Discrepancies and Transformation Between Space-Based and Ground-Based Dst Indices: Insights From Swarm and CSES

The Disturbance Storm Time (Dst) index is a critical metric primarily reflecting the intensity of symmetric part of the magnetospheric ring current. Although satellite magnetometry has enabled the calculation of a space-based Dst index, such measurements exhibit a systematic bias when compared to the conventionally used ground-based Dst. Using 12 years of scalar magnetic field data from the Swarm A satellite and over 7 years of data from the CSES satellite, this study systematically investigates the possible drivers to cause such discrepancies. Our detailed analysis reveals that while the space-based Dst indices from both satellites correlate well with ground observations, they systematically exhibit larger magnetic disturbance amplitudes. The discrepancies between the two indices are modulated by baseline offset, solar activity and local time. The baseline offset originates from the different reference of the two indices. High solar activity widens the difference, likely due to uncorrected ionospheric currents in the space-based Dst, while LT dependencies emerge because satellites sample localized asymmetric part of ring current rather than the global averages. To reconcile two indices, we developed a multivariate linear regression model that converts two Dst indices by explicitly considering effect from solar flux, local time asymmetries, and baseline shifts. Applying this transformation increases the correlation coefficient between space-based and ground-based Dst to approximately 0.95 and reduces the Root Mean Square Error to approximately 6.0 nT for both CSES and Swarm A. The results provide a quantitative characterization of the factors controlling the differences between satellite-derived and ground-based Dst measurements.