Using the CFSv2 datasets\cite{Saha_2010, Saha_2014}, the vertical profiles of water vapour, temperature and relative humidity are extracted for the grid spaces around Darwin. The Darwin site is located at -12.42^o, 130.89^o (30m ASL). The selected data has:
horizontal resolution of 0.5^o
vertical - 37 levels
hourly resolution
Two comparisons between in-situ water vapour isotopes and TES retrievals. The first is by \citet{Worden_2011} using constructed profiles from the in-situ measurements on Mauna Loa. They used the TES H_2O profile and the daily evolution of the \(\delta\)^2H as the PBL grew and decayed (assuming a constant \(\delta\)^2H.H_2O relationship). The second paper by \cite{Herman_2014} used aircraft in-situ profiles and extrapolation above the top of the flights (> 4000m). These papers mention the importance of convolving the constructed “true” profiles with the TES operator (averaging kernel and a priori) to provide an accurate comparison. The later paper by \cite{Herman_2014}, also determine the sensitivity of the error budget to the true profile. We may be able to do something similar using the different profile models (Rayeligh’s and mixing models). Both these papers show a strong positive bias for the remote sensing data, which is largest in the boundary layer ( 100 per mille).
All code written in python.
Put vertically resolved data into a single netcdf file
Write code to construct isotope profile (fixed \(\delta\)^2H)
Open Rayleigh Model
Closed Rayleigh model
mixing model
include in-situ constraint for profiles
flexible model?
\citet{Worden_2011} approach to matching in-situ H2O.dD to pressure level
validation cfsv2 profiles and water vapor
temperature and water vapour profiles vs aeri and/or sondes
surface temperature and water vapour against local met