A high-precision magnetization vector inversion method with normalized source strength and geometric shape partitioning constraints

This is the authors' abstract. We don't add key points for this paper.

Considering the effect of remanence, the magnetization vector inversion method can recover three orthogonal components of total magnetization vector using the magnetization intensity as a constraint, which makes is a more practical approach for the inversion of magnetic. However, inversion precision is relatively low when the magnetization directions of adjacent geological bodies differ significantly. Based on the characteristic that the normalized source strength (NSS) is weakly sensitive to the magnetization direction, NSS is adopted instead of magnetization intensity to constrain the three orthogonal components. To address magnetic anomalies arising from the superposition of multiple magnetization directions, a geometric-shape-based partitioning method is further developed to realize high-precision magnetization vector inversion constrained by NSS; this approach is termed the NSS-GSC inversion method. Synthetic tests show that the NSS-GSC inversion method yields more accurate inversion results compared with the magnetization intensity method, and exhibits good anti-noise performance. The NSS-GSC method is applied to the Huanggangliang-Ganzhuermiao metallic ore cluster in China to obtain the distribution of highly magnetic bodies located near NE- and NW-trending faults, with top depths ranging from 178 m to 384 m. This provides robust evidence for further mineral exploration.