From: David Rogers Date: Fri, 23 Oct 1998 15:49:50 -0600 (MDT) Subject: draft topic F2 - dust and IN Dear IN-WG members, The following draft discussion is offered for ICE Initiative topic F2, "What are the strongest arguments for connecting dust with IN?" This discussion is not yet integrated with other topics, although there are some likely ties. For example, it could connect to Marcia's list of measurements for topic W4 (aerosol and cloud measurements) and W6 (laboratory experiments). This topic arose as a suggestion from Al Cooper. The arguments he made for focusing on dust are not in this discussion, which tries to examine the evidentiary question. In a general sense, perhaps a question for upcoming field experiments would be, "What kinds of evidence should be obtained in order to connect ice nuclei with particular aerosol source regions or production mechanisms?" ..dave.. ---------------------------------------------------------------------- ICE - Nucleation Working Group ****Discussion Topics**** FUNDAMENTAL QUESTIONS F2. What are the strongest arguments for connecting dust with IN? For the purposes of this discussion, "dust" is defined as aerosol mineral particles of surface crustal origin and composed primarily of silicon, aluminum, potassium and calcium. Evidence for a relationship between dust and ice nuclei comes from several different types of studies. Some studies involve direct examinations of IN particles, and others are based on inferential evidence. For all of these studies, the generality of the connection with ice nuclei is somewhat uncertain because of various factors: the number of samples is small; the time scale or geographic range where samples were collected is limited; the full suite of measurements needed for a process type description are not available; or the studies were conducted in areas with ongoing cloud seeding operations, so there was potential for contamination of the natural aerosol. The direct evidence comes from identifying the nucleating particles in snow. In studies by Kumai, Iwai and others, precipitating snow crystals were collected, and the particles found at the crystal centers were analyzed using ion microprobe and electron microscope (EM) techniques. The chemical composition of the particles indicated clay materials, including illite, kaolinite, halloysite and other minerals, as well as particles containing sodium chloride. While this evidence is suggestive, the approach has several weaknesses: (1) although the location at the crystal centers indicate the particles were the nuclei, they may have been collected by processes other than nucleation scavenging; (2) when several particles are near the center, there is an inherent bias towards the larger ones (Mossop 1963); (3) the analysis identifies elements, weighted by mass fraction; (4) structure and chemical bonding are not characterized; (5) the technique requires particles larger than ~0.02um; and (6) nucleation occurs on the surface at a particular site, the properties of which are not characterized. In an airborne study, Heintzenberg et al. (1996) used a counter-flow virtual impactor (CVI) to separate cirrus crystals. These crystals were evaporated, and the residual particles were impacted onto transmission EM grids for single particle analysis. The dominant particles were identified as minerals, containing Si and Fe. Similar direct evidence studies used ice nuclei instruments to grow IN to detectable size crystals and then examine particles at the center of the crystal (Grosch and Georgii, Rosinski et al., Chen et al.). Grosch and Georgii collected natural aerosol particles on membrane filters, and processed the filters in a low pressure chamber at -18C and water saturation in order to nucleate ice crystals and grow them to detectable size. The crystals were then transferred to an EM substrate with a cold needle. To get a large enough sample, crystals from 100 filters were collected. X-ray energy spectrometer results showed the ice nuclei contained silicon, calcium and aluminum, and the majority of IN were larger than 1um. Their samples from Israel were heavily contaminated with AgI, apparently from cloud seeding operations. No marine elements were found in any of the ice nuclei samples. Single particle EM studies were performed by Rosinski et al. (1976a; 1976b) and Chen et al. (1998) on ice nuclei that were isolated by ice nuclei detection instruments. In the Rosinski studies, ice crystals of ~20um diameter that nucleated and grew in a mixing cloud chamber (Langer 1973) were impacted onto an inert surface. The impacted ice crystals were sublimed, then an optical microscope was used to find the residual particles, and a tungsten needle was used to pick up the particles and transfer them to a polished beryllium substrate for EM analysis. Samples with the chamber at -12C and -19C on summertime surface air in Colorado showed evidence of AgI attached to clay particles. The presence of AgI was attributed to the effects of cloud seeding operations in previous years. IN were mostly clay minerals, 1-10um diameter, with AgI present in 20-25% of the particles. In the Chen et al. study, airborne samples were taken in the upper troposphere and lower stratosphere. Crystals that grew on ice nuclei in a continuous flow diffusion (CFD) chamber were impacted directly onto a TEM grid. The chamber was operated over a range of temperatures (-27 to -36) and water supersaturations (-8 to +8%). The majority fractions of IN particles were identified as crustal or carbonaceous types. The inferential evidence of a connection between dust and ice nuclei comes from analysis of weather modification in Israel (Gabriel and Rosenfeld 1990). The results suggested that cloud seeding increased precipitation on days with low natural IN concentration, but decreased precipitation when IN concentration was high. IN were measured with membrane filters. Higher natural IN concentrations were associated with days having greater amounts of desert dust, as determined by meteorological trajectories, rain water chemistry and total suspended particulate analyses. Gagin (1965) reported that desert dust, especially loess, produces large quantities of ice nuclei. --- references --- Chen, Y., S.M. Kreidenweis, L.M. McInnes, D.C. Rogers and P.J. DeMott, 1998: Single particle analyses of ice nucleating aerosols in the upper troposphere and lower stratosphere. Geophys. Res. Lett., 25, 1391-1394. Gabriel, K.R. and D. Rosenfeld, 1990: The Second Israeli Stimulation Experiment: Analysis of precipitation on both targets. J. Appl. Meteor., 29, 1055-1067. Gagin, A., 1965: Ice nuclei, their physical characteristics and possible effect on precipitation initiation. Proc. Intl. Conf. Cloud Physics, Tokyo-Sapporo, p.155 Grosch, M. and H-W. Georgii, 1976: Elemental composition of atmospheric aerosols and natural ice forming nuclei. J. Rech. Atmos., X, 227-232. Heintzenberg, J., K. Okada and J. Strom, 1996: On the composition of non- volatile material in upper tropospheric aerosols and cirrus crystals. Atmospheric Research, 41, 81-88. Kumai, M., 1951: Electron-microscope study of snow-crystal nuclei. J. Meteor., 8, 151-156. Kumai, M., 1976: Identification of nuclei and concentrations of chemical species in snow crystals sampled at the South Pole. J. Atmos. Sci., 33, 833-841. Rosenfeld, D. and R. Nirel, 1996: Ice nuclei, rainwater chemical composition, and static cloud seeding effects in Israel. J. Appl. Meteor., 35, 1494-1501. Rosinski, J., G. Langer, C.T. Nagamoto, M.C. Bayard and F.P. Parungo, 1976: Chemical composition of surfaces of natural ice-forming nuclei. J. Rech. Atmos., X, 202-210. Rosinski, J., G. Langer, C.T. Nagamoto and M.C. Bayard, 1976: Detection of silver iodide particles in seeded storms. J. Rech. Atmos., X, 243-248.