• Baldauf, M., , A. Seifert, , J. Förstner, , D. Majewski, , M. Raschendorfer, , and T. Reinhardt, 2011: Operational convective-scale numerical weather prediction with the COSMO model: Description and sensitivities. Mon. Wea. Rev., 139, 38873905, doi:10.1175/MWR-D-10-05013.1.

    • Search Google Scholar
    • Export Citation
  • Bodas-Salcedo, A., , K. D. Williams, , P. R. Field, , and A. P. Lock, 2012: The surface downwelling solar radiation surplus over the Southern Ocean in the Met Office model: The role of midlatitude cyclone clouds. J. Climate, 25, 74677486, doi:10.1175/JCLI-D-11-00702.1.

    • Search Google Scholar
    • Export Citation
  • Bodas-Salcedo, A., and et al. , 2014: Origins of the solar radiation biases over the Southern Ocean in CFMIP2 models. J. Climate, 27, 4156, doi:10.1175/JCLI-D-13-00169.1.

    • Search Google Scholar
    • Export Citation
  • Bracegirdle, T., , and E. Kolstad, 2010: Climatology and variability of Southern Hemisphere marine cold-air outbreaks. Tellus, 62A, 202208, doi:10.1111/j.1600-0870.2009.00431.x.

    • Search Google Scholar
    • Export Citation
  • Bromwich, D. H., , J. F. Carrasco, , and C. R. Stearns, 1992: Satellite observations of katabatic-wind propagation for great distances across the Ross Ice Shelf. Mon. Wea. Rev., 120, 19401949, doi:10.1175/1520-0493(1992)120<1940:SOOKWP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brümmer, B., 1996: Boundary-layer modification in wintertime cold-air outbreaks from the Arctic sea ice. Bound.-Layer Meteor., 80, 109125, doi:10.1007/BF00119014.

    • Search Google Scholar
    • Export Citation
  • Brümmer, B., 1997: Boundary layer mass, water, and heat budgets in wintertime cold-air outbreaks from the Arctic sea ice. Mon. Wea. Rev., 125, 18241837, doi:10.1175/1520-0493(1997)125<1824:BLMWAH>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brümmer, B., 1999: Roll and cell convection in wintertime Arctic cold-air outbreaks. J. Atmos. Sci., 56, 26132636, doi:10.1175/1520-0469(1999)056<2613:RACCIW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Businger, S., , and R. J. Reed, 1989: Cyclogenesis in cold air masses. Wea. Forecasting, 4, 133156, doi:10.1175/1520-0434(1989)004<0133:CICAM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carleton, A. M., , and Y. Song, 1997: Synoptic climatology, and intrahemispheric associations, of cold air mesocyclones in the Australasian sector. J. Geophys. Res., 102, 13 87313 887, doi:10.1029/96JD03357.

    • Search Google Scholar
    • Export Citation
  • Coggins, J. H. J., , A. J. McDonald, , and B. Jolly, 2014: Synoptic climatology of the Ross Ice Shelf and Ross Sea region of Antarctica: k-means clustering and validation. Int. J. Climatol., 34, 23302348, doi:10.1002/joc.3842.

    • Search Google Scholar
    • Export Citation
  • Condron, A., , and I. A. Renfrew, 2013: The impact of polar mesoscale storms on northeast Atlantic Ocean circulation. Nat. Geosci., 6, 3437, doi:10.1038/ngeo1661.

    • Search Google Scholar
    • Export Citation
  • Condron, A., , G. R. Bigg, , and I. A. Renfrew, 2006: Polar mesoscale cyclones in the northeast Atlantic: Comparing climatologies from ERA-40 and satellite imagery. Mon. Wea. Rev., 134, 15181533, doi:10.1175/MWR3136.1.

    • Search Google Scholar
    • Export Citation
  • Cook, P. A., , and I. A. Renfrew, 2015: Aircraft-based observations of air-sea turbulent fluxes around the British Isles. Quart. J. Roy. Meteor. Soc., 141, 139152, doi:10.1002/qj.2345.

    • Search Google Scholar
    • Export Citation
  • Dee, D., and et al. , 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Doms, G., and et al. , 2011: A description of the nonhydrostatic regional COSMO model. Part II: Physical parameterization. Consortium for Small-Scale Modelling Tech. Rep. LM_F90 4.20, 154 pp. [Available online at http://www.cosmo-model.org/content/model/documentation/core/cosmoPhysParamtr.pdf.]

  • Fairall, C. W., , E. F. Bradley, , D. P. Rogers, , J. B. Edson, , and G. S. Young, 1996: Bulk parameterization of air-sea fluxes for tropical ocean-global atmosphere coupled-ocean atmosphere response experiment. J. Geophys. Res., 101, 37473764, doi:10.1029/95JC03205.

    • Search Google Scholar
    • Export Citation
  • Fairall, C. W., , E. F. Bradley, , J. E. Hare, , A. A. Grachev, , and J. B. Edson, 2003: Bulk parameterization of air–sea fluxes: Updates and verification for the COARE algorithm. J. Climate, 16, 571591, doi:10.1175/1520-0442(2003)016<0571:BPOASF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Forbes, G. S., , and W. D. Lottes, 1985: Classification of mesoscale vortices in polar airstreams and the influence of the large-scale environment on their evolutions. Tellus, 37A, 132155, doi:10.1111/j.1600-0870.1985.tb00276.x.

    • Search Google Scholar
    • Export Citation
  • Green, J. S. A., 1960: A problem in baroclinic stability. Quart. J. Roy. Meteor. Soc., 86, 237251, doi:10.1002/qj.49708636813.

  • Green, J. S. A., 1979: Topics in dynamical meteorology: 8. Trough-ridge systems as slantwise convection. Weather, 34, 210, doi:10.1002/j.1477-8696.1979.tb03366.x.

    • Search Google Scholar
    • Export Citation
  • Grossman, R. L., , and A. K. Betts, 1990: Air–sea interaction during an extreme cold air outbreak from the eastern coast of the United States. Mon. Wea. Rev., 118, 324342, doi:10.1175/1520-0493(1990)118<0324:AIDAEC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gryschka, M., , J. Fricke, , and S. Raasch, 2014: On the impact of forced roll convection on vertical turbulent transport in cold air outbreaks. J. Geophys. Res. Atmos., 119, 12 51312 532, doi:10.1002/2014JD022160.

    • Search Google Scholar
    • Export Citation
  • Hartmann, J., , C. Kottmeier, , and S. Raasch, 1997: Roll vortices and boundary-layer development during a cold air outbreak. Bound.-Layer Meteor., 84, 4565, doi:10.1023/A:1000392931768.

    • Search Google Scholar
    • Export Citation
  • Holton, J. R., , and G. J. Hakim, 2012: An Introduction to Dynamic Meteorology. Academic Press, 532 pp.

  • Hoskins, B. J., , M. E. McIntyre, , and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111, 877946, doi:10.1002/qj.49711147002.

    • Search Google Scholar
    • Export Citation
  • Irving, D., , I. Simmonds, , and K. Keay, 2010: Mesoscale cyclone activity over the ice-free Southern Ocean: 1999–2008. J. Climate, 23, 54045420, doi:10.1175/2010JCLI3628.1.

    • Search Google Scholar
    • Export Citation
  • Isachsen, P. E., , M. Drivdal, , S. Eastwood, , Y. Gusdal, , G. Noer, , and Ø. Saetra, 2013: Observations of the ocean response to cold air outbreaks and polar lows over the Nordic Seas. Geophys. Res. Lett., 40, 36673671, doi:10.1002/grl.50705.

    • Search Google Scholar
    • Export Citation
  • Iwasaki, T., , T. Shoji, , Y. Kanno, , M. Sawada, , M. Ujiie, , and K. Takaya, 2014: Isentropic analysis of polar cold airmass streams in the Northern Hemispheric winter. J. Atmos. Sci., 71, 22302243, doi:10.1175/JAS-D-13-058.1.

    • Search Google Scholar
    • Export Citation
  • Jensen, T. G., , T. J. Campbell, , R. A. Allard, , R. J. Small, , and T. A. Smith, 2011: Turbulent heat fluxes during an intense cold-air outbreak over the Kuroshio Extension Region: Results from a high-resolution coupled atmosphere–ocean model. Ocean Dyn., 61, 657674, doi:10.1007/s10236-011-0380-0.

