Description

This repository presents a post-processing technique for generating ensemble consensus precipitation forecasts, accounting for spatial and intensity differences among individual members.
📄 Read the paper: https://doi.org/10.1175/WAF-D-23-0229.1
💻 Data is available at https://osf.io/6bame/

2023 FFaIR Summer Experiment results

🌦️ You can explore the 2023 FFaIR Summer Experiment results here.
https://caps.ou.edu/clee/ens/ens_view.php

1. Required packages/libraries

• MPI
• Gfortran
• ecCodes (Version 2.27.0), for GRIB I/O
• NetCDF

2. Build program

• Set the paths of ecCodes and NetCDF in 'buildit_mpi' (Already set)
• Execute 'buildit_mpi' in the 'build' directory in the project

3. Run program

• Command line : ${path_of_program}/ensalign ${YYYYMMDDHH} ${fhr} ${input_file_path}
• Command line (with MPI): mpirun -np ${number_of_processors} ${path_of_program}/ensalign ${YYYYMMDDHH} ${fhr} ${input_file_path}

4. Output file

4.1) command line to read the header of output file

• ncdump -h ${filename}

4.2) variables

• ensfcst(nmembers, y, x) : Each member's original preicipitation field from ensemble system
• ensshffcst(nshfpass, nmembers, y, x) : Each member's aligned precipitation field
• shiftvec_u(nshfpass, nmembers, y, x), shiftvec_v(nshfpass, nmembers, y, x) : Each member's shift vector fields
• ensmean(y, x) : Ensemble mean of precipitation
• enspm(y, x) : Ensemble PM mean of precipitation
• enslpm(y, x) : Ensemble LPM mean of precipitation
• ensshfmean(nshfpass, y, x) : Spatially Aligned Mean (SAM) of precipitation
• ensshflpm(nshfpass, y, x) : Spatially Aligned Mean with LPM method applied (SAM-LPM) of precipitation