Combination of Heights

Abstract

This chapter provides a practical discussion on the combination of heterogeneous height data. Since most of the theory is discussed in detail in the previous chapters, only a brief introduction to the theoretical issues is included along with some insight into why combining geoid, orthometric and ellipsoidal height data is relevant and important on both regional and global scales. The next section is devoted to a detailed outline of a computational methodology that can be used for the optimal combination of heterogeneous height data (via least-squares adjustment). From this approach it is evident that two key elements deserve more attention, in particular the individual accuracy contributions of each of the height types (via variance component estimation) and the role of the parametric model which appears in the general linear functional model used in the adjustment. Modelling options are provided and more importantly an approach for the assessment of selected models is described in detail. These techniques are supported by numerical examples with real data sets (in Canada and Switzerland). Finally, it should be noted that this chapter is an abridged version of Fotopoulos (2003) and readers who are interested in implementing the methodologies shown herein are encouraged to view the aforementioned manuscript for further details.