UCAR > Communications > Staff Notes > March 1995 Search

Rain, Snow, Ice, Dust:
It's All Grist For
NCAR's Cloud Physicists

How do clouds form? That simple meteorological question has a devilish number of answers. Throw precipitation into the mix, and things get even more complicated. In spite of the attendant difficulties, NCAR's cloud physicists have been forging ahead with observational and theoretical research to define the processes at work in cloud development.

NCAR became a center of cloud physics research from its earliest days, when the late scientists Pat Squires and Doyne Sartor arrived in the 1960s. In the 1970s, more experts gathered here for the multiyear National Hail Research Experiment. After NHRE closed its doors and U.S. efforts at weather modification dwindled, the discipline languished. Still, NCAR retained a nucleus of cloud physicists doing basic research. Now, this group of scientists--one of the discipline's largest--is finding its work once again fashionable, this time in the quest for understanding the earth's climate.

Charlie Knight, for one, is glad. "Precipitation development ranks among the most fundamental and most important areas in atmospheric science," he says.

Charlie Knight. (Photo by Bob Bumpas.)

Among several dozen cloud physicists spread across several NCAR divisions, Charlie is the senior member. He joined NCAR in 1962 and is now a senior scientist in the Mesoscale and Microscale Meteorology Division (MMM). Charlie's wife and colleague, Nancy, also has been a long-time MMM researcher.

The books that line the walls of Charlie's office in the Foothills Lab, including Polymer Chemistry and Crystallography, bear witness to the multidisciplinary skills needed to study clouds. "I was a geologist in school," says Charlie. "Then I got into ice. Ice is a crystal, and I had some expertise in that. Now I'm studying warm rain, which I haven't worked on before."

The problem now capturing Charlie's attention is how tropical clouds form raindrops before they have built upward to the freezing level. It's called warm rain formation, and, says Charlie, it is "one of the classical problems in cloud physics--in fact, Doyne Sartor used to work on this. Warm rain occurs much faster than theory can explain. Exactly how long it should take depends on your theory; how long it does take depends on your observations. Neither is very well defined."

This summer Charlie heads to Florida for the Small Cumulus Microphysics Study, officially SCMS, although most of its participants refer to it as SCUM. The project takes place 3 July to 17 August from a base about 20 kilometers north of Cape Canaveral, the same area studied in 1991 for the Convective and Precipitation/Electrification Experiment (CaPE). NCAR's CP-2 radar will be shipped to its former CaPE site for its last-ever research expedition. The study also involves scientists from NCAR; the Universities of Illinois, Chicago, and Wyoming; the Desert Research Institute (DRI); the New Mexico Institute of Mining and Technology; and the French government's meteorological research agency.

The goal of SCMS is to capture the critical few minutes in which a building cumulus begins to form its tiniest raindrops. On hand to penetrate the cloud at those moments will be NCAR's C-130, Wyoming's King Air, and France's Merlin aircraft (which Charlie describes as "a longer King Air"). It will be the first time both radar and aircraft have focused on rainfall development in clouds this small. Though the experiment's spatial scale is modest, the plans are ambitious.

"All the real action, the important stuff that we don't understand, happens very early. By the time a cloud down there is about two kilometers deep, it starts to make rain," explains Charlie. "Once it's four or five kilometers deep, it hits the freezing level. We have to have the planes near the radar and ready to respond quickly, we hope within five minutes." The entire sequence of events, from initial radar returns to multiple aircraft penetrations, will last no more than a quarter hour.

What might SCMS uncover? "The biggest question is how the rain gets started. Another is how much variability there is," Charlie says. Of particular interest are the size distributions of droplets in each cloud. These spectra are hard to measure in experiments and even harder to explain from theory. "They're important for all kinds of things, like cloud albedo, and they're quite mysterious. There are a lot of bimodal spectra [peaking at two different sizes], and in general there are more small droplets than you would expect from simple condensation theory.

"It's very complicated, very variable. Once in a while, somebody calls me and asks what the droplet sizes are in cumulus. There's no simple answer to that."

As a visitor to Charlie's office might surmise from his potpourri of books, papers, and lab supplies, he always has "three or four things going on at one time." For instance, his crystallographic interests have led him to study the molecular structure that allows fish to survive in freezing seawater. Charlie is now pursuing work on dendritic snowflake growth with DRI's John Hallett. As for the SCMS data to come, "it'll be analyzed for at least five years. If all goes well, I'd love to go to Hawaii and do it all over again in a more maritime setting." Why? Among other reasons, the mid-Pacific air of Hawaii has only one-tenth the number of cloud condensation nuclei (CCN) that are present over Florida.

The Heart of the Matter

CCN are at the root of cloud and precipitation formation. They are also a current interest of Al Cooper, head of MMM's physical meteorology group. Al offers two somewhat tongue-in-cheek definitions for his group's focus: "Physical meteorology is everything that's not dynamic meteorology. We like to think the dynamicists have the Navier-Stokes equation and we have everything else."

Al Cooper. (Photo by Bob Bumpas.)

