Tropospheric delay corrections to differential InSAR results from GPS observations

Differential Interferometric Synthetic Aperture Radar (DInSAR) techniques have been recognised as well-suited for ground deformation monitoring applications. However, the spatially and temporally variable delay of the radar signal propagating through the atmosphere represents a major limitation to accuracy. The dominant factor to be considered is the tropospheric heterogeneity, which can lead to misinterpretation of DInSAR results. In this paper a between-site and between-epoch double-differencing algorithm for the generation of tropospheric corrections to DInSAR results based on GPS observations is proposed. In order to correct the radar results on a pixel-by-pixel basis, the GPS-derived corrections have to be interpolated. Using GPS data from the Southern California Integrated GPS Network (SCIGN) it has been found that the inverse distance weighted and Kriging interpolation methods are more suitable than the spline method. Differential corrections as much as several centimetres may have to be applied in order to ensure sub-centimetre accuracy for the DInSAR result and it seems optimal to estimate the tropospheric delay from GPS data at 5-minute intervals. The algorithm and procedures described in this paper could easily be implemented in a continuous GPS data centre. The interpolated image of between-site, single-differenced tropospheric delay can be provided as a routine product to assist radar interferometry.