Statistical geodesy—An engineering perspective

Motivated largely by the needs of the navigation community, engineers have become increasingly involved in understanding and modeling the fine structure of the earth's gravity field. In this endeavor, the engineer has augmented the applied mathematical tools of the "geodesist-scientist" with those of modern estimation and control theory, state-space mathematics, and random process theory. One of the outputs of this involvement has been the development of "statistical geodesy." In this paper, the mathematical structure and applications of statistical geodesy are reviewed, with an emphasis on the engineer's contribution. Geodetic terminology, geopotential theory, and estimation theory are briefly reviewed, and models with a random-process-theory structure are presented for uncertainties in the earth's gravity field. These models are then utilized in a variety of applications: estimation of gravimetric uncertainties, error analysis of inertial navigation systems, gravity gradiometry, satellite altimetry, etc. Finally, a new algorithm is presented-frequency-domain collocation-suitable for the efficient processing of large amounts of gravimetric data.