UBC Faculty Research and Publications

Pockets of Open Cells and Drizzle in Marine Stratocumulus Stevens, Bjorn; Vali, Gabor; Comstock, Kimberly; Woods, Robert; Van Zanten, Margreet C.; Austin, Philip H.; Bretherton, Christopher S.; Lenschow, Donald H. Jan 31, 2005

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


52383-Austin_AMS_2005_BAMS-86-1-51.pdf [ 353.59kB ]
JSON: 52383-1.0041809.json
JSON-LD: 52383-1.0041809-ld.json
RDF/XML (Pretty): 52383-1.0041809-rdf.xml
RDF/JSON: 52383-1.0041809-rdf.json
Turtle: 52383-1.0041809-turtle.txt
N-Triples: 52383-1.0041809-rdf-ntriples.txt
Original Record: 52383-1.0041809-source.json
Full Text

Full Text

POCKETS OF OPEN CELLS AND DRIZZLE IN MARINE STRATOCUMULUS BY  BJORN STEVENS, GABOR VALI, KIMBERLY COMSTOCK, ROBERT WOOD, MARGREET C. VAN ZANTEN, PHILIP H. AUSTIN, CHRISTOPHER S. BRETHERTON, AND DONALD H. LENSCHOW  The relationship between drizzle and cloud morphology, as manifest in transitions to spatially compact regions of open cellular convection, is studied using data collected from recent field studies over the northeast and southeast Pacific.  A  curious feature of the cloud climatology over regions of the globe where low-lying marine stratiform clouds predominate is the appearance of pockets of seemingly cloud-free air embedded in an otherwise homogeneous cloud field. The irregularities in the cloud field as seen in the visible satellite imagery in the left and upper panels of Fig.1 are an example of such features. Here two elongated regions of very low  AFFILIATIONS: STEVENS AND VAN ZANTEN—Department of Atmospheric Sciences, University of California, Los Angeles, Los Angeles, California; VALI—Department of Atmospheric Science, University of Wyoming, Laramie, Wyoming; COMSTOCK, WOOD, AND BRETHERTON—Department of Atmospheric Sciences, University of Washington, Seattle, Washington; VAN ZANTEN— Institute of Marine and Atmospheric Research Utrecht, Utrecht University, Utrecht, Netherlands; AUSTIN—Atmospheric Sciences Programme, University of British Columbia, Vancouver, British Columbia, Canada; LENSCHOW—National Center for Atmospheric Research, Boulder, Colorado CORRESPONDING AUTHOR: Bjorn Stevens, Department of Atmospheric and Oceanic Sciences, Box 951565, University of California, Los Angeles, Los Angeles, CA 90095-1564 E-mail: bstevens@atmos.ucla.edu doi:10.1175/BAMS-86-1-51 In final form 30 June 2004 ©2005 American Meteorological Society  AMERICAN METEOROLOGICAL SOCIETY  reflectivity stripe the southern portion of the cloud field, with more irregular pockets evident in the eastward extension of these low-reflectivity bands. Because factors that regulate the reflectivity of low clouds can critically affect the climate system as a whole (e.g., Randall et al. 1984; Ma et al. 1996), features such as these are not simply curious. To the extent they provide clues into processes that regulate cloudiness on larger scales, hence climate, they may be important as well. Upon closer examination such low reflectivity pockets or bands are rarely completely clear; instead they evince an underlying patterning reminiscent of open cellular convection (Agee 1984). The extent of this patterning varies among cases and, for any particular case, with time. In Fig. 1, hints of cellular patterning are most evident at earlier times, and in the westward extension of the bands. To highlight the particularly curious tendency of these spatially compact, cellular-patterned, low-reflectivity regions to be embedded in otherwise stratiform cloud fields, we hereafter refer to them as pockets of open cells (POCs). It remains to be seen whether such a distinction can be physically, rather than methodologically, justified. By looking at DTB, the difference between the 11- and 4-mm brightness temperature, POCs can also be detected in the nighttime satellite imagery. This is JANUARY 2005  |  51  cent East Pacific Investigation of Climate (EPIC) field study (Bretherton et al. 2004) a boundary between a region of open and closed cellular convection advected over the National Oceanic and Atmospheric Administration (NOAA) Research Vessel the Ronald H. Brown (see Fig. 2). Here the microphysical structure of the cloud fields is investigated with the help of an upward-pointing cloud radar (Moran et al. 1998), whose measurements of reflectivity at 35 GHz are