HUSILEBAI / mpi2wrf

Climate Data Processing Scripts for MPI-ESM1-2-HR Model

This repository contains scripts for processing climate data from the MPI-ESM1-2-HR model. These scripts handle data interpolation, selection, and conversion to a common grid format suitable for further analysis.

Table of Contents

• Scripts
  • Bash Script
  • Python Script
• Dependencies
• Usage
  • Bash Script Usage
  • Python Script Usage
• Author
• Acknowledgement

Scripts

Bash Script

The Bash script processes the Sea Surface Temperature (SST) data from the MPI-ESM1-2-HR historical dataset by selecting specific time steps and interpolating the data to a target grid.

File: process_var_tos.sh

Script Details

1. Input Files:

   input_files=(
       "tos_3hr_MPI-ESM1-2-HR_historical_r1i1p1f1_gn_198001010300-198501010000.nc"
       "tos_3hr_MPI-ESM1-2-HR_historical_r1i1p1f1_gn_198501010300-199001010000.nc"
       "tos_3hr_MPI-ESM1-2-HR_historical_r1i1p1f1_gn_199001010300-199501010000.nc"
       "tos_3hr_MPI-ESM1-2-HR_historical_r1i1p1f1_gn_199501010300-200001010000.nc"
       "tos_3hr_MPI-ESM1-2-HR_historical_r1i1p1f1_gn_200001010300-200501010000.nc"
       "tos_3hr_MPI-ESM1-2-HR_historical_r1i1p1f1_gn_200501010300-201001010000.nc"
       "tos_3hr_MPI-ESM1-2-HR_historical_r1i1p1f1_gn_201001010300-201501010000.nc"
   )

2. Target Grid Description: The target grid is defined in target_grid.txt:

   gridtype = lonlat
   gridsize = 73728
   xsize = 384
   ysize = 192
   xfirst = -179.53125
   xinc = 0.9375
   yfirst = -89.4625
   yinc = 0.9375

3. Processing Loop: The script loops through each input file, adjusts the start time, selects specific time steps, performs bilinear interpolation, and saves the output file.

Python Script

The Python script processes climate data from the MPI-ESM1-2-HR model by reading multiple variables, performing boundary checks, and ensuring data integrity by handling missing values. The script then interpolates the data to a common grid and saves the results for further analysis.

File: process_mpi2wrf.py

Script Details

1. Constants and Directories:

   • Start and End Dates:

     START_DATE = DT.datetime(1980, 1, 1, 6, 0, 0)
     END_DATE = DT.datetime(1985, 1, 1, 0, 0, 0)

   • Directories:

     INPUT_DIR = '/path/to/input/'
     OUTPUT_DIR = '/path/to/output/'

2. Variable File Patterns: The script handles multiple variables such as tas, huss, ps, etc.

3. Utility Functions:

   • format_time(): Formats datetime objects.
   • read_and_process_variable(): Reads and processes 2D variables from NetCDF files.
   • read_and_process_3d_variable(): Reads and processes 3D variables from NetCDF files.
   • map_to_pywinter_name(): Maps variable names to PyWinter naming conventions.

4. Processing Functions:

   • process_chunks(): Processes data in chunks.
   • process_variables(): Main function to process variables from NetCDF files.

5. Main Execution Flow: The script defines the main execution flow and processes data in chunks:

   if __name__ == "__main__":
       main()

Dependencies

Ensure the following dependencies are installed before running the scripts:

• Bash
• CDO (for the Bash script)
• Python 3.x
• numpy
• pywinter
• netCDF4
• scipy

Usage

Bash Script Usage

1. Set Execute Permission:

   chmod +x process_var_tos.sh

2. Run the Script:

   ./process_var_tos.sh

Python Script Usage

1. Set Directories: Ensure the INPUT_DIR and OUTPUT_DIR paths are correctly set in the script.

2. Run the Script:

   python process_mpi2wrf.py

Author

Husile Bai
Email: [e-mail]

Acknowledgement

The author extends gratitude to Prof. Alfonso Fernandez ([e-mail]) for recommending the 'pywinter' package and providing codes for data processing.