Modeling of Inhomogeneous Spatio-Temporal Signals by Least Squares Collocation

Abstract Through inverse modeling and adjustment techniques, the geodesists try to derive mathematical models from their measurements get a better understanding of various processes in system Earth. Sophisticated deterministic stochastic are developed achieve best possible reflection reality remaining uncertainty. The main focus this article is on further development model representations, with capability switch usual assumption homogeneous (time-stationary) inhomogeneous (time-variable) models. To accomplish we build up extend methodical framework connect filter covariance approach represented by time-variable autoregressive (AR) (inhomogeneous) for least squares collocation. We apply these describe temporal component spatio-temporal point stack surface displacements derived DInSAR-SBAS analysis ERS1 ERS2 missions Lower-Rhine Embayment North Rhine-Westphalia. construction allows use collocation predict vertical movements at any location time. Furthermore, report uncertainty prediction provided.