• Brownscombe, J. L., and J. Hallett, 1967: Experimental and field studies of precipitation particles formed by the freezing of supercooled water. Quart. J. Roy. Meteor. Soc., 93 , 455473.

    • Search Google Scholar
    • Export Citation
  • Cannon, T. W., J. E. Dye, and V. Toutenhoofd, 1974: The mechanism of precipitation formation in northeastern Colorado cumulus. II. Sailplane measurements. J. Atmos. Sci., 31 , 21482151.

    • Search Google Scholar
    • Export Citation
  • Chisnell, R. F., and J. Latham, 1976: Ice particle multiplication in cumulus clouds. Quart. J. Roy. Meteor. Soc., 102 , 133156.

  • Choularton, T. W., J. Latham, and B. J. Mason, 1978: A possible mechanism of ice splinter production during riming. Nature, 274 , 791792.

    • Search Google Scholar
    • Export Citation
  • Choularton, T. W., D. J. Griggs, B. Y. Humood, and J. Latham, 1980: Laboratory studies of riming, and its relation to ice splinter production. Quart. J. Roy. Meteor. Soc., 106 , 367374.

    • Search Google Scholar
    • Export Citation
  • Coons, R. D., and R. Gunn, 1951: Relation of artificial cloud-modification to the production of precipitation. Compendium of Meteorology, T. F. Malone, Ed., Amer. Meteor. Soc., 235–241.

    • Search Google Scholar
    • Export Citation
  • Dye, J. E., and P. V. Hobbs, 1966: Effect of carbon dioxide on the shattering of freezing water drops. Nature, 209 , 464466.

  • Dye, J. E., and P. V. Hobbs, 1968: The influence of environmental parameters on the freezing and fragmentation of suspended water drops. J. Atmos. Sci., 25 , 8296.

    • Search Google Scholar
    • Export Citation
  • Fletcher, N. H., 1962: The Physics of Rainclouds. Cambridge University Press, 386 pp.

  • Gerber, H., B. G. Arends, and A. S. Ackerman, 1994: New microphysics sensor for aircraft use. Atmos. Res., 31 , 235252.

  • Goldsmith, P., J. Gloster, and C. Hume, 1976: The ice phase in clouds. Preprints, Int. Conf. on Cloud Physics, Boulder, CO, Amer. Meteor. Soc., 163–167.

  • Griggs, D. J., and T. W. Choularton, 1983: Freezing modes of riming droplets with application to ice splinter production. Quart. J. Roy. Meteor. Soc., 109 , 243253.

    • Search Google Scholar
    • Export Citation
  • Hallett, J., and S. C. Mossop, 1974: Production of secondary ice particles during the riming process. Nature, 249 , 2628.

  • Hobbs, P. V., 1969: Ice multiplication in clouds. J. Atmos. Sci., 26 , 315318.

  • Hobbs, P. V., and A. J. Alkezweeny, 1968: The fragmentation of freezing water droplets in free fall. J. Atmos. Sci., 25 , 881888.

  • Hobbs, P. V., and A. L. Rangno, 1985: Ice particle concentrations in clouds. J. Atmos. Sci., 42 , 25232549.

  • Johnson, D. A., and J. Hallett, 1968: Freezing and shattering of supercooled water drops. Quart. J. Roy. Meteor. Soc., 94 , 468482.

  • Knight, C. A., and N. C. Knight, 1974: Drop freezing in clouds. J. Atmos. Sci., 31 , 11741176.

  • Koenig, L. R., 1963: The glaciating behavior of small cumulonimbus clouds. J. Atmos. Sci., 20 , 2947.

  • Koenig, L. R., 1968: Some observations suggesting ice multiplication in the atmosphere. J. Atmos. Sci., 25 , 460463.

  • Kolomeychuk, R. J., D. C. McKay, and J. V. Iribarne, 1975: The fragmentation and electrification of freezing drops. J. Atmos. Sci., 32 , 974979.

    • Search Google Scholar
    • Export Citation
  • Korolev, A. V., M. P. Bailey, J. Hallett, and G. A. Isaac, 2004: Laboratory and in situ observation of deposition growth of frozen drops. J. Appl. Meteor., 43 , 612622.

    • Search Google Scholar
    • Export Citation
  • Kostinski, A. B., and A. R. Jameson, 1997: Fluctuation properties of precipitation. Part I: On deviations of single-size drop counts from the Poisson distribution. J. Atmos. Sci., 54 , 21742186.

    • Search Google Scholar
    • Export Citation
  • Lawson, R. P., and T. L. Jensen, 1998: Improved microphysical observations in mixed phase clouds. Preprints, Conf. on Cloud Physics, Everett, WA, Amer. Meteor. Soc., 451–454.

  • Ludlam, F. H., 1952: The production of showers by the growth of ice particles. Quart J. Roy. Meteor. Soc., 78 , 543553.

  • Ludlam, F. H., 1955: Artificial snowfall from mountain clouds. Tellus, 7 , 277290.

  • Mason, B. J., and J. Maybank, 1961: The fragmentation and electrification of freezing water drops. Quart. J. Roy. Meteor. Soc., 87 , 113114.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., 1978: The influence of drop size distribution on the production of secondary ice particles during graupel growth. Quart. J. Roy. Meteor. Soc., 104 , 323330.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., 1985: Secondary ice particle production during rime growth: The effect of drop size distribution and rimer velocity. Quart. J. Roy. Meteor. Soc., 111 , 11131124.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., and J. Hallett, 1974: Ice crystal concentration in cumulus clouds: Influence of the drop spectrum. Science, 186 , 632634.

    • Search Google Scholar
    • Export Citation
  • Murgatroyd, R. J., and M. P. Garrod, 1960: Observations of precipitation elements in cumulus clouds. Quart. J. Roy. Meteor. Soc., 86 , 167175.

