I want to use the boundary conditions produced by the hemispherical model of CMAQ in the CMAQ simulation, because currently I am running the CMAQ model using clean profiles.
However, I would like to consider the effects of long-range transport during the simulation process.
I am not sure if the current data is available up to December 2024, or if it has been updated to the latest 2025 version. Could you please advise me on how to proceed?
According to the tutorial, one method would be to create a script to time shift from the available data to the dates that you are interested in modeling.
Thank you very much for your reply, lizadams.
So, if I want to simulate the CMAQ from January 1, 2025, to February 20, 2025, and use the boundary condition from H-CMAQ, I just need to use the latest daily average H-CMAQ output from 2019 and adjust the time to the period I want to simulate?
Hi lizadams:
I still have a little question. I still don’t quite understand why we can take the daily average H-CMAQ data from 2019, change the time, and use it to create the 2025 BCON results. Won’t the error be significant?
You are correct that simply time-shifting day-specific files to a different year would introduce errors. The section of the tutorial discussing time-shifting as a potential approach dealt with using both seasonal average and day-specific files. Using time shifting on seasonal average files might introduce fewer errors since those files would be expected to vary less across years than day-specific files, but of course interannual variability and trends in emissions can affect seasonal averages, too. You could construct seasonal average files for 2019 from the day-specific files, or you could work with the existing seasonal average files for 2016 that were based on an earlier model version.
The bottom line is that the EQUATES project does not extend beyond 2019 so there currently are no hemispheric CMAQ outputs from that project past that year. If the errors and uncertainties associated with time-shifting day-specific or seasonal average H-CMAQ output to a later year are not acceptable for your specific project, you would either have to find a way to perform your own H-CMAQ simulations for your time period of interest, or possibly obtain outputs from other existing large-scale modeling datasets and then process them into boundary conditions for your regional CMAQ domain.
As Christian says, the error certainly could be significant. You have several alternatives. You could generate boundary conditions from a forecast model that represents 2025 meteorology (e.g., RAQMS, GEOS-CF, or WACMM). The aqmbc processor will help you. aqmbc is under rapid development, so open an issue on the github site if something isn’t working for you.
To add to Barron’s suggestion of looking for other potential sources of large-scale model output to provide boundary conditions for your regional CMAQ domain for your period of interest, I’d also encourage you to carefully examine the output from those models as part of your process to generate boundary conditions for CMAQ and become aware of how these models were configured.
If your goal is to consider the effects of long-range transport as you stated in your first post, you would want to understand how these models represented various “upstream” emission sources, to which extent they might have constrained emissions with remote sensing products, how they handled anthropogenic emissions for very recent time periods when bottom-up inventories typically aren’t yet available, etc. The use of actual period-specific meteorology and maybe fire and dust emissions in those models would certainly be a strength, but you would still want to learn as much as you can about those models, their inputs, and their gas phase and aerosol schemes and how to most appropriately link them to the regional CMAQ gas and aerosol scheme (Barron’s aqmbc tool is a great starting point for that).
If you’re studying long-range transport of air pollution, don’t treat this aspect as a black box but really think through your different options and objectives, just as you have thought through that time-shifting existing H-CMAQ output may not be appropriate for your study.