CCN on the order of 0.01 millimeter (one micrometer) in diameter or larger are the most active in cloud formation. The largest ones, such as sea salt or desert dust, are important in the formation of rain. Al is looking at the impact of these large particles in SCMS. "There have been very interesting preliminary results in South Africa where clouds have been seeded with CCN on the order of one micrometer." In the past, weather modification studies have tried to use CCN as large as 20 micrometers as individual raindrop embryos. The South African work, says Al, is instead trying to stimulate the collision-coalescence process and nudge the entire droplet distribution spectrum toward larger drops. The collision-coalescence process refers to the mechanism by which droplets of different sizes fall at differing rates, causing collisions (just as might happen on an interstate if some cars moved at 80 kilometers per hour and others at 130). These collisions tend to result in larger, coalesced drops that fall even faster, provoking more collisions in a cascading process that culminates in rain.

Al will be examining the droplet distribution in SCMS with an eye toward the presence and evolution of giant particles and resulting droplets. Eventually, he'd like to see better models of the collision- coalescence process. "Do we really understand this well enough to calculate what will happen? I think we understand the basic process, but I'm not sure we have the collisions right." There are several difficulties:

One thing is certain: the earth's atmosphere is loaded with CCN. Currently observed levels are from 50 to 200 CCN per cubic centimeter over the pristine ocean (that is, away from large land sources) and from 500 to 1,000 CCN over land. Some counts in the thousands are observed downwind from urban areas. Says Al, "Worldwide, biomass burning could be the largest source of CCN, especially the burning of grasslands and forests in the tropics. We probably have more CCN now than ever before."

Ironically, the growth in CCN concentrations might work to reduce global precipitation. As more CCN compete for a given amount of water vapor, the range in droplet sizes is reduced (the droplets are smaller and greater in number) and fewer collisions occur. The sheer presence of more aerosols also helps to obscure incoming solar radiation, cooling the atmosphere and making it more stable, a process Al says was "very dramatic" during the Kuwait oil fires studied by NCAR in 1991. Thankfully, the effects were primarily local.

The connections between microscale physics and global climate are the impetus for a new CCN/IN (ice nuclei) counter being developed by Al and Larry Radke, director of NCAR's Research Aviation Facility. "We're calling it a CCN/IN counter, but it's primarily designed to count the nuclei involved in cirrus and contrail formation. If it does all of this, it'll be a great success. If it does only some, it'll still be useful." The instrument will be flown at high altitudes worldwide, first on NASA's DC-8 and eventually on NCAR's WB-57. Meanwhile, Al's group is looking at new ways to use such data. For instance, through a joint appointment between MMM and the Atmospheric Chemistry Division, Mary Barth is studying the effect of clouds on tropospheric chemistry.

Like Charlie, Al is glad to see the resurgence of interest in cloud physics for global change research after its 1960s-70s role in cloud seeding. "This is really not that great a change. Both are questions of weather modification--one intentional and the other unintentional."

Tracing the Life of Ice

Cloud watchers long have marveled at the sublime streaks of classic cirrus clouds, often referred to as mare's tails. MMM's Andy Heymsfield is studying the mechanics behind the beauty of cirrus.

Andy Heymsfield. (Photo by Bob Bumpas.)

The goal isn't to deconstruct the experience of cloud watching but to understand how cirrus evolve by "seeding" themselves and clouds below. Andy recently collaborated with Sam Oltmans (NOAA Aeronomy Laboratory) and MMM associate scientists Larry Miloshevich and Steve Aulenbach to evaluate the humidity profiles and crystal structures throughout a cirrus cloud that formed near Boulder on 10 November 1994. The study used a cryogenic (ice-based) hygrometer to measure relative humidity at temperatures as low as -60 C. A separate instrument package aboard the same radiosonde captured crystals at various levels, drenching and preserving each crystal in a quick-drying liquid plastic. With this technique, says Andy, "you don't have the problems you have with crystal breakup when aircraft are moving at 250 kilometers an hour trying to take samples. The other advantage is that you get a vertical profile instead of a pass through a cloud at one level."

Documenting cirrus is crucial to understanding clouds' effect on global climate. Andy has been at the forefront of the effort to clarify the role of cirrus on global radiation budgets, largely through his participation as a principal investigator in the First ISCCP Regional Experiment (FIRE) and the Central Equatorial Pacific Experiment (CEPEX). ISCCP is the ongoing International Satellite Cloud Climatology Project and CEPEX was a 1992 study of the radiation budget and microphysics of the tropical central Pacific. Satellites are particularly adept at mapping global cirrus distributions. Andy is working to compare research aircraft data to satellite-based water-vapor data in order to calibrate the latter. "This could give us cirrus height, optical depth, and particle size information."

Another focus of Andy's work has been "finding out under what conditions cirrus crystals nucleate." Their formation depends on the saturation vapor pressure of ice. Cloud droplets form when the relative humidity exceeds 100%, a state called supersaturation with respect to water. But for very cold air at humidities just under 100%, the air is unsaturated with respect to water but highly supersaturated with respect to ice, allowing cirrus to form. A climate model thus needs to keep close tabs on temperature and relative humidity in order to treat cirrus formation accurately.

