Gravity anomaly recovery from satellite altimetry data using least squares collocation techniques
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The prediction of mean gravity anomalies from other anomalies and satellite altimeter data is described using the technique of least squares collocation. The equations derived are then used in simulation studies, using altimeter data alone, to estimate the accuracy to which we can determine mean gravity anomalies with respect to an ellipsoidal reference field and a degree 12 reference field. In these studies we found that for the ellipsoidal reference model it was necessary to include altimeter data out to 8° from the center of the anomaly block while such data was needed only out to 4° when using the N = 12 reference field. We also saw that along track and cross track spacings of 1° were adequate for anomaly recovery purposes. In fact the increase of the cross track spacing led to only small accuracy degradation. In addition we found that as the altimeter accuracy varied from ±0.5 m to 2 m only a small change in the accuracy of the predicted anomaly was found. Assuming a uniform 1° x 1° grid of altimeter data, of 1 m accuracy, within 8° of the block center we found the following prediction accuracies with respect to the ellipsoidal reference field: 5°, ±4.2 mgals; 2°.5, ±4.9 mgals; 1°, ±10.0 mgals.