Sunday, September 22, 2013


In honor of the coming snows, Eli thought it would be interesting to read a June paper from Nature, "The importance of feldspar for ice nucleation by mineral dust in mixed phase clouds" by Atkinson, et al, including another Tamsin, Malkin. (for those interested in different names there is great fun at Lawyers Guns and Money this weekend).  There appears to be at least one follow on from this group at Atmospheric Physics and Chemistry, and there is a horse race, involving a UToronto group

The issue is whether there is a difference in the freezing temperatures of water droplets nucleated around small dust particles (solid aerosols).  This is one of those things that at first glance you think WTF, but as soon as you know the result the reason is pretty obvious.  So these folks made up a bunch of aerosols using different species of mineral dust, like the kind that comes off deserts and other dry places, drove the dust through water vapor until they picked up enough water to be between 14 and 16 microns diameter and then cooled them at 1K/min.  They could observe the freezing very Millikan like through a microscope.

The issue about which does what to the production of ice crystals depends not only on the temperature at which the particles freeze, but also on the relative amounts of  each type of dust (let the bunnies not worry about mixed dust particles at this point).  The authors used their data to model the effects of the differential freezing point on mixed clouds (those with solid and liquid water aerosols).  Even tho feldspar is not so common (the fraction, of course as in all such things varies from 25% on down to less that .1% depending), it turns out to have an important effect.

Eli is not going to get too deeply into the modeling, but indeed the idea is interesting.  Obviously something is going on at the interface which causes water molecules to bind tightly.  Turns out that there is a lot of oldish work on dielectric constants and dissolution of minerals, and IEHO the answer lurks there.  For example, from 1980 a paper on dielectric constant and loss tangent angles of Indian rock samples by Singh, Singh and Lal showing a significant enhancement of the dielectric constant of moist feldspar dust and not so much for stuff like marble and quartz.  Also, comments here and there about how the alkalai ions at the surface of the feldspar can be solubalized, leaving a charged surface.

A geologist could probably say something more than this handwaving, but it sure looks like water molecules can bind strongly to the feldspar surface providing excellent nucleation sites for freezing.

UPDATE:  Just pointing out for the aerosol fans, that Eli has been on a bit of an aerosol kick lately

Ashes to ashes,
Dust in clouds
If you ain't sneezing
Then the ice is freezing

(Horatio does it better)


penguindreams said...

No so much a question of whether, but of the meteorological importance.

Ice Forming Nuclei (IFN in the literature) are much rarer than Cloud Condensation Nuclei (CCN). In the free atmosphere, we're almost never limited in CCN. But in some regions, particularly if you're far from land, it's easy to be limited for IFN.

If you're starved for IFN, then clouds have to get thick enough for their tops to be ca. -40 in order to start making ice crystals and thence initiate efficient rain-producing processes. If you have plenty of IFN, you can start raining out of a cloud that's only -10 to -20 at its top -- 2-5 km shallower/thinner.

So the sources of IFN have substantial control on just how easy it is to produce rain clouds.

A different effect is that water clouds are brighter than ice clouds. Rather, they're more reflective == higher albedo. If the clouds are starved for IFN, they're brighter, and exert a greater cooling effect.

Anonymous said...

Apologies in advance for taking a side-burrow on this thread, but Australians and streaming-savvy international folk might be interested to know that David Suzuki is appearing as the sole interviewee on tonight's episode of the Australian ABC's Q and A:

Should be very interesting...

Bernard J.

Russell Seitz said...

Bear in mind that, geochemistry being messy , a healthy number of the alkali ions on a feldspar surface are cesium and rubidium, which are about as electropositive as thungs get.

EliRabett said...

True enough Russ, but Cs and Rb go much more easily into solution. However, there is also the point that water is extremely polar, so, if the feldspar surface is positive, the O end will attach, if negative the H end. IEHO, more likely the later based on earlier studies which show the surface to have a negative charge.

As was said, while the answer is obvious, the reasons for the answer require some more careful analysis.

Hank Roberts said...

Glad to see thatIce-Nine is well off the upper right corner of that graph, at least for now. Though it sounds like they're working out structures that could promote its formation.

Fixed Carbon said...

Have you alerted Judy Curry about this result?

WHT said...

FC said:
"Have you alerted Judy Curry about this result?"

I have alerted Curry that one of her favored skeptical blog commenters Oliver Manuel is an offender registered at the Missouri Highway Patrol site but she keeps on deleting my comment.

Yet she allows O Manuel to keep commenting in spite of this.

The point is that Curry has an iron constitution and can not be swayed by popular thought, no matter how obvious.

Remember, it's all about the uncertainty!

WHT said...

I do enjoy this topic of particulates. Cirrus clouds show an anomaly of a particular crystal size down amidst the dispersive tail of particulate sizes. I wrote about this bimodal distribution here a few years ago:

I notice some of my links are dead. These results were all inferred from sounding measurements. I can't recall if they could pinpoint the crystal seeding impurity responsible for the bump in the distribution.

Ian Forrester said...

Some of the best ice nucleating agents are certain bacteria and the proteins on their cell surface. Ice nucleation can occur at temps of 6 or 7 C below freezing.

Many ski areas use them in the water they use to make artificial snow. Ice nucleating bacteria have been found in clouds all over the globe including Antarctica.

Hank Roberts said...

It'd be most interesting to test other material that works (maybe along with mineral dust?) to form water and ice:

Too bad few microorganisms from soil and seawater grow out successfully in labs. Mostly what's there we know only from DNA probes.

Even the plankton in the ocean is barely identified; we don't have baselines yet so can't identify trends.

Damn, dropped another wrench into the gearbox ... oh well, maybe it won't hurt anything ...

Russell Seitz said...

It is hardly remarkable that an Iron Sun crank like Manuel should be magnetically attacted to Gurry's iron constitution.

It is clear to the Iron Sun community that upward temperature trends have been retarded by excessive spending on cold fusion research

Hank Roberts said...

Apropos plankton:

"The goal of the site is to classify underwater images in order to study plankton. ...

The site shows you one of millions of plankton images taken by the In Situ Ichthyoplankton Imaging System (ISIIS), a unique underwater robot engineered at the University of Miami. ISIIS operates as an ocean scanner
that casts the shadow of tiny and transparent oceanic creatures onto a very high resolution digital sensor at very high frequency.

The dataset used for Plankton Portal comes a period of just three days in Fall 2010.... they need your help!
... to distinguish different species within an order or family ... is still best done by the human eye. ...
A field guide is provided, and there is a simple tutorial. The science team will be on Plankton Portal Talk to answer any questions, and the project is also on ... "

those newfangled telecommie things the youngsters use.

David B. Benson said...

Horatio does it *much* better. Maybe offer him a commission now and again?