FINN2.5 in EMEP rv4_36

[kevinbasoa]

Dear EMEP team,

I'm running EMEP over South America using FINN2.5 to cover the forest fire emissions. The overall results are ok, however, in places were fires are present, PM concentrations and AOD are too high. I check if our antrophogenic emission invenroy was the problem but without the forest fire emission the results are ok.

My question is, are you using FINN2.5 in other simulations? how you process the emissions inventory from the database (https://www2.acom.ucar.edu/modeling/finn-fire-inventory-ncar)?

Since the emissions from the netCDF comes in molecules/cm2/s, I have some troubles to put it on the right units (molecules/day). I also check the "CMX_BiomassBurning_FINNv1p5.txt", and it is ok too (I named the variables on the inventory).

It is possible to share how you preprocess the forest fire emissions inventory from the FINN database?

In any case, thank you very much.

Best regards,

Kevin

[gitpeterwind]

Hi Kevin,
We do not normally distribute our FINN forest fire netcdf files. If you have got a file that has been converted by us, which emission year was it, and what is the creation date? (global attributes created_date and created_hour). We have had some erroneous versions, and maybe your copy is one of them.

[kevinbasoa]

Hi Peter,

I used a inventory that a made myself, with the database from NCAR (https://www2.acom.ucar.edu/modeling/finn-fire-inventory-ncar), I don't have access to any "official" forest fires inventory. Here is a dump from the inventory that I made.

I think that emissions from aerosol in molecules are or could be wrong calculated. This is a inventory for last days of 2018 and January of 2019.

dimensions:
lat = 880 ;
lon = 880 ;
time = 43 ;
variables:
float lat(lat) ;
lat:units = "degrees_north" ;
lat:long_name = "latitude" ;
float lon(lon) ;
lon:units = "degrees_east" ;
lon:long_name = "longitude" ;
float time(time) ;
time:units = "days since 1990-01-01 00:00:00" ;
time:long_name = "time" ;
time:calendar = "Gregorian" ;
float BC(time, lat, lon) ;
BC:units = "mol/day" ;
float CO(time, lat, lon) ;
CO:units = "mol/day" ;
float ACET(time, lat, lon) ;
ACET:units = "mol/day" ;
float ALK4(time, lat, lon) ;
ALK4:units = "mol/day" ;
float BENZ(time, lat, lon) ;
BENZ:units = "mol/day" ;
float C2H4(time, lat, lon) ;
C2H4:units = "mol/day" ;
float C2H6(time, lat, lon) ;
C2H6:units = "mol/day" ;
float C3H8(time, lat, lon) ;
C3H8:units = "mol/day" ;
float CH20(time, lat, lon) ;
CH20:units = "mol/day" ;
float GLYC(time, lat, lon) ;
GLYC:units = "mol/day" ;
float HAC(time, lat, lon) ;
HAC:units = "mol/day" ;
float MEK(time, lat, lon) ;
MEK:units = "mol/day" ;
float MGLY(time, lat, lon) ;
MGLY:units = "mol/day" ;
float PRPE(time, lat, lon) ;
PRPE:units = "mol/day" ;
float TOLU(time, lat, lon) ;
TOLU:units = "mol/day" ;
float XYLE(time, lat, lon) ;
XYLE:units = "mol/day" ;
float CH3CHO(time, lat, lon) ;
CH3CHO:units = "mol/day" ;
float NH3(time, lat, lon) ;
NH3:units = "mol/day" ;
float NO(time, lat, lon) ;
NO:units = "mol/day" ;
float NO2(time, lat, lon) ;
NO2:units = "mol/day" ;
float OC(time, lat, lon) ;
OC:units = "mol/day" ;
float PM25(time, lat, lon) ;
PM25:units = "mol/day" ;
float SO2(time, lat, lon) ;
SO2:units = "mol/day" ;

// global attributes:
:title = "FINN_v2.5_2018" ;

Since this inventory I made it myself, I think is builded wrong.

Could you guide me if there something wrong with this inventory?

Kevin

[gitpeterwind]

and you use the correct unit conversion? they specify
"NOTE on units: In the text files with emissions for each fire, the aerosols (BC, OC, PM2.5, PM10), as well as the gas-phase lumped species 'NOXasNO' and NMOC, are in units of kg/day, and the gas-phase species are in moles/day. In these files 'AREA' is the area burned in m^2; 'BMASS' is the biomass burned per area burned in kg/m^2. 'GENVEG' specifies the vegetation type assumed for the fire: 1 = grasslands and savanna; 2 = woody savanna/shrublands; 3 = tropical forest; 4 = temperate forest; 5 = boreal forest; 6 = temperate evergreen forest; 9 = croplands; 0 = no vegetation.
In the 0.1x0.1 degree gridded files, all compounds are converted to molecules/cm^2/sec, where aerosols have been converted from kg to molecules using molecular weight of 12 g/mole.
"

[kevinbasoa]

I read it and I think the second apply to my case, I only use the netCDF files, not the text files. I convert these to molescules/day, so I did not change molecules to mass. I do not know if EMEP convert "molecules of PM" as they do (NCAR).

Maybe this is making my results bad?

[gitpeterwind]

Yes, the emep model assumes kg/day for PM25, given you are using the "CMX_BiomassBurning_FINNv2p5.txt" conversion file

[kevinbasoa]

Thanks for the clarification.

I will change the units and come back with you.

Best regards