(RAMA2MET Tool v2.0)
[Model Evaluation] [ascii2nc] WRF-chem RAMA
Strategies for air quality improvement requieres the use of air quality numerical models. The statistical tools that can provide objective comparison between models results and air quality measurements are important. One of those verification packages is the Model Evaluation Tools (MET) with it is possible to compare results from air quality models such WRF-chem [Grell et al 2005]1 and observations from air quality networks such as RAMA in Mexico City.
However the format requiered for the MET suite is not the format provided by the air quality network database. This code provide a way to reformat RAMA data base into MET ascii format in order to be used in ascii2nc tool for providing the data to pointstat for computing the different statistical metrics.
This code reads, computes wind vectors, identifies the grib code and reformat the air quality data into METv5 ascii format. This software was developed for an specific input ascii format however fuctions and subroutines can be used to convert other databases.
The MET ascii format has 11 columns it contains one point observation per line. Each row in the file has the following data in columns:
- Message_Type
- Station_ID
- Valid_Time in YYYYMMDD_HHMMSS format
- Lat in degrees North
- Lon in degrees East
- Elevation in meters above sea level
- Grib_Code as the integer GRIB code value or variable name corresponding to this observation type
- Level as the pressure level in hPa or accumulation interval in hours
- Height in meters above sea level or above ground level
- QC_String corresponding to the quality control value
- Observation_Value in the units prescribed for the grib code
Meteorological data and chemical trace gases are measured at ground level then the Message_Type variable is set to ADPSFC.
Information about stations are obtained from reading the file est_rama.txt , it has the Station_ID, Lat, Lon and Elevation
For Mexico City the pressure level is on average 776. hPa
Stations measured it variables at 10 m above ground level.
RAMA datafile in the first 10 rows contains metadata, the 11th has the header data and it has a single data per row.
11 date,id_station,id_parameter,value,unit
12 01-01-2020 01:00,ACO,RH,70,6
13 01-01-2020 01:00,ACO,TMP,11.8,5
14 01-01-2020 01:00,ACO,WDR,340,4
15 01-01-2020 01:00,ACO,WSP,0.5,3
Because a file contains a year measurements it can be quite large (>1.5M rows) and in order to reformatting a fortran code was developed.
The GRIB site contains a set of Parameter tables versions with the codes for each variable in order to map the variable to its description, units and abbreviation. In addition to the meteorological variables the chemical trace gases are also considered. The following tables used and variables considered are presented. The grib codes follows the guide from World Meteorological Organization WMO.
| Value | Parameter | Units | Abbrev. |
|---|---|---|---|
| 001 | Pressure | Pa | PRES |
| 011 | Temperature | K | TEMP |
| 031 | Wind direction (from which blowing) | deg true | WDIR |
| 032 | Wind speed | m/s | WIND |
| 033 | u-component of wind | m/s | UGRD |
| 034 | v-component of wind | m/s | VGRD |
| 052 | Relative humidity | % | RH |
| Value | Parameter | Units | Abbrev. |
|---|---|---|---|
| 156 | Particulate matter (coarse) | ug/m^3 | PMTC |
| 157 | Particulate matter (fine) | ug/m^3 | PMTF |
| 180 | Ozone concentration | ppbv | OZCON |
| Value | Parameter | Units | Abbrev. |
|---|---|---|---|
| 141 | Nitrogen Oxide | ppbv | NO |
| 142 | Nitrogen Dioxide | ppbv | NO2 |
| 148 | Carbon Monoxide | ppbv | CO |
| 232 | Sulfur Dioxide | ppbv | SO2 |
| 249 | Hydrophobic Organic Carbon | ug/m^3 | OC |
This mapping between text variable and grib code is performed in vconvert subroutine.
The RAMA database contains wind speed and wind direction, the meteorological model provides wind components in W-E and S-N directions. A conversion from vector to it components for wind use the following equations:
u = -S SIN(DD)
v = -S COS(DD)
where
- u - eastward wind component (m/s)
- v - northward wind component (m/s)
- S - wind speed (m/s)
- DD -- Wind direction (from which blowing) deg
This computations are performed in viento subroutine
For RAMA files the date and time are in one column separated by one space, the date in DD-MM-YYYY and time HH:MM subroutine fconvert reformat the date to YYYYMMDD_HHMMSS and computes the conversion from GMT-6 time zone to GMT.
Because the air quality network in Mexico City publish data after a QA/QC process the QC flags is set to 1.
The rama2met.exe program requires as input the following files:
namelist.met
est_rama.txt
contaminantes_YYYY.csv
meteorologia_YYYY.csv
namelist.met file contains the time period information
&FECHA
anio=2020 ! Year from input data YYYY
imes=02 ! Start month
fmes=03 ! End month
idia=18 ! Start day
fdia=11 ! End day
ihr=01 ! Start hour
fhr=24 ! End hour
/
est_rama.txt contains information from the stations: a three character ID, latitud, longitud, height above sea level and description.
Alias Latitud Longitud Altitud Estacion
ACO 19.635501 -98.912003 2198 Acolman
AJU 19.154286 -99.162611 2942 Ajusco
HGM 19.411617 -99.152207 2234 Hospital General de México
meteorologia_YYYY.csv contains the air pollutants concentrations during the year YYYY (i.e. 2020). RAMA format
contaminantes_YYYY.csv contains the meteorological variables during the year YYYY (i.e. 2020). RAMA format.
The program starts reading the namelist.met file, then the stations file est_rama.txt. Then reads the meteorolgia_YYYY.csv file, reformat the date, from the ID obtains the coordinates, and from the variable name the grib code. Finally writes the information in ramaYYYY_met.txt. After finishing the meteorological file starts with the pollutants file contaminantes_YYYY.csv following the same procedure and writing the reformat data in ramaYYYY_pol.txt.
The RAMA2.sh bash script can generate files per month.
Footnotes
-
Grell, G.A., Peckham, S.E., Schmitz, R., McKeen, S.A., Frost, G.J., Skamarock, W.C., & Eder, B.K. (2005). Fully coupled "online" chemistry within the WRF model. Atmospheric Environment, 39, 6957-6975. ↩