NOTES
1.    The liquid/vapor coef for this data is from the actual experiment sonds from the ship.  This will reduce the independents of the radiometer data from the sond data.
2.    There were gain shifts in the 31 GHZ channel that made if necessary to spit the data into groups and process each group independently of each other.
3.    I hand edited out single point outlyers  from the data.
4.    Removed known wet reflector data.  Not that there still could be more but they could not be identified by just the radiometer data.
5.    This data is different from the real time data in the fact that the radiometer coef are modified based on the tip curve and the experiment goes on instead of a fixed value for the whole experiment like the real time data dose.