Table of Contents

Step 1: Getting meteorological data from IFS or ERA5 reanalysis

Boundary conditions for WRF can be freely downloaded from ftp://ftp.ecmwf.int/pub/wrf. The output comes from ECMWF's operational HRES forecast and only one day's data is available. For more information please have a look at: ECMWF WRF Test Data. For those who have access to ECMWF MARS (access restricted) they can find the recipe how to retrieve the data for the past.

Alternatively, users can download the freely available ERA5 global reanalysis data from the C3S Climate Data Store (CDS) to drive WRF meteorological component. Please read through How to download ERA5. You also need to have a look at the parameter table in ECMWF WRF Test Data for retrieving relevant data.

Surface data 
import cdsapi

c = cdsapi.Client()

c.retrieve(
    'reanalysis-era5-single-levels',
    {
        'product_type': 'reanalysis',
        'format': 'grib',
        'grid': '0.75/0.75',
        'variable': [
            '10m_u_component_of_wind', '10m_v_component_of_wind', '2m_dewpoint_temperature',
            '2m_temperature', 'land_sea_mask', 'mean_sea_level_pressure',
            'sea_ice_cover', 'sea_surface_temperature', 'skin_temperature',
            'snow_density', 'snow_depth', 'soil_temperature_level_1',
            'soil_temperature_level_2', 'soil_temperature_level_3', 'soil_temperature_level_4',
            'surface_pressure', 'volumetric_soil_water_layer_1', 'volumetric_soil_water_layer_2',
            'volumetric_soil_water_layer_3', 'volumetric_soil_water_layer_4',
        ],
        'year': '2010',
        'month': '01',
        'day': [
            '01', '02',
        ],
        'time': [
            '00:00', '03:00', '06:00',
            '09:00', '12:00', '15:00',
            '18:00', '21:00',
        ],
    },
    'era_sfc_20100101-20100102.grib')
Pressure level data 
import cdsapi

c = cdsapi.Client()

c.retrieve(
    'reanalysis-era5-pressure-levels',
    {
        'product_type': 'reanalysis',
        'format': 'grib',
        'grid': '0.75/0.75',
        'variable': [
            'geopotential', 'relative_humidity', 'temperature',
            'u_component_of_wind', 'v_component_of_wind',
        ],
        'pressure_level': [
            '100', '250', '500',
            '850', '1000',
        ],
        'year': '2010',
        'month': '01',
        'day': [
            '01', '02',
        ],
        'time': [
            '00:00', '03:00', '06:00',
            '09:00', '12:00', '15:00',
            '18:00', '21:00',
        ],
    },
    'era5_pl_20100101-20100102.grib')


If your version of the WRF model requires separate input GRIB files for each time step, you can easily split it using ecCodes:

ecCodes
$>grib_copy era_sfc_20100101-20100102.grib era_sfc_[validityDate][validityTime].grib
$>grib_copy era_pl_20100101-20100102.grib era_pl_[validityDate][validityTime].grib


Step 2: Downloading CAMS data from MARS

2.1 Retrieving the data from MARS

We tested with the CAMS reanalysis data for driving the WRF-Chem model. Please have a look at how to download the CAMS Reanalysis data so that you have a good understanding of the scripts below. Scripts below are using ECMWF's Web API.

Script for retrieving aerosols, parameter names are given in the table bellow:

ParamShort NameLong Name

tTemperature
1.210

aermr01

Sea Salt Aerosol (0.03 - 0.5 um) Mixing Ratio

2.210

aermr02

Sea Salt Aerosol (0.5 - 5 um) Mixing Ratio

3.210

aermr03

Sea Salt Aerosol (5 - 20 um) Mixing Ratio

4.210

aermr04

Dust Aerosol (0.03 - 0.55 um) Mixing Ratio

5.210

aermr05

Dust Aerosol (0.55 - 0.9 um) Mixing Ratio

6.210

aermr06

Dust Aerosol (0.9 - 20 um) Mixing Ratio

7.210

aermr07

Hydrophobic Organic Matter Aerosol Mixing Ratio

8.210

aermr08

Hydrophilic Organic Matter Aerosol Mixing Ratio

9.210

aermr09

Hydrophobic Black Carbon Aerosol Mixing Ratio

10.210

aermr10

Hydrophilic Black Carbon Aerosol Mixing Ratio

11.210

aermr11

Sulphate Aerosol Mixing Ratio

AER 
#!/usr/bin/env python
from ecmwfapi import ECMWFDataServer
   
server = ECMWFDataServer()
   