    • Search Google Scholar
    • Export Citation
  • Ono, A., 1972: Evidence on the nature of ice crystal multiplication processes in natural cloud. J. Rech. Atmos., 6 , 399408.

  • Phillips, V. T. J., A. M. Blyth, P. R. A. Brown, T. W. Choularton, and J. Latham, 2001: The glaciation of a cumulus cloud over New Mexico. Quart. J. Roy. Meteor. Soc., 127 , 15131534.

    • Search Google Scholar
    • Export Citation
  • Pitter, R. L., and H. R. Pruppacher, 1973: A wind tunnel investigation of freezing of small water drops falling at terminal velocity in air. Quart. J. Roy. Meteor. Soc., 99 , 540550.

    • Search Google Scholar
    • Export Citation
  • Rangno, A. L., and P. V. Hobbs, 1991: Ice particle concentrations and precipitation development in small polar maritime cumuliform clouds. Quart. J. Roy. Meteor. Soc., 117 , 207241.

    • Search Google Scholar
    • Export Citation
  • Rangno, A. L., and P. V. Hobbs, 2005: Microstructures and precipitation development in cumulus and small cumulonimbus clouds over the warm pool of the tropical Pacific Ocean. Quart. J. Roy. Meteor. Soc., 131 , 639673.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., and I. M. Lensky, 1998: Satellite-based insights into precipitation formation processes in continental and maritime convective clouds. Bull. Amer. Meteor. Soc., 79 , 24572476.

    • Search Google Scholar
    • Export Citation
  • Scott, B. C., and P. V. Hobbs, 1977: A theoretical study of the evolution of mixed-phase cumulus clouds. J. Atmos. Sci., 34 , 812826.

  • Stith, J. L., J. A. Haggerty, A. Heymsfield, and C. A. Grainger, 2004: Microphysical characteristics of tropical updrafts in clean conditions. J. Appl. Meteor., 43 , 779794.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., and Coauthors, 2003: Improvement of microphysical parameterization through observational verification experiment. Bull. Amer. Meteor. Soc., 84 , 18071826.

    • Search Google Scholar
    • Export Citation
  • Takahashi, C., and A. Yamashita, 1970: Shattering of frozen water drops in a supercooled cloud. J. Meteor. Soc. Japan, 48 , 369372.

  • Takahashi, T., and N. Fukuta, 1988: Observations of the embryos of graupel. J. Atmos. Sci., 45 , 32883297.

  • Vaughan, H. C., 1954: The spontaneous freezing temperatures of melted snow and of small water drops. Bull. Amer. Meteor. Soc., 35 , 5255.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Examples of particle images from the Stratton Park Engineering Company Cloud Particle Imager at temperatures between 8° and 10°C suggesting that the particles are solid. (a) Small drops with pieces missing or that appear cracked. (b) Larger drizzle-sized drops that are substantially out of round suggesting deformities produced by freezing.

  • View in gallery

    Examples of the classification of ice particles made in this paper from the Stratton Park Engineering Company Cloud Particle Imager. (a) Unidentifiable amorphous fragments, (b) columnar (note first particle in row appears to be viewed on end), (c1), (c2) frozen drops, and (d) partial frozen drops and possible fragments that might have been due to shattering–fragmentation upon freezing. The confidence that these are due to shattering–fragmentation decreases from left to right. No particle without a rounded section is included in the inventory for those particles that are considered possibly due to fragmentation of freezing drops.

  • View in gallery

    Example of cloud microstructural data obtained in level flight on 18 Jan 2001 at −2.5°C and 708 hPa. The droplet concentrations originate with a FSSP-100. The liquid water contents from the integrated FSSP-100 spectrum (dark line) and a Gerber Particle Volume Monitor-100 (light line). The ice particle concentrations are from a 2D-C for those particles at least 100 μm in maximum dimension.

  • View in gallery

    Vertical aircraft soundings on (a) 18 and (b) 20 Jan 2001 also showing the concentrations of droplets greater than 24 μm in diameter. The vertical dashed lines represent the boundaries of the Hallett–Mossop riming and splintering temperature zone as determined by Mossop (1985).

  • View in gallery

    Examples of 2-DC imagery in which it was deduced that ice particles (a) were steadily forming in a quasi–steady state process, (b) are forming in situ due to the freezing of drizzle drops, and (c) are forming where columnar ice particles have recently formed. The length of the thin vertical lines separating particles is 800 μm.

  • View in gallery

    Example of cloud microstructural data obtained in level flight on 20 Jan 2001 at −3°C and 730 hPa. The origin of the data and symbols are as in Fig. 3.

  • View in gallery

    An example of the localized regions of liquid drizzle drops that were occasionally intercepted. Note splashes at the end of some rows confirming the liquid phase of at least some drops. The length of the thin vertical lines separating particles is 800 μm.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 96 60 2
PDF Downloads 42 28 2

Fragmentation of Freezing Drops in Shallow Maritime Frontal Clouds

View More View Less
  • 1 Cloud and Aerosol Research Group, Department of Atmospheric Sciences, University of Washington, Seattle, Washington
© Get Permissions
Full access

Abstract

Images of frozen drops with pieces missing were collected on two days of airborne sampling in shallow supercooled stratiform frontal clouds in the coastal waters of Washington State. In those limited regions where ice appeared to be newly formed, ice fragments with rounded portions accounted for about 5% of the total ice particle concentrations. These results are in rough agreement with the body of literature on laboratory experiments concerning the freezing of drops in free fall that have suggested a modest, though not insignificant, role for the fragmentation of freezing drops on total ice particle concentrations when larger supercooled drops are present.

Corresponding author address: Arthur L. Rangno, Sky Guide, P.O. Box 30027, Greenwood Station, Seattle, WA 98113-2027. Email: skyguide@comcast.net

Abstract

Images of frozen drops with pieces missing were collected on two days of airborne sampling in shallow supercooled stratiform frontal clouds in the coastal waters of Washington State. In those limited regions where ice appeared to be newly formed, ice fragments with rounded portions accounted for about 5% of the total ice particle concentrations. These results are in rough agreement with the body of literature on laboratory experiments concerning the freezing of drops in free fall that have suggested a modest, though not insignificant, role for the fragmentation of freezing drops on total ice particle concentrations when larger supercooled drops are present.

