This is the Chevallier 91L diverse dataset in ARTS format. As of 2014-12-10 the data is stored in ArrayOfGriddedField4 structure and XML format applicable with arts-2-3 and newer. Older layouts (ArrayOfMatrix) and formats (plainASCII) as used by older versions can be derived from the svn repository (ArrayOfMatrix from ARTS-XML-DATA-2-3-3 and older, plainASCII from ARTS-XML-DATA-2-1-12 and older). The files in old ARTS format might be of interest, because this dataset should be particularly useful for model validation, e.g., to check whether old and new ARTS give consistent results. The dataset documentation by Chevallier is in file "nwp_saf_profiles91.pdf". The original Chevallier data can be found in "~sbuehler/scalc/2007_10_chevallier_91L/original_data/". The FORTRAN extraction program that created the ArrayOfMatrix files as well as the ARTS controlfile to convert it to ArrayOfGriddedField4 format are stored here along with the ARTS xml-data. The z profile was generated with the hydrostatic assumption from the surface pressure and elevation (using the hypsometric equation from Wallace and Hobbs). 1. Full atmospheric states ============================== The full atmospheric states are stored in: chevallierl91_all_X.xml where X for the different data sets is: t q oz ccol rcol These files contain one GriddedField4 for each atmospheric state. (Each file has 5000 states.) Stored parameters are: T[K] z[m] CLW[kg/m^3] CIW[kg/m^3] RR[kg/(m2*s)] SR[kg/(m2*s)] H2O_VMR[1] O3_VMR[1] with following abbreviations: CLW : Cloud liquid water content CIW : Cloud ice water content RR : Rain rate SR : Snow rate Note: previously existing separate clear and cloudy state file have been removed. 2. Misc ======== In future, it is forseen to later generate for each data set: chevallierl91_surface_X.xml It would contain surface parameters (T, wind speed). Then, the clear & cloud & surface or full & surface files together would hold all state information for RT calculations with clouds and variable surface conditions. Let us know if anything here is unclear, so that we can fix it. 2007-10-23 Stefan Buehler 2008-05-30 Stefan Buehler 2012-07-19 Jana Mendrok 2014-12-10 Jana Mendrok