    • Search Google Scholar
    • Export Citation
  • Kolstad, E. W., 2011: A global climatology of favourable conditions for polar lows. Quart. J. Roy. Meteor. Soc., 137, 17491761, doi:10.1002/qj.888.

    • Search Google Scholar
    • Export Citation
  • Kolstad, E. W., , and T. J. Bracegirdle, 2008: Marine cold-air outbreaks in the future: An assessment of IPCC AR4 model results for the Northern Hemisphere. Climate Dyn., 30, 871885, doi:10.1007/s00382-007-0331-0.

    • Search Google Scholar
    • Export Citation
  • Liu, Z., , and D. H. Bromwich, 1997: Dynamics of the katabatic wind confluence zone near Siple Coast, West Antarctica. J. Appl. Meteor., 36, 97118, doi:10.1175/1520-0450(1997)036<0097:DOTKWC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Louis, J., 1979: A parametric model of vertical eddy fluxes in the atmosphere. Bound.-Layer Meteor., 17, 187202, doi:10.1007/BF00117978.

    • Search Google Scholar
    • Export Citation
  • Lüpkes, C., , and K. H. Schlünzen, 1996: Modelling the Arctic convective boundary-layer with different turbulence parameterizations. Bound.-Layer Meteor., 79, 107130, doi:10.1007/BF00120077.

    • Search Google Scholar
    • Export Citation
  • Nigro, M. A., , and J. J. Cassano, 2014: Identification of surface wind patterns over the Ross Ice Shelf, Antarctica, using self-organizing maps. Mon. Wea. Rev., 142, 23612378, doi:10.1175/MWR-D-13-00382.1.

    • Search Google Scholar
    • Export Citation
  • O’Connor, W. P., , D. H. Bromwich, , and J. F. Carrasco, 1994: Cyclonically forced barrier winds along the Transantarctic Mountains near Ross Island. Mon. Wea. Rev., 122, 137150, doi:10.1175/1520-0493(1994)122<0137:CFBWAT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Pagowski, M., , and G. W. K. Moore, 2001: A numerical study of an extreme cold-air outbreak over the Labrador Sea: Sea ice, air–sea interaction, and development of polar lows. Mon. Wea. Rev., 129, 4772, doi:10.1175/1520-0493(2001)129<0047:ANSOAE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Papritz, L., , S. Pfahl, , H. Sodemann, , and H. Wernli, 2015: A climatology of cold air outbreaks and their impact on air–sea heat fluxes in the high-latitude South Pacific. J. Climate, 28, 342364, doi:10.1175/JCLI-D-14-00482.1.

    • Search Google Scholar
    • Export Citation
  • Parish, T. R., , and J. J. Cassano, 2003a: Diagnosis of the katabatic wind influence on the wintertime Antarctic surface wind field from numerical simulations. Mon. Wea. Rev., 131, 11281139, doi:10.1175/1520-0493(2003)131<1128:DOTKWI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Parish, T. R., , and J. J. Cassano, 2003b: The role of katabatic winds on the Antarctic surface wind regime. Mon. Wea. Rev., 131, 317333, doi:10.1175/1520-0493(2003)131<0317:TROKWO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Parish, T. R., , J. J. Cassano, , and M. W. Seefeldt, 2006: Characteristics of the Ross Ice Shelf air stream as depicted in Antarctic Mesoscale Prediction System simulations. J. Geophys. Res., 111, D12109, doi:10.1029/2005JD006185.

    • Search Google Scholar
    • Export Citation
  • Rasmussen, E., , and J. Turner, 2003: Polar Lows: Mesoscale Weather Systems in the Polar Regions. Cambridge University Press, 612 pp.

  • Renfrew, I. A., , and G. W. K. Moore, 1999: An extreme cold-air outbreak over the Labrador Sea: Roll vortices and air–sea interaction. Mon. Wea. Rev., 127, 23792394, doi:10.1175/1520-0493(1999)127<2379:AECAOO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schröder, D., , G. Heinemann, , and S. Willmes, 2011: The impact of a thermodynamic sea-ice module in the COSMO numerical weather prediction model on simulations for the Laptev Sea, Siberian Arctic. Polar Res., 30, 6334, doi:10.3402/polar.v30i0.6334.

    • Search Google Scholar
    • Export Citation
  • Shoji, T., , Y. Kanno, , T. Iwasaki, , and K. Takaya, 2014: An isentropic analysis of the temporal evolution of East Asian cold air outbreaks. J. Climate, 27, 93379348, doi:10.1175/JCLI-D-14-00307.1.

    • Search Google Scholar
    • Export Citation
  • Škerlak, B., , M. Sprenger, , and H. Wernli, 2014: A global climatology of stratosphere–troposphere exchange using the ERA-Interim data set from 1979 to 2011. Atmos. Chem. Phys., 14, 913937, doi:10.5194/acp-14-913-2014.

    • Search Google Scholar
    • Export Citation
  • Skyllingstad, E. D., , and J. B. Edson, 2009: Large-eddy simulation of moist convection during a cold air outbreak over the Gulf Stream. J. Atmos. Sci., 66, 12741293, doi:10.1175/2008JAS2755.1.

    • Search Google Scholar
    • Export Citation
  • Steppeler, J., , G. Doms, , U. Schättler, , H. W. Bitzer, , A. Gassmann, , U. Damrath, , and G. Gregoric, 2003: Meso-gamma scale forecasts using the nonhydrostatic model LM. Meteor. Atmos. Phys., 82, 7596, doi:10.1007/s00703-001-0592-9.

    • Search Google Scholar
    • Export Citation
  • Talley, L. D., 2008: Freshwater transport estimates and the global overturning circulation: Shallow, deep and throughflow components. Prog. Oceanogr., 78, 257303, doi:10.1016/j.pocean.2008.05.001.

    • Search Google Scholar
    • Export Citation
  • Terpstra, A., , T. Spengler, , and R. W. Moore, 2015: Idealised simulations of polar low development in an Arctic moist-baroclinic environment. Quart. J. Roy. Meteor. Soc., 141, 19871996, doi:10.1002/qj.2507.

    • Search Google Scholar
    • Export Citation
  • Tiedtke, M., 1989: A comprehensive mass flux scheme for cumulus parameterization in large-scale models. Mon. Wea. Rev., 117, 17791800, doi:10.1175/1520-0493(1989)117<1779:ACMFSF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Vihma, T., , and B. Brümmer, 2002: Observations and modelling of the on-ice and off-ice air flow over the Northern Baltic Sea. Bound.-Layer Meteor., 103, 127, doi:10.1023/A:1014566530774.

    • Search Google Scholar
    • Export Citation
  • Wacker, U., , K. V. Jayaraman Potty, , C. Lüpkes, , J. Hartmann, , and M. Raschendorfer, 2005: A case study on a polar cold air outbreak over Fram Strait using a mesoscale weather prediction model. Bound.-Layer Meteor., 117, 301336, doi:10.1007/s10546-005-2189-1.

    • Search Google Scholar
    • Export Citation
  • Williams, K. D., and et al. , 2013: The Transpose-AMIP II experiment and its application to the understanding of Southern Ocean cloud biases in climate models. J. Climate, 26, 32583274, doi:10.1175/JCLI-D-12-00429.1.

    • Search Google Scholar
    • Export Citation
  • Yuan, X., , J. Patoux, , and C. Li, 2009: Satellite-based midlatitude cyclone statistics over the Southern Ocean: 2. Tracks and surface fluxes. J. Geophys. Res., 114, D04106, doi:10.1029/2008JD010874.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Infrared (channel 4) satellite images from the NOAA AVHRR instrument valid at (left image) 0009 UTC and (right image) 0137 UTC 26 Jun 2010, with only the earlier image shown where they overlap. A synoptic-scale cyclone in the ABS (green), CAO tongues (blue), and mesocyclones (red) are indicated.

  • View in gallery

    Location map of the study domain and model topography (shading). In addition, the sea ice edge (black) and the 280-K isotherm of SST (blue) are shown. The outer boundary of the interior Ross Sea is indicated by the black lines. The latitude circles are shown at 60° and 80°S and the longitude circles every 20°. The upper-right inset indicates location and orientation of the model domain, whereby the depicted region is bounded by the 35°S latitude circle.