Corresponding author address: Arthur L. Rangno, Sky Guide, P.O. Box 30027, Greenwood Station, Seattle, WA 98113-2027. Email: skyguide@comcast.net

1. Introduction and background

It has been more than 40 years since Koenig (1963) reported on what became one of the greatest enigmas in the field of cloud physics when he described the glaciating behavior of cumulus clouds that were never colder than −10°C. Koenig wrote, “. . . the (slightly supercooled) clouds were observed to develop large concentrations of ice particles in comparatively short time (less than ten minutes).”

Koenig’s findings, acquired against the backdrop of a large randomized cloud seeding experiment in Missouri, were unexpected because it was generally believed that cloud-top temperatures lower than about −15° to −20°C were required for ice to form (e.g., Ludlam 1955; Fletcher 1962). Also, the speed at which the clouds glaciated was equally inexplicable.

A few isolated reports before Koenig’s had, in fact, suggested that ice was appearing in slightly supercooled clouds (e.g., Coons and Gunn 1951; Ludlam 1952; Vaughan 1954; Murgatroyd and Garrod 1960), but these did not receive much weight, perhaps because they were rather few, based on visual observations or, in the case of Murgatroyd and Garrod, based on a rather small sample size of ice-containing clouds.

It was thought for a time that fragmentation of freezing drops could be a strong factor in producing ice in slightly supercooled clouds, such as in those studied by Koenig (1963). Mason and Maybank (1961) reported that freezing drops shattered explosively in their laboratory experiments. However, Dye and Hobbs (1966, 1968) discovered that the high incidence of shattering observed by Mason and Maybank was an artifact caused by the excessive amount of CO2 in their cloud chamber. Solid CO2 had been used by Mason and Maybank as a chilling agent; it became dissolved in the droplets, and was explosively expressed during freezing.

Following the Dye and Hobbs reports, the thought that the shattering of supercooled drops was an important contributor to ice in clouds faded even though subsequent careful laboratory experiments confirmed that it did occur under some circumstances. For example, Brownscombe and Hallett (1967), in experiments using drop sizes between 20 and 80 μm diameter that fell on an ice substrate at temperatures between −10° and −20°C, suggested that drops might shatter when landing on points of ice that were less than one-tenth the area of the drop surface. They reasoned that the heat of fusion could not be carried away quickly enough by a narrow point of contact with ice, and thus symmetric freezing that would encapsulate still unfrozen water would occur. Johnson and Hallett (1968) demonstrated that fragmentation could occur in some situations in experiments in free convection conducted between temperatures of −4° and −25°C using drops of 100-μm diameter. Fragmentation could occur, they concluded, if the drop experienced symmetric cooling in free fall, which they deemed unlikely in the free atmosphere.

However, Hobbs and Alkezweeny (1968) reported that symmetric cooling apparently occurred when tumbling motion was initiated as drops in free fall began to freeze. The symmetric cooling produced an ice shell around the drop’s liquid center that then expanded and broke the shell as freezing progressed inward. The Hobbs and Alkezweeny experiments were conducted on drops of 50 to 130 μm diameter at temperatures between −20° and −32°C and at −8°C. In experiments similar to those of Hobbs and Alkezweeny (1968), Pitter and Pruppacher (1973) reported that a noticeable fraction of their precipitation-sized drops (280–650 μm in diameter) developed spicules when freezing and then broke off from the drop during free fall. The drops they used were cooled to between −10° and −28°C. However, they did not observe shattering of drops per se.

Takahashi and Yamashita (1970) also confirmed that drizzle and subdrizzle sized drops (75 to about 175 μm in diameter) fragmented during freezing when they dropped such drops in a 14-m-high tube located outdoors adjacent to a building. They experimented at temperatures between −6° and −24°C. They found that 37% of the drops fractured on the way down, most into only two pieces.

Most of the above experiments were summarized by Kolomeychuk et al. (1975) who also conducted drop freezing–shattering experiments. The fragmentation of freezing precipitation-sized drops (1.2 to 1.90 mm in diameter) was also observed in their experiments. In these experiments, drops were supercooled from −12° to −25°C in a stream of dry or saturated with water vapor nitrogen. The drops rose in position in a vertical wind tunnel and began to tumble and spin as they froze, the latter observation being similar to that of Hobbs and Alkezweeny (1968) and Pitter and Pruppacher (1973). Some of the drops that fractured also produced numerous splinters. They concluded that the tumbling–spinning induced by the initiation of freezing was important in the causing a symmetric cooling of the outer drop surface.

Choularton et al. (1978) found several shards of ice downstream from frost crystals being rimed by droplets at −7° to −8°C in their laboratory experiments. In these experiments, drops were collected on frost crystals approximately 100 μm in maximum dimension that fell through a supercooled cloud. Ninety-nine percent of the droplets in the supercooled cloud had diameters less than 20 μm, but well below those deemed suitable for riming and splintering (24 μm in diameter) from the results of the original experiments (Hallett and Mossop 1974; Mossop and Hallett 1974). Choularton et al. speculated that some of the droplets had completely shattered to produce the shards that they found downstream of the rimed crystals.

Choularton et al. (1980), following up the work of Choularton et al. (1978), deduced that the ice shards were likely due to the occasional explosive freezing of the spicules that formed when drops larger than 24-μm diameter froze rather than from the shattering of whole drops since only spicules were observed when drops froze. These results were more compatible with the riming and splintering hypothesis of Hallett and Mossop (1974). In these experiments, Choularton et al. (1980) moved a frosty metal rod at 1.5 to 3 m s−1 in a cloud of supercooled drops at temperatures of −3° to −8°C.

