Process Analysis - Physical Tendencies Instead of Concentrations


As far as I know, the CMAQ process analysis explains the tendency of concentration with respect to each process (VDIF, CLD and etc.). These tendencies do not only rely on the rate of each process but also on underlying concentrations. For example, if two different emissions are used with the same meteorological fields, we will see a difference in VDIF resulting from two different values for that specific compound. This will obscure the realistic portrayal of each subprocess before considering the level of compounds. I should note that I’m using CMAQ v5.2 right now. Does the new version (or will the future version) include the magnitude of physical/chemical processes instead of fluxes?


The process analysis fields are in units of mixing ratio (for gases) or concentration (for aerosols) change per output time step. There are no plans to output the concentration normalized physical/chemical timescale, for example, but this could be approximated by dividing by the average value of the variable from ACONC. In some cases, diagnostic values for individual processes do exist. For example, deposition velocities may be output in the DEPV file. For other processes, like VDIF, I’m not sure what values one would be looking for. We are happy to consider proposed metrics for additional optional diagnostic output.

Best wishes,

1 Like

Thanks, Ben, It makes sense. I guess, to have a better picture of how fast vertical diffusion is I can calculate Richardson’s number, or even looking at PBLH may drop some hints on diffusion. What do you think?

Those are good suggestions. I know other models make use of the vertical diffusion coefficients (Kzz). Could you diagnose those from your WRF inputs? My expectation is that they should be basically the same although not numerically identical to what CMAQ would output if we tried to diagnose them online.

Perhaps I should hand this off to @pleim.jon or @tlspero as I’m now out of my expertise.

I ended up scaling the process analysis output by concentration. By doing that, I was able to isolate the magnitude of each process without involving the underlying concentration.