  • View in gallery

    (a)–(d) Upper-level PV (shading) and wind speed (black contours) for 40 m s−1 and higher values, in intervals of 10 m s−1, on 310 K daily at 0000 UTC from 24 to 27 Jun 2010. Data are from the ERA-Interim (Dee et al. 2011).

  • View in gallery

    Overview on the synoptic evolution of the CAO from the COSMO simulation at (a),(b) 1200 UTC 24 Jun; (c),(d) 0000 UTC 25 Jun; (e),(f) 1200 UTC 25 Jun; and (g),(h) 0000 UTC 26 Jun 2010. (left) Upper-level PV on 310 K (shading) and geopotential height at 500 hPa (blue contours, in intervals of 100 m). (right) Cold airmass (shading), as well as SLP (gray contours, in intervals of 5 hPa), (blue contours, from 0 K in intervals of 4 K), and the sea ice edge (black, thick line). Note that SLP contours are not drawn over land. The discussed synoptic and mesoscale cyclones are labeled. In (b) the subregion shown in Fig. 6 is outlined by the black box.

  • View in gallery

    As in Fig. 4, but at (a),(b) 1200 UTC 26 Jun; (c),(d) 0000 UTC 27 Jun; and (e),(f) 1200 UTC 27 Jun 2010. In (b) the black box outlines the region shown in Figs. 8 and 1113, and in (f) the location of the cross section of Fig. 10 is shown. In addition, the box, over which the cold airmass budget is calculated, is outlined in black. The dashed and dotted lines denote the influx boundaries from the south Indian Ocean (A) and the interior RS (B), respectively.

  • View in gallery

    (a) Cold airmass and (b) the magnitude of its flux at 1200 UTC 24 Jun 2010. In addition, in (b) also the horizontal cold airmass-flux vectors and geopotential height at 500 hPa in intervals of 50 m (blue) are shown. In both figures SLP contours—omitted over topography—in intervals of 5 hPa are depicted in gray.

  • View in gallery

    Evolution of the cold airmass fluxes in the box outlined in Fig. 5f. (a) The total cold airmass tendency within the box (black), the horizontal cold airmass flux across the lateral boundaries (blue, solid), as well as the contributions of the fluxes from the south Indian Ocean (A, blue dashed) and the interior RS (B, blue dotted). In addition, the residual of the cold airmass budget within the box is shown in gray. (b) The diabatic mass flux across the upper-bounding isentropic surface of the cold air mass. Shown are the total diabatic mass flux (solid) and individual contributions from latent heating (dashed), radiation (dotted), and turbulence (dashed–dotted). In both panels the time axis is relative to the initialization time of the simulation; that is, time zero corresponds to 0000 UTC 24 Jun 2010. All tendencies are expressed in terms of the actual mass.

  • View in gallery

    (a) Cold airmass and (b)–(d) diabatic mass flux across the upper-bounding isentropic surface of the cold air mass for (b) all diabatic processes, (c) latent heating, and (d) radiation for a subregion of the CAO centered on M3 (cf. Fig. 5b) and valid at 1200 UTC 26 Jun. Gray contours show SLP in intervals of 5 hPa. In addition, the 100-, 300-, and 500-hPa contours of the cold air mass are drawn as a reference (thin black). The thick black contour delineates the sea ice edge. Furthermore, in (a) the location of the cross section shown in Fig. 9 is drawn.

  • View in gallery

    (a) Specific cloud water plus ice content at 1200 UTC 26 Jun 2010 along a cross section across the warm sector of M3 (see Fig. 8a). (b) Latent and (c) radiative heating rates. Potential temperature (black) is shown in intervals of 3 K with the 280- and 310-K isentropes highlighted in bold.

  • View in gallery

    Vertical cross section across the CAO air mass upstream of M3 valid at 1200 UTC 27 Jun 2010 (see Fig. 5f). Shown are PV (shading) and potential temperature (black contours, in intervals of 3 K), with the 280- and 310-K isentropes highlighted in bold.

  • View in gallery

    Changes in (a),(c) SLP and (b),(d) cold air mass at 1200 UTC 26 Jun, centered on mesocyclone M3 (cf. Fig. 5b), in sensitivity experiments with switched off surface fluxes of sensible (NOSHFL) and latent (NOLHFL) heat with respect to the reference run (REF). In addition, in (left) SLP in the respective sensitivity experiment (red) and in REF (gray) are shown in intervals of 5 hPa. Additionally SLP in intervals of 5 hPa (gray) and the 100-, 300-, and 500-hPa contours of the cold air mass (black) from the reference simulation (REF) are shown in (right). The thick black contour delineates the sea ice edge.

  • View in gallery

    As in Figs. 11b and 11d, but for (a),(c) the percentage change in vertically integrated water vapor relative to REF and (b),(d) the absolute change in vertically integrated cloud water and ice. Vertical integrals are taken over the entire column.

  • View in gallery

    Impact of sensible heat fluxes (a) on precipitation in the presence of latent heat fluxes (NOSHFL − REF) and (b) on cold air mass in the absence of latent heat fluxes (NOLHFL − NOHFL). The fields are shown centered on M3 (cf. Fig. 5b) at 1200 UTC 26 Jun. Gray and black contours are as in Fig. 11.

  • View in gallery

    (a) Lagrangian schematic of the relevant physical processes during CAO stages 2–4, whereby the horizontal axis represents time and the vertical axis is height. Shown are the height of the mixed layer (blue), the 280-K isentrope (black), clouds (gray), surface sensible and latent heat fluxes (red arrows), radiative cooling (blue arrow), and convective and turbulent mixing (circular arrows). Latent heat release associated with the mesocyclones essentially contributes to the erosion of the cold air mass, indicated by the large red arrow. The stages of the CAO development are discussed in detail in the text. (b) Schematic evolution of (green) in the CAO air mass.

  • View in gallery

    Cold air mass (shading), SLP (gray contours, in intervals of 5 hPa), and the sea ice edge (black, thick line) at 1200 UTC 26 Jun 2010 for simulations at a horizontal resolution of (a) 0.05° × 0.05° and (b) 1.0° × 1.0°.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 52 52 10
PDF Downloads 409 409 10

Importance of Latent Heating in Mesocyclones for the Decay of Cold Air Outbreaks: A Numerical Process Study from the Pacific Sector of the Southern Ocean

View More View Less
  • 1 Institute for Atmospheric and Climate Science, and Center for Climate Systems Modeling, ETH Zürich, Zürich, Switzerland
  • | 2 Institute for Atmospheric and Climate Science, ETH Zürich, Zürich, Switzerland
© Get Permissions
Full access

Abstract

In this study the dynamical mechanisms shaping the evolution of a marine cold air outbreak (CAO) that occurred over the Ross, Amundsen, and Bellingshausen Seas in June 2010 are investigated in an isentropic framework. The drainage of cold air from West Antarctica into the interior Ross Sea, its subsequent export, and the formation of a dome of cold air off the sea ice edge are shown to be intimately linked to a lower-tropospheric cyclone, as well as the cyclonic breaking of an upper-level potential vorticity trough. The dome formation is accompanied by an extreme deepening of the boundary layer, whose top reaches to the height of the low-lying tropopause within the trough, potentially allowing for deep stratosphere–troposphere exchange. A crucial finding of this study is that the decay of the CAO is essentially driven by the circulation associated with a train of mesocyclones and the release of latent heat in their warm sectors. Sensitivity experiments with switched off fluxes of sensible and latent heat reveal that the erosion of the CAO air mass depends critically on the moistening by latent heat fluxes, whereby the synergistic effects of sensible heat fluxes and moist processes amplify the erosion. Within the CAO air mass, the erosion is inhibited by cloud-top radiative cooling and the dissolution of clouds by the entrainment of dryer air. These findings potentially have implications for the representation of CAOs in coarse-resolution climate models.

Corresponding author address: Lukas Papritz, Geophysical Institute, University of Bergen, Allegaten 70, 5007 Bergen, Norway. E-mail: lukas.papritz@uib.no

Current affiliation: Geophysical Institute, University of Bergen, Bergen, Norway.