In a further set of experiments Griggs and Choularton (1983), building on the results of Choularton et al. (1980) found that the peak frequency of protuberance production of rimed drops was at −7°C when using drops 40 μm in diameter. The apparatus and conditions were essentially those in their previous experiments except that the temperature of the experiments was expanded from −1° to −15°C. However, when a mixture of small (<20 μm) and larger (>40 μm) drops was used, the protuberance production peak rose from −7° to −4.5°C. This reinforced the belief that a small point of contact with ice provided by a small rimed drop onto which a larger drop impacted was crucial to splinter production as Goldsmith et al. (1976) and Mossop (1978) had inferred.

While the shattering of freezing drops as a factor in the production of secondary ice was revived by Hobbs and Alkezweeny (1968) and Johnson and Hallett (1968), among others, shattering was thought to have only a relatively modest effect, far less than had been demonstrated by riming and splintering in which tiny shards are emitted by relatively few cloud drops.

Also, field evidence for drop shattering was generally lacking and was limited to reports by Knight and Knight (1974) and Takahashi and Fukuta (1988). These researchers found partial frozen drops as embryos of hailstones and graupel, respectively. Knight and Knight, in fact, found that more than half of hailstone embryos that they examined consisted of fragmented drops. A partial drop was also photographed by Cannon et al. (1974) while studying ice initiation in summer Colorado cumulus congestus clouds.

Subsequently, in view of the lack of substantial field evidence—in fact, even counterevidence by Koenig (1968) and Ono (1972), who found little or no evidence of drop shattering events in the ice crystals they examined at the ground—the importance of the fragmentation of freezing drops faded. The last numerical models of cumulus glaciation that contained factors for drop fragmentation/shattering during freezing until the current era were those by Chisnell and Latham (1976) and Scott and Hobbs (1977). These model calculations indicated that drop fragmentation decreased the time for the buildup of ice particle concentrations to high values. Recently, Phillips et al. (2001) have reincorporated fragmentation of freezing drops as a factor in their numerical model of cumulus glaciation. Further, two recent field reports containing high-resolution imagery suggested that drops had fragmented during freezing (Korolev et al. 2004; Rangno and Hobbs 2005).

In this paper we report further evidence that fragmentation of freezing drops occurs in natural clouds. The recent images of fragmented frozen drops were obtained in the coastal waters of Washington State in two shallow (tops warmer than −14°C), multilayered maritime cloud systems containing low cloud droplet concentrations, significant concentrations of supercooled drops between 50 and 500 μm diameter (embryo drizzle drops and drizzle drops), and moderately high ice particle concentrations.

2. Instrumentation and data analysis

The images shown in this paper were collected on 18 and 20 January 2001 by the University of Washington’s Convair-580 (CV-580) aircraft. These flights were part of the Improvement of Microphysical Parameterization through Observational Verification Experiment (IMPROVE 1; Stoelinga et al. 2003).

The liquid water measurements made on these two flights were derived from a Particle Measuring Systems Forward Scattering Spectrometer Probe (FSSP-100) and a Gerber Scientific Particle Volume Monitor (PVM-100: Gerber et al. 1994). The two liquid water content measurements were virtually identical to each other in liquid phase clouds and were highly correlated (r2 = 0.94), adding credibility to the droplet spectra measured by the FSSP-100.

A Cloud Particle Imager (hereafter CPI) described by Lawson and Jensen (1998) was also aboard the CV-580. This instrument carries a 2.3-μm pixel resolution, charged-coupled-device camera that allows cloud particles to be seen in remarkable detail and at much smaller sizes than had been available previously on aircraft. Generally, the phase of particles (except drops) can be detected to sizes of about 30–50 μm in maximum dimension when they are in focus. The sample rate of the CPI is described by the manufacturer as about 0.35 L s−1 at 90–100 m s−1, the true airspeeds at which the CV-580 generally flew.

However, working with the CPI images can pose challenges. For example, some liquid drops imaged at temperatures above 10°C can appear to have pieces missing or cracks in them, or appear to be fragments of solid particles (Fig. 1a). Also, out-of-round drops imaged at temperatures well above freezing can also be mistaken for frozen drops with bulges (Fig. 1b). Thus, a significant amount of inspection of images at well-above freezing temperatures is necessary to be able to become adept at recognizing spurious “frozen drops” and spurious “drops with pieces missing” in supercooled regions so that the contribution of these kinds of particles is minimized, though probably never eliminated.

It is also difficult to reliably discriminate frozen from liquid drops at sizes much below 100-μm diameter since the evidence of a bulge or spike in such a frozen drop might be too subtle in the CPI imagery. Many frozen drops at these sizes also may not exhibit spikes or noticeable bulges at all (cf. Takahashi and Yamashita 1970). Hence, only those frozen drops exhibiting spikes or bulges have a possibility of being identified. Thus, frozen drops in this inventory are certain to be underrepresented. The CV-580 did not carry a Rosemount icing meter, which allows the determination of high concentrations of spherical frozen cloud-sized drops (>10 000 per liter) as was able to done by Stith et al. (2004) in conjunction with CPI images and FSSP-100 concentrations.

Also, due to the various orientations a crystal may present to the CPI’s camera, identifying the ice crystal habit seen in the CPI imagery also poses a challenge. A columnar ice particle seen from various perspectives and not in exact focus might be misconstrued as a “frozen drop with a spicule” if the crystal is captured by the camera while presenting a mainly basal face (see Fig. 2b, first particle). It is also almost impossible to identify the nature of particles below about 30-μm diameter except those that are columnar and exhibit strong rectangular geometries to the camera. Since all identifiable particles were inventoried, columnar particles are probably overrepresented in this inventory since they are confidently recognizable to smaller sizes than are other particles in the CPI imagery.

