botany - Are trees still carrying their leaves more likely to be felled by an autumn storm?

Is there evidence that trees still carrying their leaves are more likely to be felled by an autumn storm than trees already having lost their leaves?

At first sight this seems plausible (and might be easily verified statistically), but at second sight it seems mysterious, considering a) the weak forces that keep leaves attached to their branches (at least in autumn times), b) the flexibility of the branches, and c) the chaotic way in which they are shaken by a gust.

Anyway, after storm Xavier which passed Germany and Poland on the 5th and 6th of October, this was spread ("kolportiert") by the media and in private communication:

Is there a thorough biomechanical model which explains whether or why this is true or false, resp. to which extent?

I would suspect that in this model a) eigenfrequencies (of branches, limbs and stem) and b) resonance will play a role.

(This model would give an interesting analogy to the sudden generation of action potentials in neurons (corresponding to the breaking of a limb or stem) by spatio-temporarilly integrating the synaptic inputs (corresponding to the shaken leaves) on the dendritic tree [sic!]. Addendum: There's a great book that brings botanical and neuronal trees together: Giorgio A. Ascoli, Trees of the Brain, Roots of the Mind.)

(Note, that one short gust almost never will fell a tree. It must be a rather long gust, or an appropriately long and/or timed sequence of shorter gusts.)

---

Addendum: The way how fluttering ion channels (on a very short time scale - grey dots below) give rise to a rhythmic and possibly resonant temporal pattern (on a much longer time scale - red or blue line above) may be compared to the oscillations of the tree as a whole, whose "rhythm" (red) cannot be detected when looking at single leaves fluttering (grey):

See also here.