The goal of the observation preprocessor is to provide a framework for reading in various data types and outputting injested information to common formats.
- Source g5_modules from GEOSgcm:
source /discover/nobackup/asewnath/GEOSgcm/@env/g5_modules.sh
. Feel free to use your own GEOSgcm build for this. git clone https://github.com/asewnath/observation_preprocessor.git
cd observation_preprocessor/build
- You can leave
ESMADIR
inGNUmakefile
as is or you can change the field to point to your build of GEOSgcm. make install
cd ../
Now you have the executable run_preprocess
to start preprocessing files.
./run_preprocess <input file directory> <source> <output type> <output file directory>
where
<input file directory>
: directory that contains files to be preprocessed<source>
: input file type<output>
: output file type<output file directory>
: directory where output files will be saved
Possible sources: eumetsat, jma, wisconsin, noaa, kma, avhrr Possible output types: prepbufr, text
To process feature-tracked winds from CIMISS into prepbufrs, run the following:
./run_preprocess /archive/u/asewnath/reanalysis/GOES/raw_data/data_stage_test/ wisconsin prepbufr ../output_prepbufr
. This will take files from the archive and process them into prepbufrs which will be saved in the indicated output directory one level above the git repository.