Abstract

In this study the dynamical mechanisms shaping the evolution of a marine cold air outbreak (CAO) that occurred over the Ross, Amundsen, and Bellingshausen Seas in June 2010 are investigated in an isentropic framework. The drainage of cold air from West Antarctica into the interior Ross Sea, its subsequent export, and the formation of a dome of cold air off the sea ice edge are shown to be intimately linked to a lower-tropospheric cyclone, as well as the cyclonic breaking of an upper-level potential vorticity trough. The dome formation is accompanied by an extreme deepening of the boundary layer, whose top reaches to the height of the low-lying tropopause within the trough, potentially allowing for deep stratosphere–troposphere exchange. A crucial finding of this study is that the decay of the CAO is essentially driven by the circulation associated with a train of mesocyclones and the release of latent heat in their warm sectors. Sensitivity experiments with switched off fluxes of sensible and latent heat reveal that the erosion of the CAO air mass depends critically on the moistening by latent heat fluxes, whereby the synergistic effects of sensible heat fluxes and moist processes amplify the erosion. Within the CAO air mass, the erosion is inhibited by cloud-top radiative cooling and the dissolution of clouds by the entrainment of dryer air. These findings potentially have implications for the representation of CAOs in coarse-resolution climate models.

Corresponding author address: Lukas Papritz, Geophysical Institute, University of Bergen, Allegaten 70, 5007 Bergen, Norway. E-mail: lukas.papritz@uib.no

Current affiliation: Geophysical Institute, University of Bergen, Bergen, Norway.

1. Introduction

The advection of polar air masses from the eastern boundaries of the continents across warm ocean currents or their flow off the sea ice edge over the comparatively warm open ocean in the polar regions is accompanied by enormous air–sea temperature contrasts leading to intense upward air–sea heat fluxes. These fluxes can be in excess of 1000 W m−2 (e.g., Grossman and Betts 1990). Consequently, the incidence of such marine cold air outbreaks (CAOs) has a massive influence on the spatial and temporal variability of winter mean air–sea heat fluxes. This applies in particular to the Pacific sector of the Southern Ocean, where throughout winter CAO conditions occur during more than 20% of the time and CAOs account for almost two-thirds of the upward sensible and latent heat fluxes (Papritz et al. 2015). The intense fluxes during a CAO event induce a strong cooling of the oceanic mixed layer (Isachsen et al. 2013) and contribute substantially to the transformation of oceanic water masses (e.g., Jensen et al. 2011), with associated impacts on the oceanic meridional overturning circulation and the global energy balance (e.g., Talley 2008). Understanding the dynamical mechanisms that govern the synoptic evolution of CAOs is thus key to constraining their important role in shaping the mean climate and climate variability.

Naturally intense air–sea heat fluxes have a strong influence on the dynamical evolution of CAOs themselves. The direct input of heat by sensible heat fluxes and the latent heat released during cloud formation, owing to the moistening of the convective boundary layer by latent heat fluxes, warm the CAO air mass and induce a rapid deepening of the boundary layer (Grossman and Betts 1990; Brümmer 1996; Renfrew and Moore 1999). Along with the deepening of the boundary layer, the cloud base is lowered and cloud-top height increases. These boundary layer processes in incipient marine CAOs are well documented based on data from observational field campaigns over the Gulf Stream (Grossman and Betts 1990), Fram Strait (Brümmer 1996, 1997; Hartmann et al. 1997; Brümmer 1999), and Labrador Sea (Renfrew and Moore 1999). Numerical studies of CAOs so far have been mostly aimed at understanding the relative importance of processes determining the vertical structure of the boundary layer in off-ice flows and the organization of convection utilizing two-dimensional models (e.g., Lüpkes and Schlünzen 1996; Vihma and Brümmer 2002), as well as large-eddy simulations (e.g., Skyllingstad and Edson 2009; Gryschka et al. 2014).

The characterization of the synoptic evolution of marine CAOs has received considerably less attention and is essentially limited to case studies of incipient CAOs in the Labrador Sea (Pagowski and Moore 2001) and over Fram Strait (Wacker et al. 2005). The authors of these studies noted the importance of the equatorward flow on the rearward side of extratropical cyclones, which draws cold air masses from polar latitudes, for the formation of CAOs. A climatological assessment of this linkage of cyclones and CAOs has been undertaken by Papritz et al. (2015) for the Ross Sea (RS) and the Amundsen and Bellingshausen Seas (ABS). Using feature-based methods for the identification of CAOs and cyclones, they found that about 80% of CAOs are induced by the flow associated with the passage of synoptic-scale cyclones. While this kinematic relationship is rather apparent, also a linkage to the upper-level flow can be expected. Specifically, upper-level potential vorticity (PV) troughs are associated with a doming of the isentropic surfaces in the troposphere (Hoskins et al. 1985), and therefore anomalously cold air is found below the troughs. In this case study we argue for a close relationship between the formation of upper-level PV troughs and incipient cold air outbreaks in the RS, favored by the local topography, in particular the channeling of the flow associated with the Transantarctic Mountains.

To the authors’ knowledge, as yet no study has investigated the dynamical processes leading to the decay of CAOs, neither observationally nor in numerical simulations. However, a detailed understanding of the physical processes promoting the decay of CAOs is, in addition to being of interest in its own right, highly relevant to eliminate biases in the representation of CAOs in climate models. For instance, Kolstad and Bracegirdle (2008) found a cold bias in marine CAO air masses in many climate models, which is likely caused by a too-slow decay of CAOs. Furthermore, Southern Hemisphere biases in surface downwelling radiation, existing in many climate models, have been largely attributed to errors in the representation of cloud cover in the cold sector of cyclones (Bodas-Salcedo et al. 2012; Williams et al. 2013; Bodas-Salcedo et al. 2014). To overcome such biases, it is necessary to not only thoroughly understand the turbulent and cloud microphysical processes occurring within CAO air masses but also the dynamical processes shaping their larger-scale evolution. In this spirit, it is one of the major goals of this paper to shed light on the dynamical mechanisms promoting the decay of CAOs.

One of the salient features accompanying CAOs is the development of mesocyclones (Businger and Reed 1989; Rasmussen and Turner 2003). The occurrence of mesocyclones in marine CAOs is favored by the enhanced baroclinic growth rate at the CAO’s outer edge as a result of the strong temperature contrast and intense air–sea heat fluxes, reducing static stability and providing latent energy. Mesocyclones frequently cluster as a series of individual vortices, at times organized in a merry-go-round fashion around a parent synoptic cyclone (Forbes and Lottes 1985; Businger and Reed 1989). In a satellite-based census of mesocyclones in the Pacific sector of the Southern Ocean, Carleton and Song (1997) showed that mesocyclone outbreaks occur when there is a northward transport of cold, polar air across the sea ice edge, typically on the rearward side of synoptic cyclones dragging along the Antarctic coastline. Furthermore, scatterometer-based climatologies of cyclones indicate that more than 50% of the cyclonic activity over the Southern Ocean is mesoscale (Yuan et al. 2009), whereby the highest mesocyclone frequency is found in the Pacific sector (Irving et al. 2010). This is also the region where CAO frequency is maximum (Bracegirdle and Kolstad 2010; Papritz et al. 2015) and the large-scale conditions for the genesis of polar lows—that is, mesocyclones with at least gale force near-surface winds—are the most favorable in the Southern Hemisphere (Kolstad 2011).

The prevalence of mesocyclones in cold air masses suggests that they may play an important role in the evolution of the CAOs. Specifically, the cyclonic circulation associated with mesocyclones can induce a splitting of the CAO air mass into several tongues (Papritz et al. 2015). Furthermore, the release of latent heat contributes substantially to the mesocyclones’ intensification (e.g., Rasmussen and Turner 2003; Terpstra et al. 2015). Such latent heating also warms the CAO air mass. Therefore, the question arises if and to what extent the mixing with warmer air masses and diabatic processes associated with these prevalent mesocyclones ramp up the decay of CAOs.

In this numerical process study we investigate an intense CAO in the RS and ABS that developed on late 24 June 2010 under the influence of a quasi-stationary synoptic cyclone located off the Antarctic coast of Marie Byrd Land and lived until 27 June 2010. Concomitant with the CAO, a train of mesocyclones developed, resulting in the splitting of the CAO air mass into several tongues. These tongues are clearly discernible in the satellite image shown in Fig. 1 as streets of low-level clouds, transitioning from stratiform to cumuliform clouds as one goes from south to north. While this CAO is not among the most extreme CAOs occurring in this region in terms of intensity, air–sea heat fluxes, or duration, the archetypal development of a train of mesocyclones accompanying it and the splitting of the CAO air mass into several tongues make it ideally suited to elucidate the questions raised above.

