Abstract:

With the increased use of GPS-based positioning, the demand for directly converting ellipsoidal heights to a height referred to a regional vertical datum with sufficient accuracy is also increasing. In Canada, the use of a gravimetric geoid as the national height reference surface has been under study. The recent revolutionary development of space gravimetry (i.e. CHAMP, GRACE and the upcoming GOCE mission) offers the opportunity to pursue such a dramatic change in the local vertical datum. Among others, a compulsory requirement of adopting a vertical datum based on a gravimetric geoid, rather than conventional levelling networks, is a reliable assessment of the systematic and stochastic errors. This is a challenging task as there is limited or poor information regarding the quality of the terrestrial gravity data, difficulty in quantifying the error for the gravity reduction and interpolation, as well as the reality of the estimated uncertainties of the satellite data derived gravity model. The purpose of this paper is twofold, namely (i) to provide a detailed examination of all sources of the gravimetric geoid error in Canada (e.g., GRACE gravity model: GGM02C, terrestrial gravity data, gravity reduction methods, Stokes integration error and the far-zone contribution error), and (ii) to assess an adequate method for the estimation of the gravimetric geoid error from these error sources. In particular the variance component estimation schemes MINQUE and AUE are implemented to validate/calibrate the geoid error using existing GPS-levelling data.