contoured in time–height space at the bottom of Fig. 2. The analysis suggests that when the region of open cellular convection is overhead the radar reflectivity field FIG 1. Temporal sequence of GOES-10 images spanning 16.5 h over the northhas a more cellular struceast Pacific on 27 Jul 2001. The color images are differences between channels ture, with frequent periods 4 (3.9 m m) and 2 (11 m m) of GOES, while the grayscale image is cloud albedo as of large reflectivities (a measured by channel one radiances. The visible image has a pixel size of 1 km, in contrast to the 4-km pixel of the infrared imagery. The domain of the upper proxy for drizzle, e.g., Vali panel is shown by the orange dashed box in the left panel. et al. 1998) extending to the surface. In the region to the shown in the right and bottom panels of Fig. 1, which south (sampled by the ship after 0900 local time) radar show that these POCs are coherent, long lived (the reflectivities are considerably smaller and the cloud images span 16.5 h), and embedded in regions of rela- field as measured by GOES appears more homogetively small values of DTB. The longevity and coher- neous. Although this is difficult to establish from the ence of POCs has attracted the attention of satellite me- satellite imagery alone, in part because of the conteorologists, with observational examples of POC-like founding effects of the diurnal cycle, close examinafeatures dating to the beginning of the satellite record tion of the 1200 LT imagery is indicative of a more (e.g., Krueger and Fritz 1961; Davies and Garay 2003; closed-cellular (brighter centers) cloud pattern over the Garay et al. 2004). Low values of DTB can be accounted ship, a view that is also consistent with observer records for by a reduction in the cloud optical depth from val- and shipboard photography (not shown). While not ues of 15 to unity (which is not consistent with the early- definitive, the imagery supports the idea that regions morning visible imagery), or by changes in the effec- of open cellular convection are characterized by cloud tive radius from 6 to 12 mm for a cloud with a fixed fields more conducive to precipitation, and that the optical depth of 15 (e.g., Fig. 1 of Perez et al. 2000). boundaries between the different modes of convecThus, we conclude that the POCs are favored in re- tion are coherent over long periods of time. The argugions of the cloud field where the cloud droplets are ment that regions of precipitation tend to be preferlarger and thus have a greater propensity to form pre- entially associated with regions of more open cellular cipitation. Such a finding is consistent with previous convection is, in retrospect, consistent with in situ observations (e.g., Han et al. 1995; Gerber 1996), as well measurements and satellite images from earlier field as the strong correlation between measured drizzle and studies (e.g., Fig. 1 of Austin et al. 1995), as well as the tendency for such regions to be more favorable for the DTB found by van Zanten et al. (2004). Longevity and drizzle also seem to be characteristic formation of ship tracks (Scorer 1987). To further evaluate the relationship between POCs not only of POCs, but of broader regions of open cellular convection as well. For instance, during the re- and precipitation we analyze data from the second re52  |  JANUARY 2005  search flight (RF02) of the Dynamics and Chemistry of Marine Stratocumulus (DYCOM-II) field study (Stevens et al. 2003b), which serendipitously sampled across a POC. Figure 3 shows the morning visible satellite imagery and reflectivity from a downwardlooking airborne cloud radar. In the nighttime satellite imagery the “tilde”-shaped POC, which is isolated in the upper-left panel of Fig. 3 and is the focus of this analysis, appears to break away from the broader region of open cellular convection to the west, thus reinforcing the relationships discussed above. Its subsequent propagation is consistent with it being adFIG 2. (top) Channel 1 (0.6 µm) reflectance from the southeast Pacific from GOESvected by the mean bound8 on 19 Oct 2001. On the right is a large areal view at 0900 LT, on the left, zoomed ary layer wind, and it reimages of cloud features in the region directly over the ship are shown for 0600 mains coherent for at least and 1200 LT. (bottom) Radar