From: Paul Demott Date: Wed, 11 Nov 1998 15:28:31 -0700 (MST) Subject: Draft topic F1 - Evaporation IN (fwd) Dear IN-WG members, Here is a (belated) contribution regarding the posed Fundamental Question F1 regarding evaporation ice nuclei. I welcome your comments. Paul ************************************************************************** Paul J. DeMott Research Scientist Department of Atmospheric Science Colorado State University Fort Collins, CO 80523-1371 (t) 970-491-8257 (f) 970-491-8449 (e) pdemott@lamar.colostate.edu ************************************************************************** F1. How compelling is the evidence that evaporation enhances the number or activity of IN? What are the details known so far? How to become more secure on this issue? 1. The evidence that evaporation enhances IN activity is mostly indirect or inferential. This evidence, at the moment, is perhaps more intriguing than it is compelling. Some field studies have related unusually high ice nuclei numbers or unusual increases in ice crystal numbers to circumstances in which clouds were evaporating. Langer et al. (1969) found IN enhanced in thunderstorm outflow regions compared to surrounding regions of the atmosphere. Cooper (1995) observed the onset of up to a 100 factor increase in ice crystal concentrations in the evaporation region of orographic wave clouds. The largest ice enhancements in the Cooper study were observed in clouds with temperatures approaching the onset temperature for homogeneous freezing. Smaller enhancements were found in warmer clouds and no enhancements were found warmer than about -20 degC. Cooper also noted that ice crystal concentrations do not progressively increase in wave cloud trains, as might be expected if IN were being created by the cloud cycling. Some of the observations of ice crystal number enhancement versus expected IN number in the comprehensive cloud studies of Hobbs and Rangno (1985; 1990; 1994) were also observed to originate in close proximity to regions of cloud evaporation. Nevertheless, these authors have focused attention on the stronger relation between cloud droplet diameter and high ice crystal concentration. Additional evidence for ice nuclei formed in association with evaporation come from studies in which natural air that had been through a condensation process was purposely warmed and evaporated before measuring ice nucleus activity. In three cases, Kassander et al. (1957) found high IN concentrations (~100/L at temperatures between -1 and -14 degC) in cloudy air samples that had been evaporated and warmed to 13 degC prior to cooling in a mixing cloud chamber. Ice formation at such warm temperatures was not noted in sampling any cloud that consisted only of liquid droplets, but only in clouds that also contained ice crystals. These unusual observations have never been reproduced in other airborne IN sampling programs. For example, Rogers et al. (1998) brought air into a warm aircraft cabin from ice clouds on many occasions and did not note enhanced ice nuclei compared to clear air sampling in their continuous flow diffusion chamber. In a filter-processing study of natural IN, Rosinski and Morgan (1991) observed a transient population of sorption ice nuclei created during the evaporation of droplets formed initially on aerosol particles above water saturation. These new ice nuclei remained viable only as long as the air remained supersaturated with respect to ice during evaporation. The IN enhancements of sorption nuclei were up to several per liter, stated to represent 0.00001 to 0.0001 fractions of the droplet population, with little temperature dependence between -4 and -20 degC. The enhancement factor approached a very high level, of the degree needed to explain the pervasive amounts of ice often found in mixed regions of cumuli, in only one of eleven samples. The same study found that the population of condensation-freezing nuclei is actually degraded in successive condensation/evaporation cycles and that the deactivation is greater the more the air volume is warmed during evaporation. These observations are in sharp contrast to the findings of Kassander et al. and appear unable to explain an abundance of IN in thunderstorm outflows. The measurements of Rosinski and Morgan (1991) also suggest that most IN measurement techniques will probably have difficulty detecting populations of transient IN that disappear when sample humidity falls below ice saturation. 