Fig. 1.
Fig. 1.

Infrared (channel 4) satellite images from the NOAA AVHRR instrument valid at (left image) 0009 UTC and (right image) 0137 UTC 26 Jun 2010, with only the earlier image shown where they overlap. A synoptic-scale cyclone in the ABS (green), CAO tongues (blue), and mesocyclones (red) are indicated.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

The high-resolution numerical weather prediction model of the Consortium for Small-Scale Modeling (COSMO; Baldauf et al. 2011) is employed to simulate the dynamical evolution of the CAO. For its analysis we adopt the cold airmass concept introduced by Iwasaki et al. (2014) and Shoji et al. (2014), which allows us to quantify the relative importance of the various physical processes for the evolution of the CAO air mass within a comprehensive isentropic framework. The setup of the numerical simulation and the diagnostic framework are introduced in section 2, followed by an overview of the synoptic evolution in section 3. The quantitative analysis of the cold air mass is presented in section 4, complemented by sensitivity experiments with perturbed surface fluxes of sensible and latent heat in section 5, which yield insight into the relative importance of both types of fluxes for the decay of the CAO. The findings from sections 35 are synthesized into a conceptual four-stage model in section 6, followed by conclusions in section 7.

2. Methodology

a. The COSMO model

1) Model description

The COSMO model is in operational use at various weather services, among them the German and the Swiss national weather services, and it is designed for simulations at the mesoscale with a horizontal grid spacing in the range of 50–1 km (Steppeler et al. 2003; Baldauf et al. 2011). It has previously been employed for case studies in the polar regions, in particular in the marginal ice zone (e.g., Wacker et al. 2005; Schröder et al. 2011). The model’s ability to realistically represent the boundary layer and surface conditions in these case studies was verified with in situ observations. Wacker et al. (2005) concluded that, even though the model was developed for midlatitude meteorological conditions, it is well suited also for process studies in polar regions.

The model solves the thermo–hydrodynamic equations on a rotated spherical grid with a height-based terrain-following vertical coordinate system. It includes physical parameterizations for radiation, subgridscale turbulence using a level-2.5 closure scheme based on turbulent kinetic energy (TKE), subgridscale orography, a modified version of the Tiedtke mass-flux scheme (Tiedtke 1989) for moist convection, and a detailed microphysics scheme with prognostic equations for water vapor, cloud water, cloud ice, snow, and rain. Details about the parameterizations are given in Doms et al. (2011).

The realistic representation of the CAO and the development of associated mesocyclones depends critically on the surface layer parameterization, in particular on surface fluxes (Pagowski and Moore 2001; Wacker et al. 2005). The COSMO model was developed for numerical weather prediction over continental Europe, and thus its surface layer parameterizations have neither been specifically adapted to, nor validated under high-latitude marine conditions. Furthermore, a close investigation of the transfer coefficients obtained from the standard TKE-based scheme casts doubt on their realistic dependence on stability in such conditions (not shown). To simulate the marine surface layer as accurately as possible, we extended the model by the state-of-the-art air–sea heat and momentum flux parameterization from the Coupled Ocean–Atmosphere Response Experiment (COARE; Fairall et al. 1996, 2003). Fluxes parameterized by this scheme are in good agreement with in situ observations also in relatively high-wind speed oceanic conditions (up to 20 m s−1; Fairall et al. 2003; Cook and Renfrew 2015). It is only applied in ocean grid cells, whereas above land and sea ice a formulation following Louis (1979) is employed. According to Schröder et al. (2011), using the Louis scheme over sea ice leads to a more realistic surface layer than with the newer TKE-based scheme.

A thermodynamic sea ice model (Schröder et al. 2011) is used with prognostic sea ice thickness and surface temperature, which considerably improves the representation of the boundary layer above sea ice. This is key for preconditioning the CAO air masses before they are advected across the sea ice edge (e.g., Pagowski and Moore 2001; Vihma and Brümmer 2002).

2) Setup

Simulations are performed at a resolution of 0.05° × 0.05°, corresponding to an equivalent grid spacing of approximately 5.5 km, and with 40 vertical levels ranging from 10 m above ground to approximately 22 200 m above sea level. The simulations are initialized at 0000 UTC 24 June 2010, approximately 1 day before the onset of cold air advection across the sea ice edge, and the model is run for 3.5 days. Initial and boundary conditions are taken from 6-hourly analyses of the European Centre for Medium-Range Weather Forecasts (ECMWF). The model grid is rotated such that the equator crosses the domain approximately in the middle; i.e., the North Pole of the rotated model grid is located at 15°N and 155°W in geographical coordinates. In addition, we run the model in the so-called climate mode, allowing for continuous updates of sea surface temperature (SST) from the 6-hourly analyses.

The model domain and topography are depicted in Fig. 2. The domain consists of 1100 × 740 grid points and covers the RS and the ABS, as well as parts of the Transantarctic Mountains, the Ross Ice Shelf, and Marie Byrd Land (see Fig. 2 for location names). Note that in this study we consider the part of the RS between the Ross Ice Shelf and 71°S—Cape Adare, the easternmost tip of Oates Land—as the interior RS.

Fig. 2.
Fig. 2.

Location map of the study domain and model topography (shading). In addition, the sea ice edge (black) and the 280-K isotherm of SST (blue) are shown. The outer boundary of the interior Ross Sea is indicated by the black lines. The latitude circles are shown at 60° and 80°S and the longitude circles every 20°. The upper-right inset indicates location and orientation of the model domain, whereby the depicted region is bounded by the 35°S latitude circle.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

3) Sensitivity experiments

To investigate the influence of surface sensible and latent heat fluxes on the erosion of the CAO air mass, we perform sensitivity experiments with separately switched off surface fluxes of sensible and latent heat. The impact of perturbing several physical processes at once can be strongly nonlinear. To identify the synergistic effect of surface sensible and latent heat fluxes, in addition a simulation is performed with both types of fluxes switched off, giving rise to a total of four simulations (cf. Table 1).

Table 1.

Configuration of sensible (SHFL) and latent heat fluxes (LHFL) for the set of simulations performed in this case study. A ✓ () symbol denotes that the respective surface flux is switched on (off).

Table 1.

The perturbations to the heat fluxes are applied only after 0000 UTC 25 June, the time immediately before the bulk of the cold air mass is advected across the sea ice edge. This ensures that before 0000 UTC 25 June the sensitivity simulations evolve in the same way as the reference simulation with full fluxes, herein denoted as REF. The simulations are then run for 1.5 days after the perturbations have been introduced—that is, until 1200 UTC 26 June. Longer runs would make the comparison with REF increasingly difficult, as the large-scale flow deviates more and more from REF. Nevertheless, this time span is sufficient to test for the sensitivity of the mechanisms promoting the decay of the CAO air mass to surface sensible and latent heat fluxes.

b. Diagnostic framework

1) Cold airmass diagnostic

To study the dynamical evolution of the CAO, we employ the cold airmass diagnostic introduced by Iwasaki et al. (2014) and Shoji et al. (2014). The mass per unit area of a column of air extending from the ground to a particular isentropic surface with threshold potential temperature is given by
e1
where ρ denotes total air density and () and () are heights (pressures) of the ground and the isentropic surface , respectively. The hydrostatic relationship was invoked to obtain the right-hand side. Accordingly, the mass per unit area is proportional to the pressure difference . Because of the convenient interpretation of as the depth of the air mass, we define the cold airmass simply as
e2
To ensure a horizontally bounded cold air mass, the threshold potential temperature has to be chosen such that the corresponding isentrope intersects the ground. Specifically in this study we choose 280 K. Even though 280 K is not a particularly low potential temperature, its choice is justified by the fact that the SST in the area influenced by the CAO is about 280 K at the CAO’s northernmost tip (cf. blue contour in Fig. 2). If a lower value was chosen, the isentropic surface would intersect the ground in regions where the SST exceeds this value and not all of the CAO air mass would be included.
Rewriting Eq. (1) yields
e3
where corresponds to the mean density in the column of cold air and and denote the heights of the isentropic surface and the ground surface, respectively. In regions with low surface heights, the cold air mass is proportional to the topography of the isentropic surface, as horizontal fluctuations of the column mean density are weak. In particular, a deep cold air mass goes along with a doming of the isentropic surface. However, in the presence of elevated topography, the cold air mass is reduced compared to the height of the isentrope.
Following Iwasaki et al. (2014), the tendency of the cold air mass is given by
e4
where denotes the latent heating rate and the horizontal wind. Accordingly, the cold air mass changes as a result of the divergence of the horizontal cold airmass flux and the mass flux across the isentropic surface induced by diabatic processes. Diabatic processes within the cold air mass, as well as at the surface—for instance, air–sea sensible heat fluxes—do not directly influence the cold air mass. Equation (4) allows us to trace the evolution of the CAO air mass in an essentially compact way using maps of the horizontal transport and the diabatic gain or loss of cold air mass. Conceptually and computationally this is simpler than sampling the evolution of the CAO air mass using kinematic trajectories. Furthermore, the cold airmass fluxes can be directly linked to the dynamical processes associated with specific flow features (e.g., mesocyclones).

