Using independently/externally calculated BVOC emissions in EMEP

[PaulHamer]

Hi, I am working with a visiting scientist at NILU called Heidi Trimmel on a new dataset of BVOC emissions in an around the city of Vienna. We are nearing the point of completion for estimating the BVOC emissions over our domain of interest and we are looking for ways to try to use the emissions in a EMEP. I understand from the EMEP code and the technical descriptions of the model in scientific papers that EMEP uses its own internal algorithms to calculate the BVOC emissions following methods similar to MEGAN. Is it possible to ingest our own BVOCs, if given in a suitable format, into EMEP. It is worth noting that we have calculated the BVOC emissions on the exact same spatial grid as our WRF meteorological input.

Best regards Paul

[mifads]

Hi Paul, Are you using the WRF model to drive EMEP? If so, then it should be easy enough to mask-out the default EMEP emissions, and add your emissions from an external netcdf file (using Peter's new system). The masking-out of EMEP's emissions would need to be hard-coded though; we don't have anything to remove biogenic emissions for a specific area. Also, by default, we have emissions of isoprene and monoterpens . the latter all treated as a-pinene. We have a version with a-pinene, b-pinene (fake, in that the products are the same as a-pinene) and "other" MT (similar fake system( if would prefer that.

By the way, we don't use MEGAN. We use own-developed emission factors + temperature and light functions from the older Guenther papers (1993, 1995). I'm not yet convinced that the complexities of MEGAN2 can be justified for Europe, and the emissions themselves are very different to ours. /Dave

[mifads]

P.S: Ideally we could change the model to allow such nesting for BVOC, but that will have to be a future job!

[PaulHamer]

Hi Dave,

Thank you very much for your swift response. Yes, we are planning to use the WRF model to drive EMEP. In fact, our student Heidi Trimmel had another question answered on this swiftly by Peter two days ago :)

The emissions will be in a netcdf file and we can do whatever adjustments are needed to get to the suitable format. We are working with EMEP version rv4.17. Please can you guide us to which lines of code we need to mask out with hard coding for the BVOC emissions?

The next step we were wondering if you could help with is you have any code or documentation on how the lumping of the BVOCs is or could be done. We are working with MEGAN 3.1, and unfortunately the EMEP chemical mechanism is supported with the default code release. So our plan was to either try to do this ourselves or maybe to follow the approach that you are using.

Thank you for clarifying on the point about whether MEGAN is being used. We had some debate about this internally after Tove found some references to MEGAN in the runfile, yet your paper from 2015 describes what you've just explained.

Thanks Paul

[mifads]

Hi Paul, There are many ways to do this of course, but I think it can be simplest to split the problem into (1) make and use a mask, and (2) read the emissions. I will leave the user-guide (look for e.g. Emis_sourceFiles) and @mifapw to guide you on emissions, but Peter's new methods are very flexible and will let you read your BVOC easily I think.

About the mask, I suggest you simply make a file with the sum of your BVOC emissions in the same projection as your WRF meteo. Or emissions factors, or anything to show where you have your own calculations. Then, in Biogenics_mod.f90, you can use ReadField_CDF to read that mask, let's say into a logical 2-D array vienna_mask. I think you can then put at the end of the routine MergedBVOC something like:

where ( vienna_mask ) bvocEF = 0.0

This will overwrite the contents of eg bvocEF(i,j,iL,BIO_ISOP) etc with zeros, and leave you free to add your own emissions via the Peter's emissions routines. (Please check all steps!)

I can reply about MEGAN3.1 species later, but we are today (hopefully!) releasing rv4.35, which (together with our GenChem system) will be a bit more flexible with regard to the number of BVOC you can use.

[avaldebe]

Hi @PaulHamer

Did we answer your question? If so, please close this issue. Otherwise let us know what else can we do to help you.

Cheers, Álvaro.

[mifads]

Seems to be settled. Closing.