Judging by some of the responses on here about weather not being a big deal, I beg the differ. Here is everything laid out. I put this together using Level 2 dual-pol data which is the highest resolution radar data available. The images are valid at 18:38:25Z and is the .5 degree base reflectivity and Spectrum Width from the Houston KHGX that was downloaded from the NOAA archive and opened in Gibson Ridge's GR2 Analyst.
It clearly shows the cold front (black arrows) draped SW to NE across the Houston area and its movement to the SE at around 10-12kts. Before the cold front passed over the area, surface observations show that Ellington Airport (KEFD) and Scholes International in Galveston (KGLS) both had winds varying from 200/210 varying between 12 and 15kts. As the front passed over IAH, EFD, and finally GLS, winds immediately shifted to 310 at 10-15kts with IAH Gusting to 24kts at 18:02Z along with a temperature drop of around 10 degrees from around 77 to 67/68.
At the time Atlas began to show signs of trouble it was positioned directly on top of this front. NWS Mesoanalysis Data shows that the entire area had at least 60kts of bulk windshear at the time of incident along and just ahead of the cold front. (Graphic Included)
Continuing along, the observed 18Z Sounding from Lake Charles, LA had a wind profile that I will lay out below.
10,000ft - [email protected]
9,000ft - [email protected]
8,000ft - [email protected]
7,000ft - [email protected]
6,000ft - [email protected]
5,000ft - [email protected]
4,000ft - [email protected]
3,000ft - [email protected]
2,000ft - [email protected]
1,000ft - [email protected]
Surface - [email protected]
Finally I have enclosed another dual-pol radar product called Spectrum Width. Spectrum Width depicts a measure of velocity dispersion. In a radar bin, it provides a measure of the variability of the mean radial velocity estimates (movement) due to wind shear, turbulence, and/or the quality of the velocity samples. The proper use of Spectrum Width can help the severe thunderstorm and tornado warning decision process. It is used to estimate turbulence associated with low-level boundaries, thunderstorms, mesocyclones in supercells, and mesovortices in quasi-linear convective systems (QLCSs)
Low values = smooth flow
High Values = depict variability in movement, turbulence, and chaotic flow.
The SW values along the cold front at 18:38Z in the exact location of Atlas 3591 show pretty high values that would be consistent with moderate to severe turbulence and a disrupted, chaotic air steam.
I am usually not the speculative type, but after looking at and analyzing the weather data, I feel very confident weather played a part in this terrible accident. Only time will tell as the NTSB works the investigation. Hope this information is helpful to any and all.
My thoughts and prayers are with the families of the flight crew. Godspeed.