MCSMIP DYAMOND Analysis

---

This repository contains analysis codes and Jupyter Notebooks for the MCSMIP (MCS tracking Method Intercomparison Project) for DYAMOND simulations.

The project details are available in this GoogleDoc.

Batch Processing in Slurm

---

All processing codes are in the src directory. Most codes are parallelized with Dask that can be run on a single node, which typically finishes in 10-15 min.

This script generates specific processing tasks and submits to slurm:

python make_mcs_stats_joblib.py ${PHASE} ${tracker}

Optionally, this Bash script can run all jobs for a single DYAMOND source and a single tracker:

run_alljobs_1source.sh

Below are more details for different sets of processing codes.

Regrid DYAMOND Data

---

CDO scripts used to regrid raw DYAMOND data to lat/lon grids are in the regrid directory.

Standardize Datasets

---

The ${} are command line inputs, examples:

${PHASE}: 'Summer' or 'Winter'

${runname}: 'OBS', 'NICAM', 'SCREAM'

${tracker}: 'PyFLEXTRKR', 'TOOCAN'

${env_varname}: 'intqv', 't2m'

${start_date}: '2016-08-01T00'

${end_date}: '2020-03-01T00'

--

• Combine PyFLEXTRKR MCS mask files to a single file:

python make_mcs_maskfile_singlefile.py ${PHASE} ${runname}

• Standardize other tracker mask files:

python unify_mask_files.py ${PHASE} ${runname} ${tracker}

• Standarize DYAMOND environmental files:

python unify_env_files.py ${PHASE} ${runname} ${env_varname}

Download DPR swath data

python download_dpr.py ${outdir} ${start_date} ${end_date}

${outdir}: directory where to store the downloaded data

${start_date}: start date and time, e.g. 2020-02-01T00:00:00

${end_date}: end date and time, e.g. 2020-02-28T23:00:00

Visualization

---

• Tb + precipitation + MCS mask animation for any DYAMOND sources and MCS trackers:

python make_mcs_quicklook_animation.py ${PHASE}

• MCS swath masks and counts for any DYAMOND sources and MCS trackers:

python make_mcs_maskswath_plots.py ${PHASE}

Regrid data to ERA5 grid:

---

• Regrid DYAMOND environment data to ERA5:

python regrid_envs2era5.py ${PHASE} ${runname} ${env_varname}

• Regrid Tb & precipitation data to ERA5:

python regrid_tbpcp2era5.py ${PHASE} ${runname}

• Regrid MCS mask to ERA5:

python regrid_mcsmask2era5.py ${PHASE} ${runname} ${tracker}

Regrid GPM DPR swath data to IMERG grid:

python regridding_dpr.py ${data_path} ${outdir} ${start_date} ${end_date} ${target_grid}

${data_path}: directory that contains the downloaded DPR data files

${outdir}: directory where to store the regridded files

${start_date}: start date and time, e.g. 2020-02-02T00:00:00

${end_date}: end date and time, e.g. 2020-02-28T23:00:00

${target_grid}: directory of a file with the target grid, e.g. data/olr_pcp_Winter_OBS_2020022612.nc

Calculate Statistics

---

• Calculate MCS track statistics from mask file:

python make_mcs_stats_from_maskfile.py ${config_file} ${tracker}

• Calculate time-mean MCS frequency & precipitation map:

python calc_tbpf_mcs_rainmap_mcsmip.py ${config_file} ${tracker} ${start_date} ${end_date}

• Global mean precipitation time series:

python avg_global_rain_timeseries.py ${PHASE} ${runname}

• Global mean environment time series:

python avg_global_env_map_timeseries.py ${PHASE} ${runname} ${env_varname} ${start_date} ${end_date}

• Tb and rain rate histogram:

python calc_tb_rainrate_pdf_byregion.py ${PHASE} ${runname} ${tracker}

• Extract 2D environments for MCS tracks:

python extract_mcs_2d_env.py ${PHASE} ${runname} ${tracker} ${env_varname}

• Average 2D environments for MCS tracks:

python avg_mcs_track_env_space.py ${PHASE} ${runname} ${tracker} ${env_varname}

• Calculate mean MCS precipitation bin by environment:

python calc_mcs_pcp_envs_pairs.py ${PHASE} ${runname} ${tracker} ${env_varname} ${start_date} ${end_date}

Analysis

---

Analysis and plotting are in the Notebooks directory. More details will be added later.