An Omnidirectional Filtering Method for Destriping Lunar Satellite Gravity Anomalies Data

Abstract

Abstract High‐resolution gravity field models of the Moon established from Gravity Recovery and Interior Laboratory satellite gravity data have been playing an important role in understanding the interior structure and tectonic evolution of the Moon. Due to the correlations in the high degree and order term coefficients of the gravity field models, the difference between satellite flight orbits, and the random noise of instruments, the high‐resolution gravity anomalies data derived from the models of high degree and order usually present serious multi‐directional striping and random noise, which clearly affect the subsequent interpretation of the data. We provide an omnidirectional filtering method based on the polynomial‐fitting principle to remove multi‐directional striping noise in the lunar satellite gravity anomalies data. A set of azimuth parameters are chosen to suppress all the directions of striping noise. The algorithms of data partitioning and iterative optimization are utilized to make our method suitable and stable for large‐scale data sets. Tests on the synthetic data and the real data from the Moon's Rümker region and globally verified the feasibility of our method with a better destriping effect than the traditional Gaussian filtering or degree‐order‐truncation methods.