SNARF Working Group - Report of the Second SNARF Workshop

Eos Trans. AGU, 85(17), Jt. Assem. Suppl., Abstract G21D-02 INVITED, 2004

The Hunting of the SNARF

Sella, G
Email: sellaearth.nwu.edu
Address: Northwestern Univ., 1850 Campus Dr.,
Evanston, IL 60208 United States

Stein, S
Email: sethearth.nwu.edu
Address: Northwestern Univ., 1850 Campus Dr.,
Evanston, IL 60208 United States

Wdowinski, S
Email: shimonwrsmas.miami.edu
Address: Univ. of Miami, 4600 Rickenbacker Csway,
Miami, FL 78890 United States

Dixon, T
Email: tdixonrsmas.miami.edu
Address: Univ. of Miami, 4600 Rickenbacker Csway,
Miami, FL 78890 United States

Establishing a Stable North American Reference Frame (SNARF) is important to understand the processes that occur in plate boundary zones and that affect the stable plate itself. Stable North American defined as the area east of the Rocky Mountains has a number of seismically active areas including New Madrid Seismic Zone and the Charelvoix region of Quebec whose causal mechanisms remain poorly understood. For the last 10 years we have been characterizing the kinematics of present day deformation of the stable plate interior using both Global Positioning System (GPS) data and seismic data. As the number of continuous GPS (CGPS) sites in Eastern North America and the lengths of data have increased we have seen a steady improvement in the root mean square misfit of site velocities to those expected for a perfectly rigid plate to less than 1mm/yr. However further reductions in the misfit are challenging since simply increasing the number of sites results in little to no improvement probably reflecting systematic errors. These errors include in order of importance: characterizing propagation delays caused by water vapour in the atmosphere, glacial isostatic adjustment (GIA) due to glacial unloading, regional site motions due to hydrologic effects, local site motions, reference frame, and satellite orbit errors. Obtaining site, temporally and azimuthally specific estimates of the water vapour content would result in a significant improvement to current mapping functions. Although we believe that we can identify sites that are most affected by GIA and exclude these sites from our stable plate definition, we are still unable to remove the GIA signal from the remaining sites. Global annual and semiannual hydrospheric signals can be successfully removed, but regional to local ones are still poorly understood. Part of the understanding of this signal will emerge through high-density networks being built with 50-100km spacing. Local site motions and equipment biases are very difficult to identify unless they are very large, althougwwwh integrating spatially continuous data such as InSAR may help identify these effects. Reference frame, ITRF 2000, and satellite orbital errors are probably much smaller than any of the other error sources.

Last modified: 2019-12-24  02:12:53  America/Denver