutions. We will perform inversions of the geodetic displacements later.
Updates:
3/8/01 - Results updated with data through 3/5/01
3/12/01 - Updated through 3/9/01 (Julian Days 037 - 068), site DWH1 added, inverted for slip on fault
3/19/01 - Added Yakima site, fixed vertical on time series, updated through day 074, models revised using OSU seismological solutions from John Nabelek.
Displacement vectors available upon request.

Map of continuous GPS sites (star shows epicenter of quake).

Focal mechanism from ERI (star shows epicenter of quake).

GPS time series (click here for larger image) from continuous sites in the Pacific Northwest. Each dot represents one day and the vertical line a little more than half-way along the time axis is the time of the quake. Vertical bars are 5 mm high. Sites are arranged in order of increasing distance from the quake (Del=distance in km, Az=azimuth in degrees). These data have not been corrected for common network motions.

Observed GPS (red) and theoretical (blue) displacements for steep, east-dipping fault based on OSU's long-period solution. Observed displacements have 2.5-sigma ellipses. Blue vectors are predictions of a half-space dislocation model of the quake. Most parameters were taken from the seismological solutiuon but we inverted for the amount of slip on the fault (with other parameters fixed), giving us a geodetic moment estimate. (Slip amount trades off directly with assumed fault area, so moment might be the only resolved parameter at this point.)
Geodetic moment = 1.09 +/- 0.12 x 10^19 N-m (adjusted to fit GPS ampltudes)
Seismological moment = 1.8 x 10^19 N-m
Centroid depth = 52.5 km
Centroid location: 47.1525N, 237.203E
Strike: -1 degrees
Dip: 69 degrees easterly
Along strike length = 30 km
Slip on fault: 0.906 +/- 0.103 meters
Rake: 270 degrees
Downdip width: 10 km
Rigidity: 40 GPa
Data: 24 3-component GPS derived displacements, normalized chi2 = 0.76

Observed GPS (red) and theoretical (blue) displacements for steep, east-dipping fault based on OSU's body-wave solution. Observed displacements have 2.5-sigma ellipses. Blue vectors are predictions of a half-space dislocation model of the quake. As above, we solved for the slip amount.
Geodetic moment = 1.18 +/- 0.14 x 10^19 N-m (adjusted to fit GPS ampltudes)
Seismological moment = 1.4 x 10^19 N-m
Centroid depth = 61 km
Centroid location: 47.1525N, 237.203E
Strike: -11 degrees
Dip: 71 degrees easterly
Along strike length = 30 km
Slip on fault: 0.980 +/- 0.116 meters
Rake: 270 degrees
Downdip width: 10 km
Rigidity: 40 GPa
Data: 24 3-component GPS derived displacements, normalized chi2 = 0.88

Contact: Rob McCaffrey, e-mail, my web page
Rensselaer Polytechnic Institute, Troy NY
This page powered by

Last updated March 12, 2001