Swarm Scaling Issue in Scalar Calibration

Investigations of the vector residuals of Swarm data versus field models reveal issues in the y-axis scaling for both the ASM-V and VFM Level~1b data. Here are some results pertaining the scalar calibration of the VFM using the scalar response of the ASM as reference (as usual). The currently applied calibration scheme poses constraints on the temporal evolution of the scaling factors for the three VFM axes:

1. For the first part (years) of the Mission, the y-axis scaling evolution is constrained to be in the middle of the x- and z-axis scalings
2. For the remainder of the Mission, the scaling evolution of all three VFM axes are constrained to be identical

Swarm Bravo

Using the new calibration scheme - with constrained y-axis scaling - the residuals look identical:

However, when looking in detail at the scalar residuals, the new scheme produce residuals of basically the opposite sign:

This is the overall picture for the entire mission:

In general, the changes in the field vector components using the new calibration scheme are very small. The following plot shows the changes in the three components (solid lines with crosses, scale to the left) together with the full field (dotted lines, scale to the right):

Finally, a plot of the changes in B_y as a function of B_y itself. Although this new calibration scheme was triggered by the vector residual analysis vs field models, the change in B_y with the new calibration scheme seem to be an order of magnitude too small compared to the field residual analysis (around 40 ppm, or 0.4~nT in a 10~uT field).

Swarm Alpha