server.retrieve({
	'dataset'   : "cams_reanalysis",
    'class'     : "mc",
	'type'      : "an",
	'stream'    : "oper",
	'expver'    : "eac4",
    'levtype'   : "ml",
    'repres'	: "gg",
	'levellist'	: "7/to/60",
    'param'     : "T/1.210/2.210/3.210/4.210/5.210/6.210/7.210/8.210/9.210/10.210/11.210",
	'date'      : "20100101/to/20100102",
    'step'      : "0",
    'time'      : "0/to/21/by/3",
	'grid'		: "0.75/0.75",
    'target'    : "AER_20100101_20100102_eac4.grib"
})


Script for retrieving global reactive gases, parameter names are given in the table bellow:

ParamShort NameLong Name
203.210

O3

ozone

123.210

CO

carbonmonoxide

27.217

NO

nitrogen_monoxide

121.210

NO2

nitrogen_dioxide

13.217

PAN

peroxyacetyl_nitrate

6.217

HNO3

nitric_acid

124.210

CH2O

formaldehyde

122.210

SO2

sulfur_dioxide

GRG 
#!/usr/bin/env python
from ecmwfapi import ECMWFDataServer
   
server = ECMWFDataServer()
   
server.retrieve({
	'dataset'   : "cams_reanalysis",
    'class'     : "mc",
	'type'      : "an",
	'stream'    : "oper",
	'expver'    : "eac4",
    'levtype'   : "ml",
    'repres'	: "gg",
	'levellist'	: "7/to/60",
    'param'     : "203.210/123.210/27.217/121.210/13.217/6.217/124.210/122.210",
	'date'      : "20100101/to/20100102",
    'step'      : "0",
    'time'      : "0/to/21/by/3",
	'grid'		: "0.75/0.75",
    'target'    : "GRG_20100101_20100102_eac4.grib"
})


Script for retrieving volatile organic compounds, parameter names are given in the table bellow:

paramShort NameLong Name
52.217

CH3COCH3

acetone

45.217

C2H6

ethane

9.217

PAR

paraffins

42.217

CH3OH

methanol

47.217

C3H8

propane

46.217

C2H5OH

ethanol

10.217

C2H4

ethene

12.217

ALD2

aldehydes

11.217

OLE

olefins

16.217

C5H8

isoprene

43.217

HCOOH

formic_acid

7.217

CH3OOH

methylperoxide

15.217

ONIT

organic_nitrates

GRG_voc 
#!/usr/bin/env python
from ecmwfapi import ECMWFDataServer
   
server = ECMWFDataServer()
   
server.retrieve({
	'dataset'   : "cams_reanalysis",
    'class'     : "mc",
	'type'      : "an",
	'stream'    : "oper",
	'expver'    : "eac4",
    'levtype'   : "ml",
    'repres'	: "gg",
	'levellist'	: "7/to/60",
    'param'     : "52.217/45.217/9.217/42.217/47.217/46.217/10.217/12.217/11.217/16.217/43.217/7.217/15.217",
	'date'      : "20100101/to/20100102",
    'step'      : "0",
    'time'      : "0/to/21/by/3",
	'grid'		: "0.75/0.75",
    'target'    : "GRG_voc_20100101_20100102_eac4.grib"
})


Script for retrieving h2o2:

