High Frequency Data

Data from sub-monthly frequency files (i.e. high frequency data) can be processed the same way as monthly data, however the ncclimo commands to extract the time-series files is slightly different. Here’s an example of extracting high-frequency time-series:

in_dir=./raw_data_path/
out_dir=./regridded_time_series/
native_out=./native_grid_time_series/
flags='-7 --dfl_lvl=1 --no_cll_msr --clm_md=hfs'
variables='PRECT TS'
start_year='1850'
end_year='2000'
years_per_output_file='50'
mapfile=${DATA}/maps/map_ne30v3_to_cmip6_180x360_aave.nc
ncclimo ${flags} -v ${variables} -O ${out_dir} -o {native_out} --map=${mapfile} -c historical -i ${in_dir}

Once you’ve extracted the time-series files, simply use the path to where they’re stored as the input path argument to e3sm_to_cmip, and supply the --freq flag with the appropriate frequency.