@HN39 et al. Sorry for delay in returning to an internet connection.
As a mere calculation exercise if we take a 5% variation in G from 1.4 to 1.33 as per your post (#1642) and leave Pstag/Pstatic the same in M=sqrt((2/(G-1)){(Pstag/Pstatic)^([G-1]/G)}) at 0.8, from the pitot, then if we assume SAT=-42C (231K) for this exercise, taking M=0.8 we get a TAT of -12.4C for G=1.4 and -17.6C for G=1.33 ie the calcs show a sudden 5 deg change if we assume all other parameters remained constant from M=sqrt[{(Tt/Ts)-1}{2/(G-1)}]. (The other air parameters wont remain constant but I am merely showing how measuring data from one G value atmosphere and calculating others using a different G gives incompatible results.)
For some reason my formulas didnt want to paste from my MathCad program, so in case I have missed some brackets see
Aircraft Temperature Measurement Page My point is that with a change in G there become changes in other parameters that change if the ADU calculates them using a different G than that of the outside atmosphere and the changes do not appear to be insignificant from a cross-checking perspective. Hence my concern is 'error checking' flagging issues that confuse HAL.
I was assuming there would be areas in the Cb that were above saturated as the humid tropical air provides the energy source to power the storm. My reference to the dry and saturated graphs were to show that G decreases with the addition of H2O not to suggest we were operating in this range. I was suggesting AF447 entered an 'abnormal' atmosphere.
There is a section in Atmosphere Ocean Dynamics Volume 30 International Geophysics by Adrian E Gill pg 44 equation 3.3.5 (via google books??) that shows this Gamma change.
There is some research on Effective ice particle densities for cold anvil cirrus
CRYSTAL-FACE Publications. See Heymsfield and Jensen half way down the list.
Trust this shows why I raised the issue.
kind regards. Ian