I have recently been asked this question. Okay, I’ve been asked this questions in a number of different ways: Do you set up a grid or do you run your ground penetrating radar and just mark what you see? Do you use a grid or a gps? Does your survey grid have to be square? Any way you ask it, it all comes down to this: Do you grid, or not grid?
When setting up a grid it is good practice to attempt to set your grid up square since this will provide the most systematic survey conditions and produce transects that are the same length. However, grids do not have to be square. If you attempt to set up a square survey grid, but the actual shape is a rhombus (yea, I said rhombus), you can correct for the misalignment. This type of shift in a survey grid often occurs when the survey area contains obstructions making it difficult to verify the accuracy of your survey grid length and width.
This issue can be corrected for by collecting GPS coordinates at each corner and then importing time-slice maps into a GIS program. By using the GPS coordinates to georeference your grid, the GIS program will plot the map to its actual shape, even though the processing software produced a map that was square, because the GIS program doesn’t care. It will place the corners of your grid in their actual place on earth and transform your data into a rhombus.
Square GPR survey grids may not even be appropriate in all cases. In a study published in FastTIMES a few years ago where the surveyors were mapping tree roots, they had incredible success using circular grids. Going around the trees gave them full coverage of the tree-root footprint allowing them to create a model of the tree roots in the subsurface.
In other cases, grids may not be necessary at all. There are a number of variables that must be considered when preparing for a survey, two of the most important of which are time and target size. These variables are going to ultimately drive the resolution of the survey. If the project is short on time because a client doesn’t want to pay you to locate a pipe using GPR, but you know it is there, and it is made of PVC, and your EM locator can’t locate it, rapid data collection strategies to pinpoint the pipe and flag it may not include grid set up and systematic coverage.
Or, if your instrument has the capability to integrate GPS, then you could couple the GPS unit and cover the area most effectively without setting up a grid.
Or, if you have enough people who are willing to rock the total station, then each transect beginning and end could be shot in and later plotted in the GPR software.
Or, if you are not searching for small targets, but are mapping geological structure, then strategically placed transects to capture a representative sample of the stratigraphy in multiple directions might be the best way (again, preferably georeferenced with GPS coordinates).
So the takeaway is that each survey is unique. I am of the belief that the more systematic you can be in your survey, the better off the data will be and the more confident you will be in your interpretations. But in the real world, every project is different and time, target size, funding, goals, and other variables will drive your data collection strategy.