The inspection of drops at above freezing temperatures was followed by a classification of each particle in the supercooled frontal clouds by eye. This was done in several regions of flight legs where ice formation appeared to be just underway. Some examples of how the CPI images were categorized are shown in Fig. 2. The author estimates that the reliability of these classifications, which, of course, cannot be independently confirmed, is better than 70% and, as noted, the errors in classifications are biased toward undercounting frozen drops and overcounting columnar particles.

Some of the ice fragments counted could have originated with the cylindrical entrance to the sampling tube of the CPI, so the total of fragments cannot be considered reliable without some estimate of the artifacts produced by the probe itself. However, no studies to determine the contribution of fragments generated by the inlet to the sampling tube of the CPI have been conducted.

Last, groupings of small particles in a single camera image appear regularly with the CPI. Such bursts typically contain a few to more than 10 small particles. CPI frames with five or more particles are rejected in this inventory, presuming that these bursts might be artifacts from larger crystals breaking up on the leading edge of the probe rather than a tight clustering of real particles. This is another element of uncertainty in the CPI analysis since it is known that cloud and precipitation particles do congregate in higher concentration filaments and clusters (e.g., Hobbs and Rangno 1985; Kostinski and Jameson 1997).

The ice particle concentrations quoted in this paper and shown in the figures are derived from the Particle Measuring Systems 2D-C probe and are not from the CPI due to uncertainties associated with the latter probe. For the 2D-C concentrations, for continuity purposes, we continue our past methods of deriving concentrations (e.g., Rangno and Hobbs 1991); we do not include particles smaller than 100 μm in maximum dimension in the calculation of the concentrations, though such particles are likely very numerous. Second, we do not account for the shrinking depth of field of the 2D-C for those particles less than 150 μm in maximum dimension. Consequently particles in the sizes from 100 to 150 μm are underrepresented in the 2D-C concentrations. The result of this method is that the particle concentrations quoted in this study are conservative relative to the total concentrations of particles present. Nevertheless, the maximum 2D-C concentrations calculated in this manner, 20–80 L−1, were high enough relative to minimum cloud top of −14° and −8°C, respectively, in these two frontal cloud situations and qualify as cases of “ice multiplication” (Hobbs 1969).

3. Description of the meteorological conditions under which the data were collected

Images of fragmented drops (Fig. 2d), were collected in portions of two weak, multilayered frontal systems on 18 and 20 January 2001 that lay offshore of the western Washington coastline. The low- and middle-level clouds were partitioned into several layers. The lowest layers would be classified as stratocumulus clouds and the higher layers that precipitated into the stratocumulus as altocumulus cloud layers because of their heights above sea level. Further, in some localized regions on 18 January 2001, the modestly turreted tops of the altocumulus would make them “castellanus” or “floccus” (convective) varieties. The general top of the highest layer of altocumulus clouds on 18 January 2001 was found at about 4 km MSL (or about 625 hPa) with higher, mounding tops and lower saddle regions. Therefore, cloud-top temperatures had a fairly significant range along the two flight legs flown, from −8° to −13°C. The ambient temperatures for these flight legs were −2.5° and −8.5°C. Figure 3 for 18 January shows several microstructural measurements made during several minutes at −8.5°C that were typical of the flight legs. At the highest flight leg, glaciated, mixed phase, and all liquid regions containing “embryo” drizzle drops (50–200 μm in diameter) and drizzle drops (200–500 μm in diameter) were encountered. Drops larger than drizzle sizes (>500 μm in diameter) were not encountered at subfreezing temperatures.

Liquid water contents (LWC) were moderate in the highest layer and very briefly reached almost 0.5 g m−3. Droplet concentrations were consistently less than 100 cm−3 in all cloud layers sampled with the exception of some stratus fractus clouds in the offshore-flowing boundary layer air near the coastline. The droplet concentrations in those clouds approached 200 cm−3.

The “threshold diameter” or “tail” of the FSSP-100 droplet spectrum (e.g., Hobbs and Rangno 1985)1 was consistently greater than 30-μm diameter and the effective radius was 12 μm in the layer clouds producing precipitation. These values are compatible with those values marking broad spectra found in precipitating clouds reported by Gerber et al. (1994) and Rosenfeld and Lensky (1998). Concentrations of droplets larger than 24 μm in the Hallett–Mossop (H–M) riming and splintering zone were consistently around 5–10 cm−3 with momentary peaks as high as 20 cm−3 (Fig. 4a).

The concentrations of ice particles on the two flight legs on 18 January 2001 were sometimes >50 L−1 in regions of a few hundred meters width. The ice particles consisted of sheaths, needles, frozen drops, ice fragments, and a few aggregates. Concentrations of drizzle drops in the low tens per liter were also encountered in localized regions of a few hundred meters width.

Along most of the two legs, the ice particle spectrum was relatively broad, suggesting a steady production of ice particles over a significant span of time (e.g., Fig. 5a). In other small regions, the transformation to ice from drizzle drops was just taking place via the freezing of drops, with columnar ice largely still absent (Fig. 5b). In still other small regions that contained ice, the small size of the columnar crystals, the lack of aggregates, along with the presence of liquid water, suggested that they were a region of ice that had formed quite recently region (e.g., Fig. 5c). In virtually all regions containing ice and at least some liquid water, images of fragmented drops were occasionally encountered in the CPI imagery. These frozen partial drops, however, constituted less than 10% of the frozen drops imaged (see Table). Some of the partial frozen drops imaged were ones that had collided with and stuck to a needle or sheath ice crystal and had fragmented (Fig. 2d last particle). Note that this last image shows a needle and a drop that apparently fragmented when colliding with it. Other partial frozen drop images collected were in isolation (e.g., as in Fig. 2d).

