Changchun Yin

Latitude‐dependent systematic biases in satellite magnetic vector measurements can significantly contaminate the estimation of external magnetic fields and electromagnetic induction signals. Unlike previous geomagnetic satellite missions, magnetic vector data from the China Seismo‐Electromagnetic Satellite (CSES) exhibit pronounced latitude‐dependent biases (latitude effect) primarily caused by boom deformation and sensor misalignment, posing challenges for the reliable use of these data in applications. In this study, we present a multiscale and directional framework to systematically characterize and correct the latitude‐dependent biases in CSES magnetic vector measurements. By removing the modeled internal geomagnetic fields and analyzing the residuals in the curvelet domain, we identify the dominant spatial scales and directional features associated with the latitude effect. A targeted multiscale correction strategy is then implemented at characteristic scales and directions. The effectiveness of the proposed correction approach is evaluated using statistical analyses, cumulative probability distributions, spherical harmonic coefficients, and magnitude‐squared coherence (MSC). The corrected data show substantially improved temporal coherence, reduced systematic amplitude distortions, and more stable inducing and induced field coefficients, indicating enhanced physical consistency of the external magnetic field estimates. These results demonstrate that, although the latitude effect is specific to the CSES mission, the proposed multiscale bias analysis method is essential for the reliable use of satellite magnetic vector data involving external field estimation and electromagnetic induction, which shows good potential for space and Earth scientific research.

In this paper, we implement the field-based secondary-field 3-D finite-volume method (FVM) and the rapid direct method of explicit Fréchet derivative for the crosswell electromagnetic (EM) measurement. The EM fields in the 3-D formation are decomposed into the background fields from the layered vertical transverse isotropy (VTI) formation and the secondary fields scattered by local anomalies. The 3-D scattered current density excited by each transmitter in local anomalies is discretized into numerous discrete current density elements. We employ the background field library and multivariate interpolation methods to enhance the efficiency of a large amount of calculation of the background fields for scattered current density. The discrete equations about the scattered electric field for all transmitters are solved simultaneously using the direct solver. The high-precision crosswell EM responses for multiple receivers at different depths are rapidly computed through interpolation and projection operators combined with the parallel technique. Based on the 3-D FVM and direct solver, we derive the explicit Fréchet derivative of the crosswell EM responses that involves calculations of only one forward simulation and simple back substitution. Through the same projection operator and parallel technique, we further enhance the computational efficiency of the explicit Fréchet derivatives. Finally, in the numerical results, the accuracy and efficiency of the forward and explicit Fréchet derivative are validated. The characteristics of the crosswell EM responses are systematically analyzed by deriving quantitative estimates of the explicit Fréchet derivative.

In the above article [1], the phrase “the chessboard model” in the title of Fig. 12 is incorrect and should be replaced with “the USArray data.” The corrected figure is shown below.Fig. 12.Comparison of predicted and observed data from the inversion of the USArray data. (a)–(h) Real and imaginary parts of the four impedance components Zxx, Zxy, Zyx, and Zyy, respectively.

In geophysical inversions, the choice of data misfit measure directly affects the accuracy and reliability of the results. The conventional L2 norm is overly sensitive to large outliers, which may cause overfitting to local anomalies or deviations from the true model. To address this issue, we propose a weighted hybrid data misfit measure that combines the Sinkhorn distance (SD) with L2 norm, aiming to improve the resolution and robustness of three-dimensional (3D) magnetotelluric (MT) inversions. By incorporating SD, the method preserves the sensitivity of L2 norm to amplitude variations while enhancing constraints on data distribution patterns. This balances global fitting with the influence of local anomalies, reduces both overfitting and underfitting. The synthetic experiments demonstrate that, compared with the conventional L2-based inversions, the hybrid SD-L2 approach requires slightly more iterations but achieves better fitting of the diagonal impedance tensor components (Zxx and Zyy). The recovered resistivity models are close to the true ones, and the overall resolution is improved. The method is applied to MT data from the USArray project in the northwestern United States. The results reveal a clearer, deeper electrical structure than that obtained from the conventional L2 inversion. This particularly enhances the imaging of magma conduits, thereby provides more reliable constraints for regional tectonic interpretation.

The Newer Volcanic Province (NVP) in southeastern Australia is a classic example of continental intraplate volcanism located far from a plate boundary, yet the configuration of its magma plumbing system remains unclear. Using three‐dimensional (3D) magnetotelluric (MT) inversion, we imaged a vertically continuous, trans‐lithospheric conductive structure beneath the NVP. This structure consists of a sub‐horizontal zone in the asthenospheric mantle, interpreted as 0.1–1.0 vol.% carbonate‐rich melt, vertically linked to hydrous basaltic melt zones (2–5 vol.%) in the lithospheric mantle and shallow crust. These upper crust conductive zones likely represent remnant melt or metasomatic minerals. These features define a vertically integrated magma plumbing system from the asthenosphere to the surface. The geometry and depth of the anomalies are consistent with scenarios in which the carbonate‐rich melt may form in response to shear‐driven upwelling and edge‐driven convection. These observations provide new constraints on possible mantle processes that drive intraplate volcanism.