Note that the cold airmass is defined in terms of a pressure difference conventionally expressed with units in hPa. Accordingly, the horizontal cold airmass flux has units hPa m s−1 and its divergence, as well as the diabatic mass flux across the upper-bounding isentrope, have units of hPa s−1, which for practical reasons will be expressed in hPa day−1. By virtue of relation (1), the cold air mass and its tendency can readily be converted into the actual mass and mass tendency, respectively.

2) Relevant diabatic processes

Latent heat release by phase changes associated with condensation or freezing at the isentropic level erodes the cold air mass because it increases potential temperature to a value above , whereas cloud-top radiative cooling, as well as evaporation, melting, and sublimation of cloud condensate and hydrometeors increase the cold air mass. Continuous heating by surface sensible heat fluxes warms the air column from below and thereby reduces static stability. Subsequent convective and turbulent mixing can induce a temperature tendency on the isentropic surface. Therefore, the individual tendencies from the following three diabatic processes are written out from the model: radiation (), latent heating (), and subgridscale turbulent mixing (). Specifically, the term comprises the contributions from latent heating or cooling due to subgridscale convection and large-scale microphysical processes. Furthermore, includes the tendencies from turbulent and convective mixing, explicit horizontal diffusion, as well as from the subgridscale orography scheme. To close the temperature budget, the tendencies due to Rayleigh damping at the model top and boundary relaxation are collected in . Within the study domain, however, K h−1 and therefore this term is not explicitly shown in the mass budget. To deal with the high temporal variability of the temperature tendencies, the tendencies at a certain time t are averaged over the interval , where 15 min is the time interval between two output time steps.

Based on these temperature tendencies, the diabatic cold airmass flux is calculated for each of the diabatic processes. Despite the exact closure of the temperature budget from the model output, the cold airmass budget is not fully closed and a residual is diagnosed as
e5
The imbalance arises entirely as a result of numerical errors in the diagnostic computation of the cold airmass budget and the exclusion of grid points with weak static stability at the isentropic level—that is, with K (100 hPa)−1. As such unstable situations occur predominantly in association with an upward mass flux , the total diabatic cold airmass loss is slightly underestimated and accordingly the residual typically has a negative sign.

3. Synoptic evolution

a. Overview

The evolution of upper-level PV and the jet stream before and during the CAO is illustrated in Fig. 3. On 24 June an anticyclonically breaking stratospheric PV streamer is located over the eastern RS, downstream of the intense jet streak in the south Indian Ocean and the western RS (Fig. 3a). This wave breaking event is subsequently followed by the development of an upper-level PV trough over the ABS between 25 and 27 June (Figs. 3b,c). Along with the trough formation the jet stream is diverted away from the coastline to the north and around the ABS, such that the ABS becomes relatively sheltered from migratory synoptic systems (Figs. 3c,d).

Fig. 3.
Fig. 3.

(a)–(d) Upper-level PV (shading) and wind speed (black contours) for 40 m s−1 and higher values, in intervals of 10 m s−1, on 310 K daily at 0000 UTC from 24 to 27 Jun 2010. Data are from the ERA-Interim (Dee et al. 2011).

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

On 24 June a midtropospheric cyclone, associated with the upper-level cyclonic PV over the RS, extends from the Transantarctic Mountains into the western Amundsen Sea, as indicated by the closed 4800-m contour of geopotential height at 500 hPa (Fig. 4a). With the formation of the upper-level PV trough this midtropospheric cyclone translates northeastward into the ABS, establishing a cyclonic circulation over the ABS with easterly winds over the Antarctic continent, namely over Marie Byrd Land (Figs. 4c,e,g). Subsequently, the center of the cyclone remains almost stationary over the ABS until the peak of trough formation at 1200 UTC 27 June, whereas its influence over the Antarctic continent diminishes, accompanied by a decline of the easterly winds (Figs. 5a,c,e).

Fig. 4.
Fig. 4.

Overview on the synoptic evolution of the CAO from the COSMO simulation at (a),(b) 1200 UTC 24 Jun; (c),(d) 0000 UTC 25 Jun; (e),(f) 1200 UTC 25 Jun; and (g),(h) 0000 UTC 26 Jun 2010. (left) Upper-level PV on 310 K (shading) and geopotential height at 500 hPa (blue contours, in intervals of 100 m). (right) Cold airmass (shading), as well as SLP (gray contours, in intervals of 5 hPa), (blue contours, from 0 K in intervals of 4 K), and the sea ice edge (black, thick line). Note that SLP contours are not drawn over land. The discussed synoptic and mesoscale cyclones are labeled. In (b) the subregion shown in Fig. 6 is outlined by the black box.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

Fig. 5.
Fig. 5.

As in Fig. 4, but at (a),(b) 1200 UTC 26 Jun; (c),(d) 0000 UTC 27 Jun; and (e),(f) 1200 UTC 27 Jun 2010. In (b) the black box outlines the region shown in Figs. 8 and 1113, and in (f) the location of the cross section of Fig. 10 is shown. In addition, the box, over which the cold airmass budget is calculated, is outlined in black. The dashed and dotted lines denote the influx boundaries from the south Indian Ocean (A) and the interior RS (B), respectively.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

In the lower troposphere the flow is dominated by a synoptic-scale surface cyclone and the development of a train of mesocyclones associated with the CAO. At 1200 UTC 24 June the synoptic-scale cyclone is located downstream of the midtropospheric cyclone and it has two distinct centers, as is seen from the two minima in sea level pressure (SLP) (Fig. 4b), one in the eastern RS (C1), and a second one in the Amundsen Sea (C2). While C1 decays during the following 24 h, C2 intensifies and is steered toward the coast of Marie Byrd Land by the northeasterly winds ahead of the midtropospheric cyclone (Figs. 4d,f,h). It remains stationary over the sea ice region until it decays on 27 June (Figs. 5b,d).

In addition to the synoptic surface cyclone, a train of three mesocyclones (M1 to M3) develops along the outer edge of the CAO, whereas a fourth mesocyclone (M4) has genesis on the cold front of C2 in the Bellingshausen Sea. These simulated mesocyclones are reasonably well collocated with those visible in the satellite image (Fig. 1).

b. Drainage from West Antarctica and accumulation in the interior Ross Sea

At 1200 UTC 24 June a tremendous pool of cold air extends over the interior RS and the western Ross Ice Shelf, sharply bounded by the Transantarctic Mountains, with a maximum cold air mass of more than 500 hPa (Fig. 6a and Fig. 4b). The air mass enters the region on the eastern side of the Ross Ice Shelf under the influence of easterly winds prevailing in the elevated regions of Marie Byrd Land (Fig. 6b). There, the cold airmass flux is approximately parallel to contours of the 500-hPa geopotential height, indicating that the midtropospheric cyclone leads to a nearly geostrophically balanced downslope flow. The air then descends along the sloping topography toward the Ross Ice Shelf, leading to a confluence of cold air mass near Siple Coast and over the Ross Ice Shelf.

Fig. 6.
Fig. 6.

