Interestingly enough, the plots with the most frequent standing water (which have, quite justifiably, some proper edged beds with channels about 12" deep around them) are at the top of the site.
We understand (from a session with a geophysicist - yes we did use the machine but there was no Tony Robinson around!) that the raised parts of the site are the end of a former glacier (a terminal moraine), and so I think there's a clay cap at the surface level of those higher parts of the site. Plus an underground spring somewhere that comes to the surface about halfway down the 'hill' and makes life extra interesting at times.
The plot which is most affected by the spring is one where we are not too forceful about the willow ...