reflectivity data is taken from an upward-pointing cloud radar operated from the NOAA RV Ron H. Brown, which was on station at 10 h. Similar to what was 20°S, 85°W as part of the EPIC experiment. In each satellite image the position found for the case study of the Ron H. Brown is indicated by the orange open-circle marker, and the apfrom EPIC, reflectivity data proximate trajectory of the cloud field, as estimated from surface wind meafrom the cloud radar show surements, is indicated by the orange line. that the POC is composed of mesoscale cells whose walls are loci of intensified convection, with locally higher derived by advecting the aircraft position following the cloud tops, and radar backscatter (i.e., drizzle) extend- procedure outlined in Stevens et al. (2003a). The reing to the sea surface. Cell interiors have locally lower gion of the POC is still evident in the tilde-shaped encloud tops and even regions of complete clearing. velope of near zero values of DTB,which is clearly biBecause the lidar, unlike the radar, is sensitive to even sected by both tracks. As in Fig. 1 the POCs in the trace amounts of cloud, the clearings in Fig. 3 are best vicinity of RF02 are embedded in a broader cloud field identified by regions where the lidar pulse penetrates characterized by relatively small values of DTB, or large to the surface. As was evident in the EPIC data, more droplets. Indeed, droplet concentrations during RF02 overcast regions in the satellite imagery are associated were among the lowest of all flights, averaging about with more uniform clouds with little or no precipita- 60 cm-3 (van Zanten et al. 2004). Data from both flight tion. This provides further evidence that regions of periods are shown as time series in the figure: the northopen cellular convection are coincident with precipi- ern, near-surface, flight leg was flown in a clockwise tating regions of the flow. direction and bisected the POC during the first 900 s of In situ measurements allow us to be more quanti- the time series (shown in upper panel); the southern tative. In Fig. 4 we show data from two 30-min legs flown leg was flown in a counterclockwise fashion 300–400 m near sunrise, first at 100 m and then at 1100 m. For above cloud top, crossing over the POC near the orientation, our estimate of the airmass sampled by middle of the leg. For both flight legs the time period the flight legs is overlaid on the DTB field as derived during which the aircraft bisected the POC is highfrom Geostationary Operational Environmental Sat- lighted, both on the flight track and by the red arrow ellites (GOES). Because the POC appears to translate within the time series panels. The liquid water content with the mean boundary layer wind this estimate is ql from the northern (surface) track and the reflectivity AMERICAN METEOROLOGICAL SOCIETY  JANUARY 2005  |  53  reflectivity tend to have locally elevated cloud tops. The cellular structure of the POC is manifest in the in situ near-surface measurements of the northern flight leg. Here the quasiperiodic regions of precipitation are associated with sharp increases in the total water mixing ratio, and reduced values of the temperature—cool but moist pools driven by the evaporation of precipitation in the subcloud layer (cf. Barnes and Garstang 1982; Jensen et al. 2000). Unlike the cold pools of deep precipitating convection that tend to be dry and cool and hence are characterized by large depressions of qe , moisture FIG 3. (top right) Channel 1 (0.6 µm) reflectance over the northeast Pacific from enhancement in the precipiGOES-10 at 0730 LT (1430 UTC) for 11 Jul 2002. (top left) Zoomed image of tating regions we sample reflectance field from boxed region in regional image; overlaid on this image is more than offsets the tema flight segment from RF02 that spans the time of the overpass and from which perature perturbations, that radar and lidar data is presented in top left panel. The zoomed image highlights a tilde-shaped POC boxed in the image. (bottom) Time–height radar reflectivities is, these are regions of filled, with cloud top height as estimated by downward-looking lidar shown by slightly elevated qe . The white line. Regions where lidar detects no cloud are shown by a lidar trace at the mechanism for generating surface. The time for which the