GRG_h2o2 
#!/usr/bin/env python
from ecmwfapi import ECMWFDataServer
   
server = ECMWFDataServer()
   
server.retrieve({
	'dataset'   : "cams_reanalysis",
    'class'     : "mc",
	'type'      : "an",
	'stream'    : "oper",
	'expver'    : "eac4",
    'levtype'   : "ml",
    'repres'	: "gg",
	'levellist'	: "7/to/60",
    'param'     : "3.217",
	'date'      : "20100101/to/20100102",
    'step'      : "0",
    'time'      : "0/to/21/by/3",
	'grid'		: "0.75/0.75",
    'target'    : "GRG_h2o2_20100101_20100102_eac4.grib"
})

Script for retrieving surface pressure:

PS 
#!/usr/bin/env python
from ecmwfapi import ECMWFDataServer
   
server = ECMWFDataServer()
   
server.retrieve({
	'dataset'   : "cams_reanalysis",
    'class'     : "mc",
	'type'      : "an",
	'stream'    : "oper",
	'expver'    : "eac4",
    'levtype'   : "ml",
    'repres'	: "gg",
	'levellist'	: "1",
    'param'     : "152",
	'date'      : "20100101/to/20100102",
    'step'      : "0",
    'time'      : "0/to/21/by/3",
	'grid'		: "0.75/0.75",
    'target'    : "PS_20100101_20100102_eac4.grib"
})

2.2 Converting GRIB to NetCDF

Before running the script below you will need to have these packages installed:

You also need to download the get_tablecol script and table_tm5ver15htap.txt file into the same directory where script is to be run. The next step is to run the script below:


cdo 
#!/bin/bash

module load netcdf4
module load cdo
module load nco

fd="20100101_20100102_eac4"

box_lab1='EU'
box_s=15.0
box_n=78.0
box_e=81.0
box_w=303.0
box1="${box_w},${box_e},${box_n},${box_s}"

chnames="go3,O3,so2,SO2,no2,NO2,co,CO,hcho,CH2O,no,NO,pan,PAN,hno3,HNO3,ald2,ALD2,c2h4,C2H4,c2h5oh,C2H5OH,c2h6,C2H6,c3h8,C3H8,c5h8,C5H8,ch3coch3,CH3COCH3,ch3oh,CH3OH,ch3ooh,CH3OOH,hcooh,HCOOH,par,PAR,ole,OLE,onit,ONIT,h2o2,H2O2"
chlist='O3/CO/NO/NO2/PAN/HNO3/CH2O/SO2'

tablefile=table_tm5ver15htap.txt

grouplist=("AER" "GRG" "GRG_voc" "PS")
for group in ${grouplist[*]} ; do

cdo -f nc -b 32 -invertlat ${group}_${fd}.grib ${group}_${fd}.nc
#
if [[ $group = 'PS' ]] ; then
  # calculate surface pressure
    cdo -expr,'ps=exp(lnsp)' PS_${fd}.nc PS1_${fd}.nc
    mv  PS1_${fd}.nc PS_${fd}.nc
    ncatted -O -h -a long_name,ps,c,c,"surface pressure" PS_${fd}.nc dum
    mv dum PS_${fd}.nc
    ncatted -O -h -a unit,ps,c,c,"Pa" PS_${fd}.nc dum
    mv dum PS_${fd}.nc
fi
#
cdo -sellonlatbox,$box1 ${group}_${fd}.nc ${group}_${fd}_${box_lab1}.nc
    if [[ $group = 'GRG' || $group = 'GRG_voc' ]] ; then
        fll=${group}_${fd}_${box_lab1}.nc
        cdo -chname,$chnames $fll dum
        mv dum  $fll
      # put together list of grib codes
        for species in `echo $chlist | perl -pe 's{/}{ }g'`
        {
        grib_no=`get_tablecol -w name=$species $tablefile grib`
        longname=`get_tablecol -w name=$species $tablefile long_name`
        molmass=`get_tablecol -w name=$species $tablefile molm`
        cunit="kg kg**-1"
        ncatted  -O -h -a units,${species},o,c,"$cunit"   $fll
        ncatted  -O -h -a long_name,${species},o,c,"$longname"  $fll
        ncatted  -O -h -a molar_mass,${species},a,c,"$molmass"  $fll
        }
    fi
#
    nccopy -d1 ${group}_${fd}_${box_lab1}.nc ${group}_${fd}_${box_lab1}.nc.tmp
    mv ${group}_${fd}_${box_lab1}.nc.tmp ${group}_${fd}_${box_lab1}.nc

done

If the above script runs successfully, you should have the following files created:

