GitLab

A quick update of the chemistry scheme to the JPL19 catalogue. Note that now all functions with jpl16 in name actually use jpl19 data. Was too lazy to change all of these names (which would also break many start scripts)

Some changes to the tracer option CRISTEN_TRACER that proved to be helpful. Some other changes in this commit seem only to be changes in spaces

Discovered that get_data_path has never been uploaded to the repository, although it is needed to run Atlas... Things like this happen if you are the (almost) only user

Large update to make all parts of Atlas work with ERA5 data. In addition, took the opportunity to overhaul the parallel trajectory functions to catch up with the changes and improvements in the main trajectory function.

... and update geopotential height calculation to take into account the change in treatment of orography

Changes to loading orography. Load only once for Interim etc. Add missing function to load orograraphy for ERA5

Fix bug that minimum temperatures and crossing of 80 percent psurf were not recorded if happening at the first trajectory time step

Add possibility to use ERA5 for convection and heating rates, change time step for ERA5, update some default pathes, streamline code for loading heating rates and convection variables a bit

Change time steps loaded from ERA5 files, insert a dot in definition of theta_factor to indicate elemtn-wise division

Added checks in the loading routines for ECMWF data if latitudes start at -90 degrees, longitudes start at 0 degrees and levels and at model level 1

All NCEP reanalysis data was not read in correctly. It seems that this was due to all variables saved as a short not as float. I am not sure if this always was the case and if the routine was working correclty when I wrote it, because the netcdf files used different data types then. Anyway, this has been corrected by introducing the correct typecast.

1. Switch off vertical diffusion during convection. 2. Add zeta coordinate as possible coordinate for vertical diffusion. 3. Change default path.

Do heterogenous chemistry in sulfur scheme not only in convection, but also in non-convective clouds

Introduce a scaling function for nitrogen, change the default initialization for most species from MLS version 3 to version 4

Do use the changes in the species initialization introduced in the last commit in the default init fucntions for the JPL11 and JPL16 chemistry models