satellite image is valid is indicated on the flight qe perturbations in the tracks. subcloud layer remains something of a puzzle, but from the southern (above cloud) track reinforce the we note that regions of elevated qe in association with idea that POCs delineate regions of locally enhanced precipitation rates commensurate with what we meaprecipitation. Based on locally tailored reflectivity–rain- sure are also evident in the analysis presented by Barnes rate relationships (an extensive discussion of which is and Garstang (1982). The variability in the thermodyprovided in van Zanten et al. 2004), and in situ data namic structure in the vicinity of the POCs is in marked taken along the surface leg, we estimate mean surface contrast to the flow far from the POC, where the precipitation rate from the radar to be on the order of boundary layer thermodynamic structure is more uni1 cm day-1 within the precipitating cell walls of the POC. form, and there is no evidence of surface precipitation. These are the largest values measured in stratocumu- Although it is difficult to discern the detailed dynamics lus during DYCOMS-II and more than twice the sur- of the circulations in the vicinity of the cells from these face evaporation rate. Cloud-base rain rates are two– data, the emergence of cells supports previous theoto threefold larger yet. Without replenishment, pre- retical work (Bjerknes 1938; Stevens et al. 1998) that cipitation rates of this magnitude would deplete the argues precipitation from initially stratiform clouds cloud of all its liquid water in tens of minutes. The per- drives a more diabatic, cumulus-like circulation, charsistence, and coherence, of these vigorously precipi- acterized by narrower regions of convective ascent and tating regions suggest that the open cells are relatively broader regions of stabilized descent. stable-flow configurations that organize to maintain POCs tend to occur within cloud layers with large the moisture supply to the precipitating cell boundaries. mean droplet radius and small droplet concentration, Circulations capable of such transport are consistent suggestive of a low-aerosol air mass. However, the spewith the observation that regions of enhanced radar cific location of POCs may also reflect other environ54  |  JANUARY 2005  mental factors. One possible factor is hinted at by the counterintuitive tendency for the boundary layer to be moister within the vicinity of the POC—in the absence of other processes drizzle would be expected to dry the boundary layer. Might the observed moistening reflect the tendency for the air directly above the POC (as measured by the above cloud track in the lower panels) to also be more humid? To the extent the moist layer sampled above the cloud in the vicinity of the POC actually impinges on the cloud layer one would expect a reduced rate of entrainment drying—in other words, a relative moistening of the FIG 4. (left) Contours of DTb from GOES-10 illustrating region of tilde-shaped POC, PBL—which would tend to with flight tracks valid at time of snapshot superimposed. (upper right) Time lower cloud base and favor series of (from top to bottom) water vapor mixing ratio (q) as measured by a precipitation formation. a probe, potential temperature (q q ), and liquid (drizzle) water (ql) for Lyman-a This may explain why the northern (surface) flight track. The time during which this flight track is over POC formed in the first the POC is indicated by red in left panel and by annotation in time series data. place and why it is more lo(lower right) Similar to upper-right panel, except for southern flight track, flown calized than the region of at 1100 m and with radar reflectivities substituting for drizzle measurements. large cloud droplet sizes seen in Fig. 4. Such an explanation is consistent with more global correlations evi- depression in free-tropospheric temperatures just dent among other DYCOMS-II flights: the average above the POC. Nonetheless, the presence of a disspecific humidity just above the boundary layer (as tinct character to the thermodynamical structure of the measured by all cloud-top penetrations) on the three flow directly above the POC does suggest that either flights with the least precipitation was 2.2 g kg-1 com- the POC is induced by such features, or that both the