I'd be interested to learn more if there are any statistically significant results showing clustering - proving that there is spatiotemporal clustering that is different enough from what might be expected at random is pretty hard. I know that Sellafield has been studied in detail many times, with no definitive evidence yet found.