(a) Cold airmass and (b) the magnitude of its flux at 1200 UTC 24 Jun 2010. In addition, in (b) also the horizontal cold airmass-flux vectors and geopotential height at 500 hPa in intervals of 50 m (blue) are shown. In both figures SLP contours—omitted over topography—in intervals of 5 hPa are depicted in gray.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

Despite the fact that the cold air mass over the elevated terrain of Marie Byrd Land amounts only to about 50% of the cold air mass over the Ross Ice Shelf, the magnitude of the cold airmass flux over Marie Byrd Land exceeds that over the Ross Ice Shelf. This is due to considerably stronger winds over Marie Byrd Land throughout the entire depth of the column of cold air (not shown). The most intense drainage flow occurs near the surface with a wind speed maximum of more than 25 m s−1, which is thought to be caused by a combination of katabatic forcing and the topographic adjustment of the large-scale flow (cf. Parish and Cassano 2003a,b). Furthermore, note that the drained air masses from West Antarctica have a higher potential temperature than the original near-surface air over the Ross Ice Shelf (not shown) and thus lead to a warming over the Ice Shelf. Such a warming is typical for events of strong drainage from West Antarctica (Coggins et al. 2014). Nevertheless, the drainage is sufficiently cold to increase the cold air mass over the Ross Ice Shelf.

Over the Ross Ice Shelf, the cold airmass flux is influenced by the cyclonic flow associated with the surface cyclone C1 with its center in the eastern RS. The air mass is advected across the Ross Ice Shelf toward the Transantarctic Mountains, which cannot be overflown by most of the column of cold air mass. Thus, the cold air mass is blocked and piles up when its flux impinges on the topographic barrier, accompanied by the development of an east–west pressure gradient over the Ross Ice Shelf. Geostrophic adjustment gives rise to an intense barrier jet (O’Connor et al. 1994), which promotes the rapid northward advection of the cold air mass in a narrow band along the mountain range toward the border of the ice shelf. This barrier jet is a variant of the so-called Ross Ice Shelf airstream, a frequent, rather persistent, and intense southerly wind over the western Ross Ice Shelf (e.g., Parish et al. 2006). The flux of cold air mass associated with the Ross Ice Shelf airstream is more than twice as large as the outflow of cold air mass leaving the interior RS (Fig. 6b). Even though this intense flux occurs in a fairly narrow band, it transports a huge amount of cold air mass, which subsequently accumulates in the northern part of the ice shelf and farther north.

The observed flow pattern characterized by easterly winds over Marie Byrd Land, the descent of the cold air mass into the Ross Ice Shelf region, and its rapid northward advection along the Transantarctic Mountains is characteristic of situations in which a synoptic-scale cyclone is located in the eastern RS (e.g., Fig. 2 of Parish et al. 2006; Coggins et al. 2014; Nigro and Cassano 2014). Such a configuration of the large-scale flow occurs during almost 25% of the time, with the highest frequencies in autumn and spring (Coggins et al. 2014). In the absence of the midtropospheric cyclone the low-level flow over the Ross Ice Shelf would still drain air from Marie Byrd Land as a density current. However, as noted by Bromwich et al. (1992) and Liu and Bromwich (1997), an easterly large-scale flow over Marie Byrd Land substantially amplifies this drainage, suggesting that the midtropospheric cyclone is key for the accumulation of cold air mass in the interior RS.

c. Formation of a cutoff dome of cold air

The further evolution of the cold air mass is depicted in the right columns of Figs. 4 and 5. At 1200 UTC 24 June the incipient CAO off the sea ice edge in the Amundsen Sea, evident from the region of elevated air–sea potential temperature difference (cf. Papritz et al. 2015), is associated with a tongue of deep cold air mass (Fig. 4b). The mesocyclone M1 develops in the right exit region of the jet maximum, which is favorably aligned with the western flank of the cold airmass tongue. This western flank goes along with an enhanced slope of the isentropic surface. This is evident from the horizontal gradient of DP, which in the absence of surface topography is predominantly determined by spatial variations of —that is, the topography of the isentropic surface. In general, baroclinic growth requires motion in a plane that is less steep than the corresponding isentropic surfaces and the maximum baroclinic growth rate is the larger the steeper the isentropic surfaces are (e.g., Green 1960, 1979). Accordingly, the enhanced slope of the isentropic surface along the outer edge of the CAO (which also corresponds to an enhanced Eady growth rate; not shown) goes along with an increased potential for baroclinic development and it provides favorable conditions for the growth of the mesocyclones.

During the day after 1200 UTC 24 June an intense east–west surface pressure gradient is established over the RS (Figs. 4d,f), which causes a rapid northward export of the air mass out of the interior RS. Farther to the north, over the sea ice–covered part of the RS, midtropospheric and low-level southwesterly winds become vertically aligned by 1200 UTC 25 June, as the midtropospheric cyclone moves into the ABS. This leads to a vertically extended, deep transport of air from the interior RS toward and across the sea ice edge. Accordingly, the depth of the cold air mass increases over open ocean, and by 0000 UTC 26 June the isentropic surface assumes the shape of a dome in the ABS (Fig. 4h), which, however, remains connected to the interior RS. This dome is relatively well collocated with the zero isoline of the air–sea potential temperature difference , indicating the transition from statically stable to unstable conditions. During the formation of the dome, the mesocyclones M2 and M3 have genesis at its baroclinic outer edge, whereby M2 is located downstream and M3 upstream of the deepest cold air mass (Fig. 4f).

Comparing the upper-level PV charts with the cold air mass during the formation of the dome reveals a remarkable spatial correlation (cf. Figs. 4 and 5): the deepest cold air mass, and thus the most elevated regions of the isentropic surface, are located beneath the upper-level PV trough. This is consistent with what is expected from the invertibility principle of PV, which implies that below a cyclonic upper-level PV anomaly, such as an upper-level PV trough, isentropic surfaces are bent upward toward the anomaly (e.g., Hoskins et al. 1985).

The dome of cold air becomes cut off from the RS within the following 12 h (Fig. 5b) because the southwesterly winds weaken and the supply of air from the RS ceases. Concomitantly, a narrowing occurs of the band of strongly cyclonic upper-level PV over the interior RS, which connects the polar vortex with the PV trough (see Fig. 5a). Furthermore, cyclone C2 on its southern flank continuously advects warm air westward into the Amundsen Sea, thereby creating a wedge of warm air displacing the cold air mass between the coast and the sea ice edge. Accordingly, regions of cold air mass deeper than 400 hPa become restricted to the region off the sea ice edge. The intensifying cyclonic circulation associated with the mesocyclones M2 and M3 causes a splitting of the dome into two major tongues of cold air over the ocean, which are also seen in the air–sea potential temperature difference (Figs. 5b,d). Thereby, two processes separating the easterly tongue from the westerly one come into play: first, the advection of relatively warm air in the warm sector of M3, displacing the cold air mass, and second, the erosion of cold air mass by diabatic processes. These processes will be analyzed further in the following section.

4. Cold airmass evolution

a. Cold airmass budget

To quantitatively assess the role of horizontal and diabatic mass fluxes for the evolution of the CAO, we consider the cold airmass budget integrated over the box outlined in Fig. 5f. The evolution of the cold air mass in this box can be divided into two major phases (Fig. 7):

  1. The first two days are characterized by a strong influx of cold air with a peak about 1 day after the start of the simulation—that is, at 0000 UTC 25 June.
  2. The cessation of this strong influx after about 2 days and the strengthening of the diabatic mass loss herald the decay of the cold air mass.
Fig. 7.
Fig. 7.

Evolution of the cold airmass fluxes in the box outlined in Fig. 5f. (a) The total cold airmass tendency within the box (black), the horizontal cold airmass flux across the lateral boundaries (blue, solid), as well as the contributions of the fluxes from the south Indian Ocean (A, blue dashed) and the interior RS (B, blue dotted). In addition, the residual of the cold airmass budget within the box is shown in gray. (b) The diabatic mass flux across the upper-bounding isentropic surface of the cold air mass. Shown are the total diabatic mass flux (solid) and individual contributions from latent heating (dashed), radiation (dotted), and turbulence (dashed–dotted). In both panels the time axis is relative to the initialization time of the simulation; that is, time zero corresponds to 0000 UTC 24 Jun 2010. All tendencies are expressed in terms of the actual mass.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

On early 25 June a large amount of cold air mass rapidly enters the domain from the sea ice covered part of the south Indian Ocean (dashed line Fig. 7a), which converges with the weaker influx from the interior RS east of Cape Adare (dotted line). Traced further, this cold air mass forms the CAO tongue between M2 and M3 (cf. Figs. 4d,f,h). The export from the RS peaks a few hours after 1200 UTC 25 June. The magnitude of the flux from the RS remains below the maximum flux from the south Indian Ocean, but it persists over a longer period. It provides the cold air mass for the CAO tongue upstream of M3 (Figs. 4f,h and 5b).