pared to a value of 4.6 g kg–1 on the flights with the most POC and the overlying thermodynamic structure were precipitation. Similar correlations have been noted in imprinted by some other process—perhaps sometime the past field studies (Yuter et al. 2000), and are also in the past. Unlike broader regions of open cellular convection, apparent in the EPIC data. That said, it is difficult to raise these arguments above the level of suggestion. for which boundary forcings (e.g., SSTs) or airmass Data from aircraft soundings show evidence of the differences may often explain the observed cloud moist layer extending to the cloud top in the region of morphology, POCs share a relatively similar large-scale the POC; however, these data also show evidence of environment with neighboring regions of more stratidirectional shear, which could serve as a basis for dis- form cloudiness. Although such a local feature could rupting the vertical continuity in the moisture field in principle be maintained by mesoscale circulations, it above cloud top. For this reason, and because the is difficult to envision why such circulations would sounding data are sparse, it is difficult to definitively maintain their integrity for such extended periods of establish that the features 300-400 m above the bound- time, unless they were tied to the development of the ary layer extend down to cloud top. In addition, such POC in the first place. A more likely, and intriguing, presumed moistening effects could in part be offset by explanation is that a given set of large-scale meteorolarger entrainment velocities in association with the logical conditions is able to support multiple cloud reAMERICAN METEOROLOGICAL SOCIETY  JANUARY 2005  |  55  gimes depending, for instance, upon the propensity of precipitation to form. In this respect our data provide empirical support for the idea that by modulating precipitation formation, perturbations in the atmospheric aerosol can affect patterns of cloudiness (e.g., Albrecht 1989; Rosenfield 2000; Lohmann and Lesins 2002). Although it may well turn out that the dynamics of POCs are not substantially different from broader regions of open cellular convection, their spatially compact nature make them well suited to comparisons with the surrounding, or background flow. As such they can provide a natural laboratory for investigating the interactions among cloud dynamics, precipitation, and properties of the atmospheric aerosol. In summary, data from two recent field experiments in the northeast and southeast Pacific stratocumulus regions suggest that coherent, long-lived regions of open cellular convection are coupled to the development of precipitation. These findings are consistent with theoretical work that suggests that the transition to more strongly precipitating flows should induce such a cloud transition. Future experimental work that targets such features may greatly advance our understanding of the relationship between precipitation and cloudiness, one of the key uncertainties in relating perturbations of the atmospheric aerosol to changes in climate. ACKNOWLEDGMENTS. This research was supported by the NSF through Grants ATM-0097053, ATM-0094956, ATM-0082384, and its support of the National Center for Atmospheric Research, as well through funding by NSERC/ CFCAS, ONR (through the EPSCoR program, NASA Grants NAGS5-10624, NAG 5-12559). The contributions of MVC were supported by the Stichting voor Fundamenteel Onderzoek der Materie (FOM), which is financially supported by the Nederlandse Organisatie voor Wetenschappelijk Onderzoek (NWO), and those of KC by an NDSEG fellowship. The authors wish to thank Christopher Fairall, Taneil Uttal, and Duane Hazen of NOAA ETL, and the staff and crew of the NOAA RHB for their assistance in collecting and interpreting EPIC data, as well as the scientists and staff of the Research Aviation Facility of NCAR in support of DYCOMS-II. Conversations and many insights collegially shared by Mike Garay, Melanie Wetzel, and Sandra Yuter are also gratefully acknowledged. Scott Loehrer and the Joint Office for Scientific Support (JOSS) are thanked for their help with the GOES imagery, as well as the reviewers for comments which considerably improved the presentation of this work. We would like to give special thanks to Bruce Albrecht, whose group it turns out has been working on a careful study of POC-like features (which they call rifts), for his careful and critical evaluation of our ideas; his attention to the details of our arguments has challenged and im-  56  |  JANUARY 2005  proved our understanding, substantially improving the final manuscript.  