I like your hypothesis. Back in the 70's I was studying stream improvement/reclamation strategies (we don't have cold-water species in KS but I was still interested). One of the best places for finding working strategies seemed to be WI. I remember that with good improvements, average stream temps fell. I would think that similar "improvements" in the watershed could lead to the same result assuming that changes in the watershed might lead to increases in ground water sources for the stream. Perhaps your "spring" creeks are becoming more "springy"