A broad learning approach for magnetic reduction to the pole

Reduction to the pole (RTP) is a fundamental method for interpreting magnetic anomaly data. It faces challenges at low magnetic latitudes and uneven observation surfaces, potentially leading to inaccurate localization of magnetic sources. To address these issues, we develop a novel approach under the machine learning framework for achieving stable and rapid magnetic RTP in arbitrary regions. Firstly, a single magnetic dipole with random size, position and magnetization magnitude is utilized in the forward model to generate samples of the original total magnetic intensity (TMI) anomaly data and the corresponding vertical TMI, which simplifies the establishment of training data. Secondly, the broad learning system (BLS) is employed to establish a mapping relationship between the original and vertical TMI data, as the BLS exhibits fast network training process due to its concise network structure and strong mapping ability. The combination of sample generation strategy and BLS enables rapid network construction and RTP completion. The feasibility of the proposed method is validated by both synthetic models and field applications. Synthetic data tests suggest that the BLS approach can be effectively applied for magnetic RTP for various regions, including the magnetic equator and undulating surfaces. It demonstrates superior accuracy and noise robustness compared to the wavenumber domain method. Field applications show that BLS prediction results facilitate further exploration of the mining area and save more than 91% of the time compared to the equivalent source and deep learning methods. This study is useful for the rapid and accurate localization of magnetic sources and providing horizontal positional constraints for three-dimensional magnetic inversion.