Interpolation of GPS results incorporating geophysical and InSAR information

Continuous GPS networks, typically with a station spacing of about 30km, are still not dense enough to accurately characteristise the dynamics of active faults. Interpolation of these GPS results can improve our understanding of active faults and hence promote related studies. Moreover, even when the networks are densified in order to recover the signature of active faults, the station configuration design may not be ideal. Interpolation at these points, based on the GPS results from a well-designed station network, can provide a good quality control measure. As a first step in the interpolation process an irregular grid pattern is formed, based on the locations of the GPS stations, by using the indexed sorting algorithm. In order to interpolate objectively, the GPS stations and the intended interpolating points are classified into different sub-regions according to their positions in relation to the faults, which are expressed by open- and closed-curve models. GPS results from stations in the same sub-region are used to derive a dynamic model for interpolation at grid points in the same sub-region. A deformation distribution model based on GPS and differential Synthetic Aperture Radar Interferometry (InSAR) results is used as constraints to scale the time series generated using the dynamic model. 1