Henghengniceman / pyWRF-automation

Python automation script to download the Global Forecast System (GFS) data from NOMADS NOAA with spatial resolution 0.25 degree and execute Weather Research & Forecasting (WRF) model.

PyWRF-Automation

Python automation script to download the Global Forecast System (GFS) data from NOMADS NOAA with spatial resolution 0.25⁰ and execute Weather Research & Forecasting (WRF) model.

Prerequisites

To using this script, you must complete following prerequisites such as:

1. Linux/Unix distribution
2. Python 3.7+
3. MPI Package (OpenMPI/Intel MPI/MPICH)
4. WRF-ARW Model with em_real case using dmpar compiler selection.

This repository includes two script, which is main.py as an executable script and utils.py as a collection of function that will be used by main.py script.

How to use

1. Open main.py file and change the values below Arguments comment (line 11):

# Arguments
gfsout_path             = "/home/your_username/wrf_model/gfs_dataset"
wps_path                = "/home/your_username/wrf_model/wps"
wrf_path                = "/home/your_username/wrf_model/wrf/test/em_real"
wrfout_path             = "/home/your_username/wrf_model/wrf_output"
gfs_num_workers         = 4
gfs_download_increment  = 3
gfs_left_lon            = 110
gfs_right_lon           = 115
gfs_top_lat             = -2
gfs_bottom_lat          = -5
wrf_forecast_duration   = 1 
num_proc                = 1
wrfout_domain_data      = 1

Argument	Data Type	Range	Description	Note
gfsout_path	str	-	Path to GFS dataset folder	-
wps_path	str	-	Path to compiled WPS folder	-
wrf_path	str	-	Path to compiled WRF met_em folder	-
wrfout_path	str	-	Path to wrfout folder	-
gfs_num_workers	int	1 to nproc	Number of workers will be assigned to download gfs data concurrently	Higher value will make download faster, but it will cost CPU cores
gfs_download_increment	int	-	Which GFS forecast hours data will be downloaded	Set to 1 if you want to download GFS dataset for every forecast hours
gfs_left_lon	float	-180 to 180	Longitude	gfs_left_lon < gfs_right_lon
gfs_right_lon	float	-180 to 180	Longitude	-
gfs_top_lat	float	-90 to 90	Latitude	gfs_top_lat > gfs_bottom_lat
gfs_bottom_lat	float	-90 to 90	Latitude	-
wrf_forecast_duration	int	1 to 16 (in days)	Length of simulation days	-
num_proc	int	1 to nproc	Number of CPU cores will be used to execute real.exe & wrf.exe	-
wrfout_domain_data	int	1 to number_of_domain	Which wrfout file will be saved	Set to 1 if you want to save wrfout file on domain 1 etc

Note: in order to download GFS dataset for several hours (eg: every 3 hours), set gfs_download_increment argument to 3.

2. Open namelist.wps file then set start_date and end_date value to 2000-01-01_00:00:00.
3. Open namelist.input file then set the following variables using this value below:

Variable	value
run_days	0
start_year	2000
start_month	01
start_day	01
end_year	2000
end_month	01
end_day	01

4. Export libraries path LD_LIBRARY_PATH that will be used by wps.exe and wrf.exe.
5. Execute the program by typing python main.py

Credit

Copyright (c) 2020-present Reza Pahlevi and Agung Baruna Setiawan Noor.