The clouds on 20 January 2001 were remarkably similar to those on 18 January 2001. The situation was again multilayered with the highest tops of an altocumulus layer separate from the boundary layer clouds lower and warmer than on 18 January 2001. Stepped ascents were made from the lowest cloud bases to highest cloud tops. On the highest leg the aircraft flew in clear air above the tops of this frontal band with the exception of a brief penetration of a very smooth appearing top at 3.3 km and −6°C. Although a heavy layer of altostratus clouds overlay the tops that the aircraft flew over, no ice particles were intercepted in the clear air. The only ice particles intercepted were tiny columns encountered in the brief intercept of the top. Figure 6 shows several minutes of microstructural data for this flight that characterizes the legs flown at −3°C.

The legs of interest on 20 January 2001 were the two reciprocal legs flown at −3°C, or about 400–600 m below cloud top. The base of this layer was at 2 km MSL and 0°C. The legs consumed 31 minutes of flight time, not counting a six minute loss of data due to a computer outage on the first leg. The microstructure encountered in these reciprocal legs at −3°C was similar to that on 18 January 2001. Droplet concentrations were again low in these clouds (<100 cm−3). The concentrations of droplets greater than 24 μm in diameter at −3°C were very low, peaking at only about 5 cm−3 (Fig. 5b). Since the clouds were likely contiguous, however, to at least −6°C, higher concentrations than this probably existed above the −3°C level.

Liquid water content was highly irregular, and on the first leg at −3°C, frequently reached as much as 0.4 g m−3. This irregularity of LWC at a constant flight level can be seen as reflecting locally thicker cloud below the flight level. Liquid water on the second pass at −3°C, though along the reciprocal, peaked at only 0.25 g m−3.

Alternating regions of moderately high concentrations of ice particles, 50–100 L−1, and isolated regions of drizzle drops in concentrations of tens per liter (e.g., Fig. 7), were encountered. The ice particles were largely needles and sheaths, compatible with the lowest cloud-top temperature.2 In some cases, the ice particles intercepted were not collocated with LWC, but fell out from the liquid top of the layer, where the ice was apparently forming, in a manner resembling altocumulus clouds with virga trails underneath them. The CPI imagery for these two flight legs showed occasional images of partial, frozen drops and ice fragments, though the majority of ice particles were columnar.

4. Results

Table 1 summarizes the results of an inventory of ice particles assembled from the CPI and categorized as shown in Fig. 2. These data are for approximately 3300 ice particles in selected periods determined by 2-DC imagery when ice appeared to be relatively newly formed, judged by their small size, and the continuing presence of liquid water, or where drizzle drops were beginning to freeze. Ice columns, which could be identified to smaller sizes (<50 μm in maximum dimension) than the other categories of ice particles, comprise the majority (about 55%) of the identifiable ice particles. This is compatible with cloud-top temperatures, mainly greater than −10°C, and the production of ice splinters through the H–M mechanism in which the two datasets were collected. Recognizable frozen drops larger than 50-μm diameter, either in isolation or on columnar ice particles, together made up about 12% of all ice particles in these newly formed ice regions.

Ice fragments, those particles that could not be assigned a habit, made up the remaining one-third of all ice particles seen in the CPI imagery. In the last column of Table 1 the percentage of the total that appeared to be most directly associated with fragmenting drops is shown, that is, those fragments having rounded portions that may have been from drop freezing events. These percentages are 3%, 6%, and 20%, respectively, of all ice particles in those limited regions of the three flight legs where ice had appeared to have formed very recently or was forming through the freezing of drizzle drops. The highest percentage was associated with the cloud system having the highest top temperature, about −6° to −8°C. Fragments having round portions therefore constituted from less than one to a few per liter in ice-forming regions. However, the contribution of fragmenting embryo drizzle and drizzle drops (50–500 μm in diameter) could have been significantly greater if the smaller unidentifiable fragments lacking rounded portions were also linked to freezing events.

5. Conclusions

The findings reported here suggest that ice particle concentrations in supercooled clouds containing drizzle drops (200–500 μm in diameter), and smaller “embryo” drizzle drops (50–200 μm in diameter) can be modestly enhanced (from a few percent to 20% in individual cases) due to the fragmentation of freezing drops in those regions where ice is first appearing. Images of partial frozen drops of the above size were found both as isolated particles or on sheaths or needle ice crystals and on some graupel particles and these images were accompanied by occasional fragments having rounded portions. However, fragments having rounded portions made up only a small subset of all the particles deemed ice fragments. The majority of fragments, about one-third of all ice particles in these new ice-forming regions, could not be identified with the available airborne technology such as might be done on the ground with a microscope.

The semiobjective nature of the classifications done in this study necessarily means that there is a substantial “error bar” around the above percentages. Nevertheless, it was encouraging that these tentative results for those fragments having rounded portions were in rough agreement with the laboratory results cited earlier that have suggested a very modest role for drop fragmentation in the production of secondary ice particles and also in prior airborne studies of ice fragments. It is suggested that this might be a further research avenue to pursue in the development of initial ice in clouds in view of the other recent reports of drop fragmentation upon freezing (Korolev et al. 2004; Rangno and Hobbs 2005).

Acknowledgments

The author is grateful to Mr. John Locatelli for his independent review of samples of the author’s crystal classifications. The data collection by the University of Washington CV-580 research aircraft during IMPROVE was made possible by the National Science Foundation under Grant ATM-9908446. Analysis of the data was supported by the NSC under Grant ATM-0242592 and by NASA under Grant NNG04GD64G.

REFERENCES

  • Brownscombe, J. L., and J. Hallett, 1967: Experimental and field studies of precipitation particles formed by the freezing of supercooled water. Quart. J. Roy. Meteor. Soc., 93 , 455473.

    • Search Google Scholar
    • Export Citation
  • Cannon, T. W., J. E. Dye, and V. Toutenhoofd, 1974: The mechanism of precipitation formation in northeastern Colorado cumulus. II. Sailplane measurements. J. Atmos. Sci., 31 , 21482151.

    • Search Google Scholar
    • Export Citation
  • Chisnell, R. F., and J. Latham, 1976: Ice particle multiplication in cumulus clouds. Quart. J. Roy. Meteor. Soc., 102 , 133156.