2. Some field and laboratory work has been done on this topic in the last several years, but results have not been published in the reviewed literature. This probably speaks for the difficulty of reproducing the conditions under which evaporation ice nuclei are thought to form. At least three laboratories have attempted experimentation on the effects of evaporation on IN as part of broader research projects. During the time period of NCAR's Winter Icing in Storms Project (1991-1994), Colorado State University conducted evaporation/recondensation experiments in a controlled expansion chamber. Aerosol particles from surface-level air and air collected by aircraft in the free troposphere were used in the experiments. Ice enhancements during forced cloud evaporation and in subsequent reformation of clouds ranged from 0 to several times in the temperature range -15 to -33 degC, but concentration changes only rarely exceeded 10/L (Paul DeMott, personal communication). The University of Missouri-Rolla has conducted similar measurements during the last few years. The University of Illinois has also initiated related studies in the laboratory and in field programs. Cooper (1995) performed a more comprehensive analysis of enhancements noted in the downstream regions of wave clouds than appeared in his conference paper, but these results are also not yet published. 3. Mechanisms have been proposed to explain ice enhancements of the type observed by Rosinski and Morgan (1991). These mechanisms involve the effects of organic films or surface charges on droplet freezing (Beard, 1991) and surface kinetic effects on droplet cooling during evaporation (Cooper, 1995). These hypotheses have not been rigorously pursued experimentally. These mechanisms are attractive because only small fractions of cloud drops would need to freeze to greatly enhance ice concentrations. 4. How to become more secure on the relation between evaporation and IN number? a. New measurements in wave clouds, including IN measurements upstream and downstream of clouds, would help address the effects of the cloud processing and evaporation on IN. Use of a counterflow virtual impactor upstream of an IN processing device is one means of providing only cloud particle residues. Keeping air at least ice saturated in the process may not be an easy task. In such an experiment, one must assure that ice in downstream leg of cloud could not have initiated aloft of flight path. Cloud profiling with a millimeter radar in combination with aircraft data would provide useful guidance. Measurements of cloud particle populations should be made down to small particle sizes (e.g., using the CPI device) in order to more accurately determine the evolution of ice particle concentrations in evaporation regions. Similar measurements, including greater detail on particle size spectra, in cloud top and edge regions of cumuli may also be useful, albeit harder to interpret. b. There are some laboratory experiments that should give clear answers to the question of the effects of evaporation on ice nuclei concentration. These experiments must be designed to consider the likely transient nature of resulting IN populations. One example is an expansion chamber experiment in which the condensation level is repeatedly crossed at various supercooled temperatures. Experiments can also be envisioned using a continuous flow diffusion chamber (CFD). For example, it is possible to expose air to a condensation/evaporation cycle at a low temperature just upstream of processing through a CFD. Some optical and electrodynamic levitation devices now exist to isolate and observe the freezing behavior of single droplets at low temperatures and controlled humidity. Finally, analysis of the contact freezing ability of evaporated cloud residues should be pursued using a cold-stage precipitator (Vali, 1976). Any of these experimental methods could also explore the effects of droplet composition, organic contaminants and particle charge on ice formation. References: Cooper, W.A., 1995: Ice formation in wave clouds: Observed enhancement during evaporation. Conf. on Cloud Physics, Amer. Meteor. Soc. (Boston), 147-152. Hobbs, P.V. and A.L. Rangno, 1990: Ice particle concentrations in clouds. J Atmos. Sci., 42, 2523-2549. Hobbs, P.V. and A.L. Rangno, 1990: Rapid development of high ice particle concentrations in small polar maritime cumuliform clouds. J Atmos. Sci., 47, 2710-2722. Kassander, A.R., L.L. Sims and J.E. McDonald, 1957: Observations of freezing nuclei over the southwestern U.S., In: Artificial Stimulation of Rain. Pergamon, New York, 392-403. Langer, G., G. Morgan, C.T. Nagamoto, M. Solak and J. Rosinski, 1979: Generation of ice nuclei in the surface outflow of thunderstorms in northeast Colorado. J. Atmos. Sci., 36, 2484-2494. Rangno, A.L. and P.V. Hobbs, 1994: Ice particle concentrations and precipitation development in small continental cumuliform clouds. Q.J.R. Meteorol. Soc., 120, 573-601. Rogers, D.C., P.J. DeMott, S.M. Kreidenweis, and Y. Chen, 1998: Measurements of ice nucleating aerosols during SUCCESS. Geophys. Res. Lett., 25, 1383-1386. Rosinski, J. and G. Morgan, 1991: Cloud condensation nuclei as a source of ice-forming nuclei in clouds. J. Aerosol Sci., 2, 123-133. Vali, G., 1976: Contact-freezing nucleation measured by the DFC instrument. In: Intnl. Workshop on Ice Nucl. Meas. (G. Vali, Ed.), Univ. of Wyoming Report, Laramie, 159-178.