Further examination of divergence on 1982

In working on the chromosome painting I noticed there seem to be a really high number of SNPs between SCA and SCB in 1982, like we had observed before. So I did a PCA using just the SNPs from 1982, and only the section of 1982 that is past the large indel, so that should not be influencing what we see. I also included D Barb for perspective. First, here is a PCA using only the SNPs from 951, one of our largest scaffolds. You can see that D Barb is very different from everyone else, and that the W ponds, D10, and D8/surrounding all cluster together.

However, the story is very different when you look at the SNPs from 1982. Also, interestingly, there appear to be three clusters, with hybrids between some of these. Would love to hear your thoughts on this. Further support for something interesting going on on this scaffold? Of course, maybe other scaffolds would also look this way if I did a PCA on their SNPs? I guess I need to test more to see if this is really something unique? I know I should be doing reading and not data analysis, but was trying to figure out how much I can say about things.