The least squares correlation method for gravity data separation

摘要

The least squares correlation method separates the observed gravity data into two parts,the regional anomaly and the residual anomaly,based on the goal of maximizing the correlation between the residual anomaly and the depth to the targeted density interface.The goal can be realized by setting up a linear least squares equation with respect to the residual anomaly and the depth to the interface and solving it iteratively using the conjugate gradient algorithm.The test on the synthetic data shows that the method is a relatively simple but effective tool for gravity data separation.