  • Choularton, T. W., J. Latham, and B. J. Mason, 1978: A possible mechanism of ice splinter production during riming. Nature, 274 , 791792.

    • Search Google Scholar
    • Export Citation
  • Choularton, T. W., D. J. Griggs, B. Y. Humood, and J. Latham, 1980: Laboratory studies of riming, and its relation to ice splinter production. Quart. J. Roy. Meteor. Soc., 106 , 367374.

    • Search Google Scholar
    • Export Citation
  • Coons, R. D., and R. Gunn, 1951: Relation of artificial cloud-modification to the production of precipitation. Compendium of Meteorology, T. F. Malone, Ed., Amer. Meteor. Soc., 235–241.

    • Search Google Scholar
    • Export Citation
  • Dye, J. E., and P. V. Hobbs, 1966: Effect of carbon dioxide on the shattering of freezing water drops. Nature, 209 , 464466.

  • Dye, J. E., and P. V. Hobbs, 1968: The influence of environmental parameters on the freezing and fragmentation of suspended water drops. J. Atmos. Sci., 25 , 8296.

    • Search Google Scholar
    • Export Citation
  • Fletcher, N. H., 1962: The Physics of Rainclouds. Cambridge University Press, 386 pp.

  • Gerber, H., B. G. Arends, and A. S. Ackerman, 1994: New microphysics sensor for aircraft use. Atmos. Res., 31 , 235252.

  • Goldsmith, P., J. Gloster, and C. Hume, 1976: The ice phase in clouds. Preprints, Int. Conf. on Cloud Physics, Boulder, CO, Amer. Meteor. Soc., 163–167.

  • Griggs, D. J., and T. W. Choularton, 1983: Freezing modes of riming droplets with application to ice splinter production. Quart. J. Roy. Meteor. Soc., 109 , 243253.

    • Search Google Scholar
    • Export Citation
  • Hallett, J., and S. C. Mossop, 1974: Production of secondary ice particles during the riming process. Nature, 249 , 2628.

  • Hobbs, P. V., 1969: Ice multiplication in clouds. J. Atmos. Sci., 26 , 315318.

  • Hobbs, P. V., and A. J. Alkezweeny, 1968: The fragmentation of freezing water droplets in free fall. J. Atmos. Sci., 25 , 881888.

  • Hobbs, P. V., and A. L. Rangno, 1985: Ice particle concentrations in clouds. J. Atmos. Sci., 42 , 25232549.

  • Johnson, D. A., and J. Hallett, 1968: Freezing and shattering of supercooled water drops. Quart. J. Roy. Meteor. Soc., 94 , 468482.

  • Knight, C. A., and N. C. Knight, 1974: Drop freezing in clouds. J. Atmos. Sci., 31 , 11741176.

  • Koenig, L. R., 1963: The glaciating behavior of small cumulonimbus clouds. J. Atmos. Sci., 20 , 2947.

  • Koenig, L. R., 1968: Some observations suggesting ice multiplication in the atmosphere. J. Atmos. Sci., 25 , 460463.

  • Kolomeychuk, R. J., D. C. McKay, and J. V. Iribarne, 1975: The fragmentation and electrification of freezing drops. J. Atmos. Sci., 32 , 974979.

    • Search Google Scholar
    • Export Citation
  • Korolev, A. V., M. P. Bailey, J. Hallett, and G. A. Isaac, 2004: Laboratory and in situ observation of deposition growth of frozen drops. J. Appl. Meteor., 43 , 612622.

    • Search Google Scholar
    • Export Citation
  • Kostinski, A. B., and A. R. Jameson, 1997: Fluctuation properties of precipitation. Part I: On deviations of single-size drop counts from the Poisson distribution. J. Atmos. Sci., 54 , 21742186.

    • Search Google Scholar
    • Export Citation
  • Lawson, R. P., and T. L. Jensen, 1998: Improved microphysical observations in mixed phase clouds. Preprints, Conf. on Cloud Physics, Everett, WA, Amer. Meteor. Soc., 451–454.

  • Ludlam, F. H., 1952: The production of showers by the growth of ice particles. Quart J. Roy. Meteor. Soc., 78 , 543553.

  • Ludlam, F. H., 1955: Artificial snowfall from mountain clouds. Tellus, 7 , 277290.

  • Mason, B. J., and J. Maybank, 1961: The fragmentation and electrification of freezing water drops. Quart. J. Roy. Meteor. Soc., 87 , 113114.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., 1978: The influence of drop size distribution on the production of secondary ice particles during graupel growth. Quart. J. Roy. Meteor. Soc., 104 , 323330.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., 1985: Secondary ice particle production during rime growth: The effect of drop size distribution and rimer velocity. Quart. J. Roy. Meteor. Soc., 111 , 11131124.

    • Search Google Scholar
    • Export Citation
  • Mossop, S. C., and J. Hallett, 1974: Ice crystal concentration in cumulus clouds: Influence of the drop spectrum. Science, 186 , 632634.

    • Search Google Scholar
    • Export Citation
  • Murgatroyd, R. J., and M. P. Garrod, 1960: Observations of precipitation elements in cumulus clouds. Quart. J. Roy. Meteor. Soc., 86 , 167175.

    • Search Google Scholar
    • Export Citation
  • Ono, A., 1972: Evidence on the nature of ice crystal multiplication processes in natural cloud. J. Rech. Atmos., 6 , 399408.

  • Phillips, V. T. J., A. M. Blyth, P. R. A. Brown, T. W. Choularton, and J. Latham, 2001: The glaciation of a cumulus cloud over New Mexico. Quart. J. Roy. Meteor. Soc., 127 , 15131534.

    • Search Google Scholar
    • Export Citation
  • Pitter, R. L., and H. R. Pruppacher, 1973: A wind tunnel investigation of freezing of small water drops falling at terminal velocity in air. Quart. J. Roy. Meteor. Soc., 99 , 540550.

