Curie Point Depth Inversion From Magnetic Data Using a Cauchy-type Integral Interface Framework: Application to the Gonghe Basin (northwest China)

Mapping crustal thermal structure is fundamental to studies of lithospheric rheology, tectonic evolution, and deep geothermal resource assessment. Curie point depth (CPD) estimates derived from magnetic anomalies are commonly used to infer geothermal gradients, but many approaches remain sensitive to simplifying assumptions about magnetization. Here, a computationally efficient, spatial-domain framework is developed to invert CPD topography by representing the CPD as an effective magnetization-contrast interface and computing its magnetic response using Cauchy-type surface integrals. This formulation replaces three-dimensional volume integration with a two-dimensional surface integral while preserving the governing potential-field physics, which facilitates high-resolution forward modelling and regularized interface inversion. Synthetic experiments are conducted to evaluate numerical accuracy and inversion robustness. The method is applied to magnetic anomalies over the Gonghe Basin on the northeastern Tibetan Plateau, a high-temperature geothermal region. The inverted CPD is interpreted as an effective magnetic-thermal boundary conditional on the assumed susceptibility model, and is compared with results from spectral techniques, equivalent-source reconstructions. Finally, the CPD constraints are integrated with independent geophysical and geological information to construct a three-dimensional temperature model of the Gonghe Basin, which provides quantitative insight into the distribution of thermal anomalies and the likely heat-source characteristics and driving mechanisms.