script for getting and analyzing gridded aerosol products
GEOS-5 data from: https://gmao.gsfc.nasa.gov/GMAO_products/NRT_products.php
Instructions for running scripts to download GEOS-5 2D aerosol data, partition it into seperate forecast files and finally put it in a format (temporary ungrib files) ready to be used in metgrid.exe:
- Compilation of the fortran code needs to be done first
$ ifort -convert big_endian -I/usr/include write_aerosols_for_metgrid.f -L/st1/luong/lib/netcdf-4.3.3.1/lib -lnetcdf -lnetcdff -o write_aerosols_for_metgrid.Linux {likely will need to change pathways to netcdf library}
- Run python script to download and process GEOS5 2D aerosol data
Library Requirements are: os, subprocess, glob, sys, numpy, netCDF4, pandas, requests, get_aeronet
A soft link to get_aeronet.py script from the get_aeronet repo in UARENForecasting need to be created
Now the script can be run: $python create_geos5_aod_aux_input.py
This should create GEOS5.FILE:* type ungrib files
After the GEOS5.FILE:* type ungrib files have been made, namelist.wps needs to be told to read them at the metgrid.exe stage. In namelist.wps add something like this;
&metgrid fg_name = 'FILE','GEOS5.FILE'
Then, the instructions from David Ovens are applicable from 4.1 onwards on this page https://atmos.washington.edu/~ovens/running_wrf_with_geos5_aerosols/README
Attached is a the namelist.input file being used for real.exe/wrf.exe. The key lines are:
auxinput15_inname = 'wrfaerinput_d',
auxinput15_outname = 'auxinput15_d_',
auxinput15_interval = 180,
io_form_auxinput15 = 2,
force_use_old_data = .true.
&physics mp_physics = 8, -1, -1, cu_physics = 3, -1, 0, ra_lw_physics = 4, -1, -1, ra_sw_physics = 4, -1, -1,
aer_opt = 2, aer_aod550_opt = 2, aer_angexp_opt = 1, aer_angexp_val = 1.3,