I have examined the data of Lago di Fosses from October to December in the period when the data logger worked (with snow, without snow, before and after days with temperatures below -10°C when the lake could has been frozen) but I have found nothing of interesting.
In fact it is not easy to analyze and understand the few cases with snow at ground and "black ice" on the lake.
About you theory of the air circulation triggered by the difference of radiation balance between the frozen lake (without snow) and the surrounding meadows covered by snow, I think it is credible, but it is probably very weak and not so strong to influence the radiation cooling of that huge frost hollow.
But, you know, we should do an accurate monitoring of temperatures and winds there, to have the answer.
Anyway, I'm looking forward to hearing your news from Lake Sämtis!