And by the way implanere, ALL of the factors you mentioned in the beginning of your thread are involved. The intensification of the 'doughnut' effect is directly due to an upward incoming airflow (does not matter whether that is due to ROD or updraft, the effect is identical) which 'pushes' the doughnut up to disc level, when previously it would have been held below the disc (if a/s is low or zero)therfore increasing recirculation of air that has already been accelerated downward. All tht remains is to consider the effect of that upward incoming airflow on the ROOT area of the blade(s) (as you say, it increases a of a and brings that part of the blade to stall) and the anatomy of Vortex Ring and its exponential development is explained. and of course, the more it develops, the more ROD increases....
You can easily demonstrate the effect with a desktop fan. If you rig up some way of opposing the fan's induced flow with another (stronger) airflow it will begin to develop recirculation and its RPM should increase audibly as it takes less 'bite' at unaccelerated air. Also interesting is what happens when you ADD to its induced flow with another airflow, this time from behind. As the induced flow of the fan is increased RPM will again rise audibly. This shows the effect of recirculation in a different (but less accurate) way.
I may struggle to get graphics onto this post as my software is not easy to convert (any tips here would be welcome, but I'll have a go. If not I could post them on my website for a short time) Anyway, I'll try first !
SPS, still wondering how they managed to convince us that 'Nestles' were always called 'Nestle' (accent on the E)
because in the 'Milky Bar Kid' adverts it was always NESTLE'S milky bar.......