Before beginning, it is worth defining the some of the terms used here, as there is some confusion about their definition. When we refer to resolution, we mean the ground resolution of the data, the standard at this time being 25cm in one direction along 1m traverses. By sensitivity, we mean the ability of the instrument to detect different levels of magnetic strength, measured in nanoTesla (nT). In all cases, both fluxgate gradiometers were set to detect magnetic variations in the order of 0.1 nT.
The methodology described here refers to the way in which we routinely carry out surveys, and the subsequent processing of the data.
Two methods of collecting the data are possible. The parallel (also unidirectional or normal) traverse method requires that the instrument is maintained in the same orientation, and that each line is walked in the same direction (i.e. south to north). This requires the operator to walk the first 30 metre line, then return to the beginning of the next line, before beginning the second traverse. The zigzag (also bi-directional) traverse method has the operator returning down the second traverse north to south. If using the FM36 the machine is turned so it is facing in the same direction as in the first traverse. The Bartington Grad 602-1 does not require to be turned round in this way. The choice of traverse method depends to a large extent upon the experience of the surveyor. With experienced surveyors, no appreciable difference will be noted between the two traverse methods.
Each field we surveyed was treated as a separate entity (or site), and divided into a number of whole or partial 30 metre square grids. A base line along one of the field boundaries was established, and then the grids were generally set out using ranging rods and an optical square, although when required either a total station or kinematic Global Positioning System (GPS) was used. Where possible, all surveys carried out from September 2004 have also had the corner of each 30m square logged by a kinematic GPS, to allow variations in slope across the surveyed area to be accounted for. Logging the exact location of each corner is more vital when the work is conducted on a slope, as normal geophysical survey assumes a flat plane. These accurate 3-dimensional coordinates allow the results of the survey to be warped onto a digital elevation model.