Re: Frozen rivers
Perhaps we can agree on a point here, I hope. A stream or river that is born primarily of run off in its watershed and considering 'that' stream or river is in a zone where temperatures are conducive to freezing, the entire waterway may freeze over. Conversely, a stream or river that acquires a good percentage of its flow from underground sources may exhibit many areas that do not freeze over due to the general higher water temp.
One of the differences but not the only difference between a 'freestone' flow and a spring creek are as described in the preceding paragraph. Slate Run, located in North Central PA. can freeze over completely from the upper branches all the way down to the town of Slate Run. The fish seem to winter over in the deepest pools and the population always was stable. The occurrence of 'anchor ice' in the shallow or riffles, did have a noticeable effect on the hatches in years following an anchor ice event. A real bad and prolonged freeze of all the shallow areas could and often did result in the loss of many nesting beds used by the brown & brook trout the preceding fall. having fished the run often for over 20 years I noticed the relationship between harsh freezes and the overall fish population. There were obvious holes or voids in whole generations of young in some areas. This was noted by the lack of sightings of 'young of the year' taking refuge in the shallows and pools along the shore of the run. I could not think of any other explanation for this void of fry other than the anchor ice of the preceding winter.
In closing, my opinion is that many a good river and creek freeze over but sustain a population of trout or salmon. Speaking on the spring creeks I've visited over the years, they were the life blood of my winter fishing and on many shelf ice was as bad as it got regardless of the severity of the winter. There are of course exceptions to any example given but what I have taken time to recount is a good account of my observations on freezing waters.