Developing FINN climatology for forecasting

[Abhinav Sharma]

Hello all,

I am trying to automate WRF-Chem for forecasting atmospheric chemistry for next two-three days. I wish to provide biomass  burning inputs to the model, for this I have planned to generate a climatology of FINNv2.5 MOD + VIIRS (MOZART chem) from 5 years (2016,17,18,19,21 skipping 2020). I have downloaded the global netcdf files from RDA (https://rda.ucar.edu/datasets/ds312-9/). 

However, FINN requires burned area for different veg. types which is not available in these files. If I understand it correctly, such information is needed by plumerise module. So if I just compute emission climatology and interpolate for my domain, would it be fine (if switching off plumerise)?

Is there any other option available. Thanks.

[Gabriele Pfister]

Hi Abhinav

the FINN emissions have all the inputs you need for running WRF-Chem with online plumerise (see https://www.acom.ucar.edu/Data/fire/data/finn2/README_FINNv2.5_Feb2022.pdf) . I assume you are using our fire_emiss tool to generate the wrffirechemi* files? 

You might have a specific reason to run the forecasts with climatological fire emissions but we do also create FINNv2.5 emissions in NRT. These could be used assuming persistence in fires during the forecast period (https://www2.acom.ucar.edu/modeling/finn-fire-inventory-ncar). 

Gabi

--

================================
Gabriele Pfister

Atmospheric Chemistry Observations & Modeling (ACOM) Lab

NSF National Center for Atmospheric Research (NCAR)

Email: pfi...@ucar.edu

Work Phone: +1 303 497 2915

Web: https://staff.ucar.edu/users/pfister

[Abhinav Sharma]

Dear Dr. Pfister,

Thanks for your suggestion. I have planned to download the FINNv2.5 0.25 degree gridded data to make the average of 5 years for each day. I would then interpolate them to my domain grid and write the wrffirechemi files using python just as fire_emiss tool would produce. The diurnal factors from WRAP will also be applied. I was thinking of doing the same averaging and regridding of FireCCI burned area (for FIRESIZE in wrffirechemi, description: https://climate.esa.int/en/projects/fire/). 

Regarding persistence burning, if I understand it correctly, you would assume a fire detected at day 0 to continue on next day with same or different intensity? I would like to know how this could be implemented with fire_emiss. If I download the text file for day 0, should I copy it with dates changed and the fire_emiss tool will take care of it?