    • Search Google Scholar
    • Export Citation
  • Rangno, A. L., and P. V. Hobbs, 1991: Ice particle concentrations and precipitation development in small polar maritime cumuliform clouds. Quart. J. Roy. Meteor. Soc., 117 , 207241.

    • Search Google Scholar
    • Export Citation
  • Rangno, A. L., and P. V. Hobbs, 2005: Microstructures and precipitation development in cumulus and small cumulonimbus clouds over the warm pool of the tropical Pacific Ocean. Quart. J. Roy. Meteor. Soc., 131 , 639673.

    • Search Google Scholar
    • Export Citation
  • Rosenfeld, D., and I. M. Lensky, 1998: Satellite-based insights into precipitation formation processes in continental and maritime convective clouds. Bull. Amer. Meteor. Soc., 79 , 24572476.

    • Search Google Scholar
    • Export Citation
  • Scott, B. C., and P. V. Hobbs, 1977: A theoretical study of the evolution of mixed-phase cumulus clouds. J. Atmos. Sci., 34 , 812826.

  • Stith, J. L., J. A. Haggerty, A. Heymsfield, and C. A. Grainger, 2004: Microphysical characteristics of tropical updrafts in clean conditions. J. Appl. Meteor., 43 , 779794.

    • Search Google Scholar
    • Export Citation
  • Stoelinga, M. T., and Coauthors, 2003: Improvement of microphysical parameterization through observational verification experiment. Bull. Amer. Meteor. Soc., 84 , 18071826.

    • Search Google Scholar
    • Export Citation
  • Takahashi, C., and A. Yamashita, 1970: Shattering of frozen water drops in a supercooled cloud. J. Meteor. Soc. Japan, 48 , 369372.

  • Takahashi, T., and N. Fukuta, 1988: Observations of the embryos of graupel. J. Atmos. Sci., 45 , 32883297.

  • Vaughan, H. C., 1954: The spontaneous freezing temperatures of melted snow and of small water drops. Bull. Amer. Meteor. Soc., 35 , 5255.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Examples of particle images from the Stratton Park Engineering Company Cloud Particle Imager at temperatures between 8° and 10°C suggesting that the particles are solid. (a) Small drops with pieces missing or that appear cracked. (b) Larger drizzle-sized drops that are substantially out of round suggesting deformities produced by freezing.

Citation: Journal of the Atmospheric Sciences 65, 4; 10.1175/2007JAS2295.1

Fig. 2.
Fig. 2.

Examples of the classification of ice particles made in this paper from the Stratton Park Engineering Company Cloud Particle Imager. (a) Unidentifiable amorphous fragments, (b) columnar (note first particle in row appears to be viewed on end), (c1), (c2) frozen drops, and (d) partial frozen drops and possible fragments that might have been due to shattering–fragmentation upon freezing. The confidence that these are due to shattering–fragmentation decreases from left to right. No particle without a rounded section is included in the inventory for those particles that are considered possibly due to fragmentation of freezing drops.

Citation: Journal of the Atmospheric Sciences 65, 4; 10.1175/2007JAS2295.1

Fig. 3.
Fig. 3.

Example of cloud microstructural data obtained in level flight on 18 Jan 2001 at −2.5°C and 708 hPa. The droplet concentrations originate with a FSSP-100. The liquid water contents from the integrated FSSP-100 spectrum (dark line) and a Gerber Particle Volume Monitor-100 (light line). The ice particle concentrations are from a 2D-C for those particles at least 100 μm in maximum dimension.

Citation: Journal of the Atmospheric Sciences 65, 4; 10.1175/2007JAS2295.1

Fig. 4.
Fig. 4.

Vertical aircraft soundings on (a) 18 and (b) 20 Jan 2001 also showing the concentrations of droplets greater than 24 μm in diameter. The vertical dashed lines represent the boundaries of the Hallett–Mossop riming and splintering temperature zone as determined by Mossop (1985).

Citation: Journal of the Atmospheric Sciences 65, 4; 10.1175/2007JAS2295.1

Fig. 5.
Fig. 5.

Examples of 2-DC imagery in which it was deduced that ice particles (a) were steadily forming in a quasi–steady state process, (b) are forming in situ due to the freezing of drizzle drops, and (c) are forming where columnar ice particles have recently formed. The length of the thin vertical lines separating particles is 800 μm.

Citation: Journal of the Atmospheric Sciences 65, 4; 10.1175/2007JAS2295.1

Fig. 6.
Fig. 6.

Example of cloud microstructural data obtained in level flight on 20 Jan 2001 at −3°C and 730 hPa. The origin of the data and symbols are as in Fig. 3.

Citation: Journal of the Atmospheric Sciences 65, 4; 10.1175/2007JAS2295.1

Fig. 7.
Fig. 7.

An example of the localized regions of liquid drizzle drops that were occasionally intercepted. Note splashes at the end of some rows confirming the liquid phase of at least some drops. The length of the thin vertical lines separating particles is 800 μm.

Citation: Journal of the Atmospheric Sciences 65, 4; 10.1175/2007JAS2295.1

Table 1.

Ice particle inventory from the Cloud Particle Imager for portions of three flight legs in shallow stratiform clouds.

Table 1.

1

The “threshold diameter” or, preferably, “tail of the droplet spectrum,” was calculated by Hobbs and Rangno (1985) when ice particle concentrations were less than 10 L−1, thereafter (e.g., Rangno and Hobbs 1991) only when ice particle concentrations were less than 1 L−1 to insure that the FSSP-100 spectra were not appreciably affected by ice particles.

2

The flight tracks were fixed along the ground. Therefore, the clouds flown through on the first leg, in view of southwesterly flow, would be to the northeast of the second leg. It was observed when the aircraft overflew the tops of the clouds in clear air at −6°C that slightly higher, mounding cloud tops estimated to have reached −8°C were to the north and northeast of the flight track and likely were the clouds flown under during the first leg.

Save