Interaction of Atmosphere Modeling and GPS Analysis Strategy

GPS estimates of the vertical component of position are sensitive to the minimum elevation of observations included in the analysis. Both the mean value and the scatter are affected by the choice, which is usually a global parameter that applies to all antennas. The uncertainties in the estimates, as well as errors in the estimates, have a spectrum that varies from sub-daily to decadal. Both will vary with geographic location (primarily latitude) and will differ for individual sites. Three effects are strongly elevation dependent and compete in determining what minimum elevation should be adopted: formal uncertainty of the height estimate, uncertainties from the atmosphere delay model, and antenna effects. Improvements are being made in each of these areas. Thus it is desirable to reassess the choices made for analysis strategies, including minimum elevation. Comparisons will be presented of the magnitudes and characteristics of these three effects, including recent advances in atmosphere mapping functions, that might help to address the following questions: 1. How should the minimum elevation be chosen? a) for best repeatability? what time scale? or b) for best accuracy (to the extent it can be determined)? 2. Which type of mapping function should be used? a) independent of external information (similar to NMF)? or b) makes use of time-dependent external information (such as Numerical Weather Model)? 3. Should the minimum elevation vary among sites? a) by latitude? or b) by site? 4. Should the minimum elevation differ for different types of solutions, analogous to the meteorologists' differing analyses for climate modeling and for forecasting, for example: a) global solutions, which might desire better accuracy when averaged over long term (such as for TRF), or b) regional solutions, which might seek better short-term repeatability (such as for detecting precursory motion).