Reply to N. C. Steenland by the authors

Abstract

In response to Steenland’s points we offer some comments. The spectral division results of Figure 6 reproduce the original model to a greater degree of accuracy due to the longer (implicit) filter length used (256 points) compared to 37 points for the Kalman filter run of Figure 4b. Obviously, the spectral division method advocated by previous workers (e.g., Grant, 1973) could have been used to calculate an apparent susceptibility map similar to Figure 9. However, it was our intent to provide an alternative approach that would not suffer from the limitations inherent to processing in the frequency domain.