REFERENCES Agee, E. M., 1984: Observations from space and thermal convection: A historical perspective. Bull. Amer. Meteor. Soc., 65, 938–949. Albrecht, B. A., 1989: Aerosols, cloud microphysics and fractional cloudiness. Science, 245, 1227–1230 Austin, P., Y. Wang, R. Pincus, and V. Kujala, 1995: Precipitation in stratocumulus clouds: Observational and modeling results. J. Atmos. Sci., 52, 2329–2352. Barnes, G. M., and M. Garstang, 1982: Subcloud layer energetics of precipitating convection. Mon. Wea. Rev., 110, 102–117. Bjerknes, J., 1938: Saturated-adiabatic ascent of air through dry-adiabatically descending environment. Quart. J. Roy. Meteor. Soc., 64, 325–330. Bretherton, C. S., and Coauthors, 2004: The EPIC 2001 stratocumulus study. Bull. Amer. Meteor. Soc., 85, 967–977. Davies, R., and M. Garay, 2003: Actinoform clouds of the South Pacific: Self-organization in the marine boundary layer. Proc. Seventh Int. Conf. on Southern Hemisphere Meteorology and Oceanography, Wellington, New Zealand, Amer. Meteor. Soc., 136– 137. Garay, M. J., R. Davies, C. Averill, and J. A. Westphal, 2004: Actinoform clouds: Overlooked examples of cloud self-organization at the mesoscale. Bull. Amer. Meteor. Soc., 85, 1585–1594. Gerber, H., 1996: Microphysics of marine stratocumulus clouds with two drizzle modes. J. Atmos. Sci., 53, 1649–1662. Han, Q., R. Welch, J. Chou, W. Rossow, and A. White, 1995: Validation of satellite retrievals of cloud microphysics and liquid water path using observations from FIRE. J. Atmos. Sci., 52, 4183–4195. Jensen, J. B., S. Lee, P. B. Krummel, J. Katzfey, and D. Gogoasa, 2000: Precipitation in marine cumulus and stratocumulus: Part I: Thermodynamic and dynamic observations of closed cell circulations and cumulus bands. Atmos. Res., 54, 117–155. Krueger, A. F., and S. Fritz, 1961: Cellular cloud patterns revealed by TIROS I. Tellus, 13, 1–7. Lohmann, U., and G. Lesins, 2002: Stronger constraints on the anthropogenic indirect aerosol effect. Science, 298, 1012–1016. Ma, C.-C., C. R. Mechoso, A. W. Robertson, and A. Arakawa, 1996: Peruvian stratus clouds and the tropical Pacific circulation: A coupled oceanatmosphere GCM study. J. Climate, 9, 1635–1645.  Moran, K. P., B. E. Martner, M. J. Post, R. A. Kropfli, D. C. Welsh, and K. B. Widener, 1998: An unattended cloud-profiling radar for use in climate research. Bull. Amer. Meteor. Soc., 79, 443–455. Perez, J. C., F. Herrera, F. Rosa, A. Gonzales, M. Wetzel, R. Borys, and D. Lowenthal, 2000: Retrieval of marine stratus cloud droplet size from NOAA AVHRR night imagery. Remote Sens. Environ., 73, 31–45. Randall, D. A., J. A. Coakley Jr., C. W. Fairall, R. A. Kropfli, and D. H. Lenschow, 1984: Outlook for research on subtropical marine stratiform clouds. Bull. Amer. Meteor. Soc., 65, 1290–1301. Rosenfeld, D., 2000: Suppression of rain and snow by urban and industrial air pollution. Science, 297, 1793–1796. Scorer, R. S., 1987: Ship trails. Atmos. Environ., 21, 1417– 1425. Stevens, B., W. R. Cotton, G. Feingold, and C.-H. Moeng, 1998: Large-eddy simulations of strongly  AMERICAN METEOROLOGICAL SOCIETY  precipitating, shallow, stratocumulus-topped boundary layers. J. Atmos. Sci., 55, 3616–3638. ——, and Coauthors, 2003a: On entrainment in nocturnal marine stratocumulus. Quart. J. Roy. Meteor. Soc., 129, 3469–3492. ——, and Coauthors, 2003b: DYCOMS-II. Bull. Amer. Meteor. Soc., 84, 579–593. Vali, G., R. D. Kelly, J. French, S. Haimov, D. Leon, R. E. McIntosh, and A. Pazmany, 1998: Finescale structure and microphysics of coastal stratus. J. Atmos. Sci., 55, 3540–3564. van Zanten, M. C., B. Stevens, G. Vali, and D. H. Lenschow, 2004: Observations of drizzle in nocturnal marine stratocumulus. J. Atmos. Sci., 62, 88–106. Yuter, S. E., Y. L. Serra, and R. A. Houze Jr., 2000: The 1997 Pan American Climate Studies Tropical Eastern Pacific Process Study. Part II: Stratocumulus region. Bull. Amer. Meteor. Soc., 81, 483–490.  JANUARY 2005  |  57  


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items