Extrapolate Molecular profiles to ground

[HavardStridBuholdt]

Continuing from issue #62.

Issue:
Though the height offset in the molecular profiles now is fixed, the meteorological model data still need to be extrapolated down to ground in scenarios where the model ground is higher than the station altitude. The currently used methodology uses the last metrological model data values before the model ground for all values below.

The plot below shows an example of a current set of interpolated meteorological model data and the resulting molecular backscatter for the 355 nm FR channel for Dushanbe (case: 23.08.2025 18:00-19:00) where the red lines indicate the model ground level.

[HavardStridBuholdt]

I have added my first attempt to extrapolate the meteorological model data to the development branch 67-extrapolate_molecular_profiles_to_ground. the extrapolation of the temperature was done by using the average moist adiabatic lapse rate of 5 K/km. The pressure values was extrapolated by using the end values as reference and calculating the pressure according to U.S. standard atmosphere. The result of the extrapolation can be seen in the plot below.

The extrapolation of the pressure values looks good however, the method for the temperature could be improved. Especially as the model temperature decreases by 5 K in the last 100 m before the model ground. Additionally, no extrapolation method is yet added for the relative and specific humidity profiles.

[HolgerPollyNet]

barometric_interpolation_report.pdf
There is a little report about interpolation model profiles - see attachment. One could use virtual temperature as well... However, for our problem, I see no other chance rather using the average temperature lapse rate of -0.65K/100 m to interpolate to ground (compromise between dry and wet conditions). For the humidity we also can just assume constant values (for specific humidity). This would then lead to a decrease in RH towards ground when considering the increasing temp.