[RegCNET] Query about internal mixing aerosols direct effect in RegCM3 (fwd)

[Solmon Fabien]

Hello

bixq wrote:
>
> Could someone please answer the question from Dr. Bingliang Zhuang ?
>
> ---------- Forwarded message ----------
> Date: Mon, 18 Oct 2010 23:02:56 +0800
> From: 炳亮庄 <dzhblyzh at gmail.com>
> To: bixq at ictp.trieste.it
> Subject: Query about internal mixing aerosols direct effect in RegCM3
>
> I am Zhuang Bingliang from School of Atmospheric Sciences, Nanjing
> University, Nanjing, China. I am major in the effects of aerosols on
> climate by using RegCM3.
>
> Now I want to investigate the direct effect of anthropogenic aerosols
> with
> different mixing states. However, I am confused about the treatment of
> the
> internal mixing of aerosols in RegCM3. As we know that there are two
> types
> of internal mixing states of aerosols (one is *well mixed*; the other
> one is
> * core-shell mixed*). Which type is used in RegCM3?
>
The default version is external mixing for aerosol. The option mixtype=2
consider a core shell mixing where the core can be dust or Black carbon.
The table have been precalculated using a mie code . But hypothesis on
the diameter of the particles have been made as I 'll explain here after.


> Additionally, I have a inquiry about the optical file “optbat.bin”
> downloaded from the ICTP Physics of Weather and Climate Group DATA
> Server,
> which is used when internal mixing aerosols are considered. It is a 6-D
> optical database (e.g.: *dextmix(4,19,11,11,11,11)*). “19” and “11” in
> the
> array represent wave bands and aerosols volume fractions, respectively.
> However, what does the first dimensional number (*4*) represent?
>

I guess 4 stands for a class of diameter , which should be the one of
dust particles :  The problem is that calculating a good internal mixing
with RegCM  simple aerosol module is nearly impossible . You don't know
the evolution of the diameter of the mixed particle so you need to
assume it for mie calculation. Since we were interested by the mixing of
dust and other aerosol for this experiment over Africa , we assumed that
we could have 4 class of diameter equal to the dust diameter ( 4
possible bins) . I checked the code  and it turns out that in reality,
we even use just the 1 diameter of this table in the calculation of
optical properties ( and it is hard coded in aeroppt.F ... )

My advice is :

It is great that you want to study the mixing effect !

But be cautious with this internal table , it need more testing you
might want to revisit this parameterization yourself. For example you
could  redifine simply the  optical properties of a  mix aerosol type as
constrained by observation .. it would not be less accurate I think ...

In the future , we 'll have a full gas-aerosol dynamic ( not before one
year) , so the mixing will be better parameterized

Also  I advise you to switch to RegCM4,  and to be critical with regards
to the default optical properties of the code.. you might have more
accurate data for your specific case.


Best regards

Fabien

> Looking forward to your reply.
>
> Sincerely yours
>
> Dr. Bingliang Zhuang
> ------------------------------------------------------------------------
>
> _______________________________________________
> RegCNET mailing list
> RegCNET at lists.ictp.it
> https://lists.ictp.it/mailman/listinfo/regcnet