Tabulated parameters in SSEMIS.F

Dear all,
I am trying to understand how to compute the parameters in SSEMIS.F from Gong function.
More in details, I would like to reproduce the values of the array listed below:

C Lognormal parameters fit to the open-ocean flux function
REAL, SAVE :: DGNJ( 136 ) ! geom mean diam of accum-mode [um]
REAL, SAVE :: DGNK( 136 ) ! geom mean diam of coarse-mode [um]
REAL, SAVE :: SIGJ( 136 ) ! geom std deviation of accum-mode flux
REAL, SAVE :: SIGK( 136 ) ! geom std deviation of coarse-mode flux
REAL, SAVE :: FNJ ( 136 ) ! magnitude of accum-mode flux [1/m2/s]
REAL, SAVE :: FNK ( 136 ) ! magnitude of coarse-mode flux [1/m2/s]

C Open-ocean flux numerically integrated over size range of interest
REAL, SAVE :: VFLUX( 136 ) ! total particle volume flux [m3/m2/s]

I read the comment that they are pre-computed using MATLAB for THETA = 8 but I could not find any further information. Do you know if either the MATLAB code is available or where I can find more specific information?
Thank you very much in advance.
Best Regards,
Mario

Hi Mario

At the beginning I need to say that I do not have the MATLAB code and I did not reproduce those parameter arrays successfully. But my experience may give you some information (which of course may be misleading).

The Gong function (and its modification by Gantt et al. (2015)) has two peaks, and CMAQ uses two lognormal distributions to fit it, one for fine mode particles (“J”) and one for coarse mode particles (“K”). Each lognormal distribution can be characterized by three parameters, that is, geometric mean (DGN), geometric standard deviation (SIG), and the amplitude (FN). Once the fitting is done, you get the three parameters. (Wind speed is 1 for the fitting.)

One fitting can only provide one value for each parameter. The 136 values in each array come from the RH-dependence of the Gong flux function, that is, there are many flux functions at different RH. Usually, only a flux function at RH=80% is provided, and you need to calculate the flux function at other RH. CMAQ calculates the functions from RH=45% to RH=99% based on the work of Zhang et al. (2005, 2006). From RH=45% to RH=90%, there is one flux function for each 1% RH, and from 90.1% to 99%, there is one flux function for each 0.1% RH. The total number of flux functions is therefore 90-45+1+(99-90.1)*10+1=136.

Because the Gong function seems not strictly a combination of two lognormal distributions, you may need to judge when the fitting is the best.

By the way, from the information you provide (THETA=8), it seems that you are using CMAQv5.2 or v5.3 where the SSA source function is not the Gong function but the modified version by Gantt et al. When I tried to compare these two source functions, I tried to calculate the parameters for the original Gong function myself, but finally I gave up and just copied those parameters from old version of CMAQ. But if you want to investigate the impact of THETA values or want to use other source functions, you have to calculate the parameters yourself.

Best Regards,
Zhizhua

References
Gantt, B., Kelly, J. T., and Bash, J. O.: Updating sea spray aerosol emissions in the Community Multiscale Air Quality (CMAQ) model version 5.0.2, Geoscientific Model Development, 8, 3733-3746, 10.5194/gmd-8-3733-2015, 2015.
Zhang, K. M., Knipping, E. M., Wexler, A. S., Bhave, P. V., and Tonnesen, G. S.: Size distribution of sea-salt emissions as a function of relative humidity, Atmospheric Environment, 39, 3373-3379, 10.1016/j.atmosenv.2005.02.032, 2005.
Zhang, K. M., Knipping, E. M., Wexler, A. S., Bhave, P. V., and Tonnesen, G. S.: Reply to comment on “Size distribution of sea-salt emissions as a function of relative humidity”, Atmospheric Environment, 40, 591-592, 10.1016/j.atmosenv.2005.08.044, 2006.

1 Like

Hi Zhizhua,
sorry for my late reply.
Thank you very much for your kind explanations and the references.
I will check the papers.
Best regards,
Mario