AER_20100101_20100102_eac4_EU.nc
GRG_20100101_20100102_eac4_EU.nc
GRG_voc_20100101_20100102_eac4_EU.nc
PS_20100101_20100102_eac4_EU.nc

Step 3: Using mozbc to interpolate CAMS data for WRF

With a set of additional utilities, it is possible to use the mozbc package provided by the Atmospheric Chemistry Observations and Modelling Lab (ACOM) of NCAR to pre-process CAMS data with some minor modifications to the original code. The set of additional utilities has been developed as part of the AQMEII project and kindly provided by A. Lupascu (IASS-Posdam) and C. Knote (NCAR): public_box.tar.gz. The package contains:


Directory ./ECMWF2WRFChem/
MACC_BC2MOZART.ncl

ncl script to convert CAMS files containing surface pressure, global reactive gases, and aerosols into a single file ready to be used by modified MOZBC package.

convert.sh
Runs the ncl script. 
Directory ./
mo_mozart_lib.patch

Patch for mo_mozart_lib.f90 program (part of original mozbc package). The file should be copied to mozbc/ and then executed:

$>patch < mo_mozart_lib.patch

saprc99.inp

Namelist. Defines input and output directories and which variables to interpolate to WRF grid. It is possible to use only those variables already defined in wrfinput_d01, which depends on chemical package and options selected in &chem section of WRF's namelist.input. Here, an example for SAPRC99 chemical mechanism coupled with the MOSAIC aerosol chemistry module with 4 size bins.

The users have to provide their own mapping to get the chemical boundary condition for their preferred chemistry option.

mozart.inpA namelist example for mozart package.

Step 4: Running the model


The first step is to run the WRF model but only up to real.exe. Instruction on how run the model are provided in WRF official user guide. Note that you should not execute the wrf.exe because we still do not have a chemical BC interpolated to the WRF grid. This will be created using the utility provided by AQMEII and the mozart package. Before running this step you should check and edit paths in convert.sh and MACC_BC2MOZART.ncl if necessary.

$> cd ECMWF2WRFChem/
$> ./convert.sh

This step converts four NetCDF files (AER_*, GRG_*, GRG_voc_*, and PS_*) into a single AQMEII_*.nc file.

Once it is completed you can run the next step. Unpack, mozbc.tar, go to the mozbc main directory, copy mo_mozart_lib.patch, install, edit path to WRF run directory in the saprc99.inp namelist, and finally run mozbc.

$> tar -xvf mozbc.tar
$> cd mozbc/
$> cp public_box/mo_mozart_lib.patch .
$> patch < mo_mozart_lib.patch
$> make_mozbc
$> mozbc < saprc99.inp > mozbc.out

This program will read NetCDF files created in the previous step (AQMEII_*.nc), chemical variables from it will be interpolated on WRF grid and added to wrfinput_d01 and wrfbdy_d01 files. Once this step has completed successfully you are ready to execute wrf.exe.

This document has been produced in the context of the Copernicus Atmosphere Monitoring Service (CAMS).

The activities leading to these results have been contracted by the European Centre for Medium-Range Weather Forecasts, operator of CAMS on behalf of the European Union (Delegation Agreement signed on 11/11/2014 and Contribution Agreement signed on 22/07/2021). All information in this document is provided "as is" and no guarantee or warranty is given that the information is fit for any particular purpose.

The users thereof use the information at their sole risk and liability. For the avoidance of all doubt , the European Commission and the European Centre for Medium - Range Weather Forecasts have no liability in respect of this document, which is merely representing the author's view.

Feedback: C3S User Satisfaction Survey - CAMS User Satisfaction Survey

Web: C3S Help and Support - CAMS Help and Support