SNARF Working Group - Report of the Second SNARF Workshop

Eos Trans. AGU, 85(17), Jt. Assem. Suppl., Abstract G21D-04 INVITED, 2004

Realization of a Stable North America Reference Frame

Herring, T A
Email: tahmit.edu
Address: Thomas A Herring, Room 54-618, MIT 77 Massachusetts Avenue,
Cambridge, MA 02139 United States

In this paper we address the realization of a stable North American frame (SNARF). The long term realization can be acheived through the positions and velocities of a set of sites on a stable part of the Northern American plate that will have zero average velocity in the reference frame combined with sites which have well defined velocities relative to this stable region. These other sites may be on different tectonic plates with a rigid body rotation rate between the plates. The estimated velocities of realizations of this type using GPS data collected over the last decade differ from zero or the predicted motion with root-mean-square (RMS) scatter of ~0.6 mm/yr for the horizontal components of motion. Vertical motion RMS scatters are aronnd the 1 mm/yr level. The greater challenge for SNARF is the day-to-day realization of the frame. For this application, deviations of sites from zero or secular motion will introduce daily rotations and translations of the frame. These deviations arise from atmospheric pressure loading which should be well modeled with high temporal resolution and water loading variations which are not well modeled currently. Other complications arise from instrument failures and equipment changes. The net effect of these deviations is that frame realizations over small regions (such as Southern California) are more precise (daily position RMS of 0.6 mm) while continental scale realizations have RMS position deviations of 1-2 mm. We discuss methods that could be used to achieve day-to-day continental frame accuracy similar to regional realization precision.

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