Reduction to the pole of large-scale magnetic anomalies via sliding window accounting for variable field direction and projection distortion

Reduction to the pole (RTP) is a common method in the processing and interpretation of lithospheric magnetic anomaly data. RTP requires knowledge of both the main geomagnetic field direction and the source total magnetization direction. For local studies, these directions can be treated as constant values, whereas for regional or large-scale studies they vary significantly. Moreover, when a large area on the Earth is projected onto a single plane, different parts undergo varying degrees of distortion, which can introduce additional errors into the RTP result. To address these problems for large-scale magnetic RTP, we propose a framework based on sliding windows. For each calculation point, the neighbouring total-field magnetic anomaly (TFA) observations are first selected and then projection transformed to plane Cartesian coordinates with the calculation point as the projection centre. Subsequently, gridding and conventional frequency domain RTP are applied to the transformed TFA data, and the value at the central point is assigned as the RTP result for that calculation point. After the RTP results for all calculation points are obtained, the large-scale RTP is completed. Two synthetic tests and a real data application over the contiguous United States all demonstrate the effectiveness and validity of the proposed framework. Furthermore, this framework can also be used for other magnetic transforms and gradient calculations, ranging from regional to global scales.