First of all I am grateful for the supportive reception given to comments by a non-aviator. I should stress, in considering what I write, that I am a hydrologist and normally meteorology is merely a means of producing rain for the rivers I study. However, the disciplines are linked, my first degree involved a dissertation on thunderstorms and I am an observer for the Tornado and Storm Research Organisation. Therefore, while it is not directly my discipline, I have a strong scientific interest/responsibility in at least recording anything in this field that I view as significant.
interesting to hear from someone involved with the weather there. I was and am a tad confused by the METAR's coming out of there, going something like BKN006 OVC027CB. Any OVC -- CB condition is something that will catch attention, but a broken layer below that, and the overcast is a bit on the unusual side for this central european met observer.
Can you give a bit more information on this? How would you rate the met conditions as they were reported there with what your experience shows?
If I read the METAR correctly, the most interesting things that I note are the variability of the wind, the number of temporary changes over a short time and presence of cumulonimbus. This accords rather well with my observations during July 2004 and August 2005 in Rostov Region. In each case, there was evidence, or it was possible to infer, electrical activity in the vicinity but not close to my observations. The wind suddenly became sharply variable with, in two cases, a significant downward component.
Work in the USA on tornado formation shows a similar, but seemingly not identical, pattern. In particular, Lemon and Doswell III (1979) report
"observations reveal the existence of a downdraft (originating at 7–10 km AGL) on the relative upwind side of the updraft". In Russia, Kochin (2001) ties downdraft activity to changes in electrical charge within the thundercloud (also interestingly, but in noting the METAR temperature seemingly not relevant in this case, commenting that this occurs in the 20C zone
"where lightning strokes to aircraft are observed most frequently").
One additional observation from my time in Adler, and in particular one trip into the mountains, is how the mountains appear to have a strong localising effect on thunderstorm activity. It seems that a storm in one valley will remain isolated from neighbouring valleys. However, using the work by Lemon and Doswell III, there is strong reason to believe that the winds associated with a storm will not remain quite so isolated. I do wonder if the variable wind in the METAR might also include, or even mask, a similar downdraft component.
Once again, I stress that I am not qualified to make any specific suggestion as to what went wrong with Flight 967, and I note a number of alternative ideas are under discussion. All I can do is give what scientific facts that are available to me and appear useful in the hope that it contributes, in a positive manner, to the discussion.
As you will appreciate, I have spent a considerable time in Russia and find great sadness in this tragedy. Today was a day of mourning in Russia and Armenia.