The graves that were marked are likely real. If you look at the image, there is no GPR return at the crosses. That is normal. GPR can only detect coffins, not decomposed bodies. That GPR image that was posted in the article is essentially useless unless they're detecting coffins. It is highly likely that wooden coffins were used if any coffins were. My interpretation of that product is that the returns that are marked as graves (aside from the crosses) are likely not coffins, as they are random and the marked graves show a pattern of decomposition which would render a GPR product nearly useless.
I'm sure there were bodies buried there, but the analysis given is flawed IMO!
“The graves that were marked are likely real.”
You realize there were originally no markers in that field at all and the pipe crosses were put up decades later by people who really had no idea exactly where the burials were, right?