The magnetometer data were processed in the following manner: (a) the COG 

 as collected by the Differential Global Positioning System (DGPS) was smoothed. 

 This removed the pitching by the sea conditions of the research vessel from the 

 navigation data which were collected every 2 sec. The SOG was smoothed for 

 the same reason, to remove the variations due to the vessel moving from sea 

 conditions. Subsequent measurements, collected every 2 m, were used to 

 compute the magnetic gradient parallel to the traverse. This gave a very close 

 approximation of the total horizonal magnetic gradient since the gradient was then 

 both perpendicular and parallel to the track lines. This gradient, either negative 

 (dashed lines) or positive (solid lines) was used to vector toward and triangulate 

 upon the pole and dipole locations of various ferrous objects. Examples of these 

 data are shown in Appendix A. Three figures are generated for each anomaly, 

 the upper left is the total anomalous magnetic field (in nanoTeslas x 10 4 ) as 

 measured by the two cesium vapor sensors separated by 2 m traverse to the track 

 line of the vessel. Sensor "A" is to the left or port of the course, and sensor "B" 

 is to the right or starboard of the track line. The right side of the figure is to the 

 south or north as indicated by "S" or "N" in the caption. With only a few 

 exceptions, the majority of the detected magnetic objects have a magnetic "low" 

 response to the north of the magnetic "positive" response. In the northern, 

 latitudes such as New Jersey, this is indicative of anomalous magnetic effects 

 originating from mainly the induced magnetic field effect, and only a smaller 

 portion is from remnant magnetization. Ultimately, if physical measurements on 

 some recovered items demonstrate that this is correct, the data can be processed 

 using more straightforward and simpler assumptions. The horizontal magnetic 

 field gradients are displayed in the lower left figure. These are "G" "east-west" 

 gradients (perpendicular to the track line) and "H" north-south" magnetic 

 gradients (parallel to the track line). Both measurements are in nanoTeslas/meter. 

 The right figure on each page displays the smoothed track line. The portion of the 

 track line which is inclusive of the detected anomaly is plotted in relative northing 

 and easting locations (units in feet). The intensity and horizontal direction of the 

 resultant magnetic gradient are then plotted in reference to the smoothed COG. In 

 these plots, the length of the magnetic gradient vector is proportional to the 

 strength of the gradient. Since the target objects generally respond as dipoles 

 (each generates a positive [south end] and negative [north end] magnetic anomaly) 

 the gradient vector from the track line is dashed in its decreasing direction and 

 solid in its increasing direction. This is necessary since a magnetic low anomaly 

 on one side of the track line can have the same gradient as a magnetic high on the 

 opposite side. However, as the sensors pass by the anomaly, the gradients will 

 converge on the source location. From this method, ordnance-type dipole objects 

 can be even further identified by the location of a magnetic negative gradient 

 (dashed lines) being generally immediately northward of a magnetic positive 

 gradient (solid lines). 



Almost all of the detected magnetic responses were locatable within distances 

 of about 3 m on each side and beneath the cesium vapor magnetic sensors. This 

 gives a detection and location swath width of about 8 m for survey purposes. 

 Over 95 percent of the detected anomalies were determined to X-Y locations of a 

 meter. The major exception to plotting an object's location were circumstances 

 where it was located in a debris field and thus in a complicated magnetic gradient 

 environment. Many of the objects are most likely elongated dipole objects (much 



Chapter 6 Magnetometer 



25 



