wet radius

[ngam]

    The one thing that's still quite unclear to me is how do we get the "wet radius" exactly?

Originally posted by @ngam in #8 (comment)

[ngam]

@Gorkowski

[Gorkowski]

So one way would be to find the wet radius from the Kohler curve.
graph

Given a saturation ratio (dependent variable), you can get the two possible wet radii non-activated and activated.

Was that what you were thinking? @ngam

There is a situation in this approach as when the saturation ratio goes to 1 (or supersat -> 0) the radius -> maga huge.

[ngam]

Yes!

Given a saturation ratio (dependent variable), you can get the two possible wet radii non-activated and activated.

So, which one do we choose? I assume we use the CCN activation fraction to activate some and leave some wet but unactivated, right? How does time come into play? Is this instantaneous? (I assume sat ratio is time-dependent as the parcel goes up, right?)

More importantly, what happens the following time step? Let's look at a concrete example.

At 0.2% supersat, let's say I have 100 particles of size 0.05 um (following the curve you linked). Then we get 0.25 um and 1.0 um wet radii unactivated and activated, respectively. Let's say I do a calculation elsewhere and I find that 25% will activate. Then, these 100 particles will transform into 75 of size 0.25 and 25 of size 1 um.

The next time step, the latter 25 are activated and will have runaway growth immediately (i.e. will be treated as a droplet after that point). The former 75 will undergo the same process as before, except on a different curve, but that curve will likely never touch 0.2% supersat so what do we do? Are they all activated basically?

[ngam]

Or is this just a parameterization that we can get around if we solve this on our own?

Can we activate particles by simply letting them grow? The growth will increase (as the condensation driving force increases due to supersat, etc.) so we could potentially get this through without having to use any of these elaborate schemes...

[ngam]

From this https://www.nature.com/articles/nature18271, go to "aerosol growth model" --- I envision dynamically resolving that can resolve things. I will ask NMD about it out of curiosity, "can this be applied across scales, e.g. for droplet activation?"

[ngam]

Like there is nothing "magical" about this so-called activation; it is arbitrary. Or am I misunderstanding things? I really don't like to think about equilibrium/steady-state settings. I am more of a dynamics person... lol

[Gorkowski]

Can we activate particles by simply letting them grow? The growth will increase (as the condensation driving force increases due to supersat, etc.) so we could potentially get this through without having to use any of these elaborate schemes...

I think that is how it should be done. I'm trying to understand what they (pyrcel, Nenes, etc.) currently do, as I think they just approximate this step.

There is nothing special about being activated except that the condensation driving force is size dependent due to kelvin effects.

[ngam]

Well, it could be a good exercise to replicate this kohler curve from a dynamics starting point. One of my most productive exercises was to get steady states from arbitrary initial states, that's how I came to trust most of my charge-related work. If your given dynamic model can reproduce these well-known steady states, then you know your model is excellent! This exercise will likely give us a lot of insight. I will try to set it up fully later this week following the paper above, I already have most of the pieces, and a solver, but I will need your help to make sense of these pressures/saturations things because I don't have a good physical intuition for their meanings (yet)

xref uncscode/particula#307