Digital data grids for the magnetic anomaly map of North America

We obtained all available aeromagnetic surveys. Some data were digitized from analog maps if digital data were unavailable. The remainder were processed using digital flight-line data.

Details of data acquisition are described in the booklet that accompanies the printed map.

The marine data were obtained from the National Geophysical Data Center of the National Oceanic and Atmospheric Administration spanned the years 1958 through 1997.

Aeromagnetic data were processed so that residual magnetic values were calculated in a consistent manner:

- The DGRF, updated to survey date, was removed

- The flight elevation was used to mathematically calculate the equivalent magnetic field at 1,000 ft. above terrain

- obvious errors were corrected

- x-y locations were calculated for the described DNAG projection

- data were gridded at 1/3 - 1/4 of the flight-line spacing, then regridded to 1 km.

Using various USGS in-house programs (before 1999) or Geosoft/Oasis Montaj (after 1999), gridded data were mathematically merged together. For the U.S. part, scientists merged groups of surveys by state or groups of states. States were merged into regions, and regions into countries.

Procedures and software used to compile the 1-km grid of magnetic data for Canada are detailed on the Geophysical Data Centre website.

Creation of a wavelength filtered grid.

Because wavelengths greater than roughly 150 km are unreliable in the compilation, applying a high-pass wavelength filter would appear to be a viable solution to remove these unreliable wavelengths. However, removing wavelengths less than 500 km from the merged grid creates artifacts, such as spurious separation of continuous anomalies. Therefore, we removed anomalies with wavelengths greater than 500 km from the merged grid to reduce the effects caused by the erroneous long wavelengths but maintaining continuity of anomalies. The correction was accomplished by transforming the merged grid to the frequency domain, filtering the transformed data with a long-wavelength cutoff at 500 km, and subtracting the long-wavelength data grid from the merged grid.

An equivalent source method based on long-wavelength characterization using satellite data was used to correct for long-wavelength shifts.

We produced an aeromagnetic grid in which the wavelengths longer than 500 km have been replaced by downward-continued CHAMP satellite data. Steps 0 and 6 were performed by Bob Kucks. Steps 1-4 were performed by Tiku Ravat. Step 5 was performed by Jeff Phillips.

0. The North American 1 km merged grid was decimated to 5 km.

1. This 5 km grid was converted to a 0.05 degree grid. This grid was low-pass filtered using a Gaussian filter with a 500 km cutoff, then decimated to 1 degree.

2. A joint inversion of this 1 degree low-pass aeromagnetic grid and CHAMP satellite data, with the aeromagnetic data weighted very low, produced a stabilized downward continuation of the CHAMP data.

3. The inverted data were interpolated to 0.05 degree and again low-pass filtered using the same Gaussian 500 km filter to remove

4. The low-pass grid from step 1 was subtracted from the original 0.05 degree aeromagnetic grid to create a 500 km high-pass aeromagnetic grid. This grid was added to the low-pass inverted grid from step 3 to get a corrected 0.05 degree aeromagnetic grid.

5. The corrected 0.05 aeromagnetic degree grid was projected to the DNAG projection and regridded to 5 km. This was subtracted from the decimated 5 km aeromagnetic grid to generate a 5 km correction grid. A matched filter was used to remove short-wavelength artifacts resulting from the projection and regridding process.

6. The resulting 5 km correction grid was regridded to the original 1 km grid and subtracted from the original 1 km aeromagnetic grid to generate the final 1 km corrected aeromagnetic grid.