The picture becomes even more complex when ingredients such as sulfuric acid come into play, says Andy. Solutions of water and sulfuric acid are very concentrated at humidities close to the saturation point with respect to ice, and these particles cannot freeze. But as the humidity increases, the solutions become more dilute (due to the added moisture) and freezing becomes more likely.

Andy has joined a number of other NCAR researchers in a modeling and observational study on the effect of cloud ice on precipitation development in the Southwest. The Arizona Project is now wrapping up two months of field work centered on winter storms in the mountainous area between Phoenix and Flagstaff. The project examined high-level mountain wave clouds formed as frontal systems crossed the Mingus Mountain region. These clouds deposit ice crystals into lower-level clouds formed by upslope winds that ascend the Mogollon Rim, which extends southeast from Flagstaff. Later this year, a mesoscale model developed by MMM's Terry Clark will be coupled with an in-depth microphysical model to test how cloud seeding in the region might enhance wintertime precipitation and help to build up Arizona's water storage.

The experiment is building on a number of findings from the multiyear Winter Icing and Storms Project (WISP), in which scientists from NCAR's Research Applications Program have observed and modeled ice and supercooled water in clouds near the Front Range since 1990. The Arizona Project's principal investigator is Roelof Bruintjes (MMM/RAP) .

Next Stop: The Stratosphere

Some cloud physics questions range even higher than cirrus clouds. Over the past decade, the Antarctic's greatly depleted "hole" of stratospheric ozone has been documented, along with lesser but still significant ozone depletions at other latitudes. The crux of the argument for ozone's human-induced depletion has been an interaction between chlorofluorocarbons (CFCs), polar stratospheric clouds, and sunlight. Now NCAR has developed an instrument to help view this and other processes at work on the microscale for the first time.

Jim Dye (MMM) and Darrel Baumgardner and Bruce Gandrud (both of ATD) combined forces with other staff from their divisions to create the multiangle aerosol spectrometer probe (MASP). Winner of the 1994 Technology Advancement Award (see the December 1994 Staff Notes Monthly), MASP promises to shed light on the ozone depletion story, as well as on a more recent development in cloud modeling that's just as sobering in its own way.

Darrel Baumgardner. (Photo by Bob Bumpas.)

"This instrument's advantage," says Darrel, "is that, in addition to measuring the size and concentration of particles, we can derive information on their optical properties. If we know something about their optical properties, we can then deduce something about their chemical compositions." MASP trains a laser beam on a particle at several angles. The particle's refractive index can be derived from the amount of light that is redirected or refracted as it passes through the particle.

The first flights of MASP took place last year aboard NASA's ER-2 and went "quite well," says Darrel, with some interesting results already: "The jury's still out, but I think we're going to find some surprises." The instrument was deployed four times from Christchurch, New Zealand, between March and October 1994 to document the ozone hole and its causative factors. A trek to the Arctic aboard NASA's DC-8 will follow next year. Darrel awaits the findings with interest. "Chlorine is released during reactions on the surfaces of PSC [polar stratospheric cloud] particles that form in the polar winter night. The composition of these PSCs is still in question, but the MASP measurements of these particles hopefully will help to clarify theoretical speculations."

MASP may also prove useful in addressing a newly discovered dilemma in cloud modeling. A study published in Science on 27 January reveals that the global energy budgets normally used in climate models appear to seriously underestimate the percentage of energy absorbed by clouds. NCAR's Jeff Kiehl is one of the coauthors. The paper reports on a comparison between satellite- and ground-based observations that found clouds absorbing more than 25 watts per square meter of energy, rather than the 6 watts predicted by theory.

"Where are those extra watts per meter going? That's where the big mystery is now," says Darrel. He thinks that haze droplets not yet large enough to count as full-fledged cloud droplets may be part of the answer, absorbing or scattering the radiation now unaccounted for. "In cloud measurements, it's typical to ignore droplet sizes below a certain level." MASP will measure particles as small as 0.3 micrometers, which "could be an important size for this kind of scattering."

Darrel knows about unexpected findings in cloud physics. On an expedition off the Florida coast in November 1993, he and King Air pilot Mike Heiting were unnerved by an intense rain gush that ruined one water-sensing probe and damaged another.

The probes, both about three centimeters long and one to two millimeters wide, were inundated by "an incredible mass of water," recalls Darrel. "The Russian probe truly was trashed--the water abraded the probe away." The other probe, which measures liquid water content by monitoring the amount of heat needed to evaporate the water, responded too slowly to the very brief gush and burnt itself out by pumping too much heat to the probe after the gush was over.

"The audio inside the cabin was amazing. It sounded like a giant 'splat.' It makes you ask the question, 'What's going on that could cause such a gush of rain?' It says that there's some incredible variability. Every time we go out, we discover something that tells us just how diverse the atmosphere is." --BH

In this issue...
Other issues of Staff Notes Monthly


Edited by Bob Henson, bhenson@ucar.edu
Last revised: Wed Mar 29 12:54:26 MST 2000