Integrated over the entire period of 3.5 days, the flux of cold air mass from the interior RS ( kg) approximately equals that from the sea ice covered south Indian Ocean ( kg). Hence, about 38% of the cold air mass enters the domain via the Ross Ice Shelf and the interior RS. This corresponds closely to the about 35% of CAO trajectories that pass the Ross Ice Shelf in the climatological mean (Papritz et al. 2015).

Overall, the net loss of cold air mass observed after 0000 UTC 26 June is of diabatic nature (Fig. 7a), whereas the dilution and loss by horizontal fluxes is weak (Figs. 7a,b). Even though the net effect of diabatic processes is to erode the cold air mass, various diabatic processes have competing effects on the cold airmass tendency. Cloud-top radiative cooling leads to a gain of cold air mass, which is relatively constant throughout the entire simulation period. Most of the loss of cold air mass is due to latent heat release. Only between 1200 UTC 25 June and 1200 UTC 26 June turbulent mixing contributes to a reduction of cold air mass, albeit at a weaker rate than latent heat release. Consistent with the argument given in section 2, the residual of the cold airmass budget is negative (Fig. 7a), which is due to diabatic cold airmass loss under weakly stable situations at the 280-K isentropic level.

The increase of the diabatic cold airmass erosion by latent heat release after 0000 UTC 26 June coincides with the intensification of the mesocyclones, in particular of M3. Because of the doming of the isentropic surfaces within the cold air mass, the associated cyclonic circulation enforces ascent, accompanied by the release of latent heat and the loss of cold air mass. This results in filaments of shallow cold air mass spiraling inward into the core of the mesocyclones (Figs. 5b,d).

b. Diabatic cold airmass fluxes

To analyze in more detail how specific diabatic processes modify the cold air mass, we consider the subregion outlined in Fig. 5b at 1200 UTC 26 June. The subregion is centered on mesocyclone M3 and comprises its warm sector, the CAO tongue farther downstream, and parts of the incipient CAO tongue upstream, as well as mesocyclone M2 (Fig. 8). The air mass of the downstream CAO tongue is older in terms of a longer exposure to surface sensible and latent heat fluxes and it is collocated with a deep boundary layer, which is in the process of being eroded. In contrast, the upstream CAO air mass has been advected across the sea ice edge less than a day ago and sensible heat fluxes have caused an increase of the depth of the boundary layer, while latent heat fluxes have gradually moistened the air mass. Cloud water and ice accumulates in the upper part of the boundary layer and as the boundary layer gradually deepens with increasing age of the CAO air mass, the elevation of the clouds increases too (Fig. 9a).

Fig. 8.
Fig. 8.

(a) Cold airmass and (b)–(d) diabatic mass flux across the upper-bounding isentropic surface of the cold air mass for (b) all diabatic processes, (c) latent heating, and (d) radiation for a subregion of the CAO centered on M3 (cf. Fig. 5b) and valid at 1200 UTC 26 Jun. Gray contours show SLP in intervals of 5 hPa. In addition, the 100-, 300-, and 500-hPa contours of the cold air mass are drawn as a reference (thin black). The thick black contour delineates the sea ice edge. Furthermore, in (a) the location of the cross section shown in Fig. 9 is drawn.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

Fig. 9.
Fig. 9.

(a) Specific cloud water plus ice content at 1200 UTC 26 Jun 2010 along a cross section across the warm sector of M3 (see Fig. 8a). (b) Latent and (c) radiative heating rates. Potential temperature (black) is shown in intervals of 3 K with the 280- and 310-K isentropes highlighted in bold.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

The total diabatic cold airmass flux (Fig. 8b) is characterized by a strong cold airmass loss in the warm sector of M3 and a mass gain in the downstream CAO tongue, whereas no net diabatic flux is observed in the upstream CAO tongue. This spatial distribution of diabatic mass loss and gain, respectively, is caused by latent heating or cooling and cloud-top radiative processes, as discussed in detail in the following.

Depending on the sign of the diabatic temperature tendency, microphysical processes can either reduce or enhance the cold air mass. Latent heat release in the warm sector of mesocyclone M3 is accompanied by an upward cross-isentropic mass flux reducing the cold air mass (Fig. 8c), which contributes substantially to the cold airmass erosion. In the CAO tongues, cloud formation is mostly due to convection, fuelled by surface sensible and latent heat fluxes. The latent heat release associated with this convective cloud formation occurs below the 280-K isentropic surface (Fig. 9b), such that its effect is limited to warming the CAO air, while not giving rise to a mass flux across the upper-bounding isentrope. As the mixed layer grows, dryer air is entrained from above and the clouds dissolve, giving rise to latent cooling at the top of the mixed layer. In order for this process to induce a cold airmass gain, the cloud tops must be in close proximity to the upper-bounding isentropic surface. Therefore, it is only at the later stages of the CAO evolution, when the boundary layer is sufficiently deep, that cold air mass is gained (see again Fig. 9b). Accordingly, no mass gain is observed in the upstream CAO tongue, whereas in the downstream tongue a mass gain occurs (Fig. 8b). This mass gain is most intense on the cold side of the warm front of M3, where the ageostrophic frontal circulation is associated with descent and the dissolution of the frontal clouds in the cold air.

While the cloud base is weakly warmed by radiation, radiative cooling at cloud tops can be intense (Skyllingstad and Edson 2009). In the upstream cold air mass the cloud layer is located below the 280-K isentropic surface and cloud-top radiative cooling essentially cools the cold air but does not increase its mass (Fig. 9c). In contrast, in the downstream CAO tongue the boundary layer has grown to the height of the 280-K isentropic surface, such that cloud-top radiative cooling increases the cold air mass over a substantial fraction of the CAO tongue (Fig. 8d).

Within the downstream cold airmass tongue the entrainment and turbulent mixing of potentially warmer air counteracts the mass gain by radiative and evaporative cooling (not shown). However, comparing Figs. 8b–d reveals that the contribution of turbulent mixing remains considerably below that of radiative and evaporative cooling. Therefore, the contributions of latent cooling and radiation dictate the spatial distribution and intensity of the total diabatic cold airmass flux within the cold airmass tongues. In contrast, in the warm sector the intense cold airmass loss is almost exclusively caused by latent heat release.

c. Formation of a deep boundary layer

The well-mixed atmospheric boundary layer grows with distance from the sea ice edge—that is, the longer the air column is exposed to strong surface heat fluxes, while entraining air from above. This boundary layer can grow approximately up to the isentropic level, where potential temperature equals SST, which is about 280 K at the northernmost tip of the CAO. Thus, the top of the boundary layer will always be below or at the upper-bounding isentrope of the cold air mass.

The insulation of the CAO tongues by cloud-top radiative and latent cooling ensures that the potentially warmer air entrained from above is cooled to a potential temperature below 280 K, which effectively prevents the erosion of the cold air mass. This deepening benefits further from reduced stability below the upper-level PV trough—that is, a weak increase of potential temperature with height—such that, in the CAO tongue upstream of M3, the boundary layer reaches a depth of almost 500 hPa by 1200 UTC 27 June 2010, as evident from the cross section shown in Fig. 10.

Fig. 10.
Fig. 10.

Vertical cross section across the CAO air mass upstream of M3 valid at 1200 UTC 27 Jun 2010 (see Fig. 5f). Shown are PV (shading) and potential temperature (black contours, in intervals of 3 K), with the 280- and 310-K isentropes highlighted in bold.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

The greatest depth of the boundary layer coincides with the position of the upper-level PV trough. Within the trough, the dynamical tropopause (defined as the −2-PVU surface; 1 PVU = 10−6 K kg−1 m2 s−1) is almost as low as 500 hPa. It is thus in close proximity to the top of the boundary layer. In such circumstances, cloud radiative processes and turbulent mixing can cause the exchange of mass between the boundary layer and the stratosphere, known as deep stratosphere–troposphere exchange. According to a climatological analysis of deep stratosphere–troposphere exchange based on ERA-Interim data by S̆kerlak et al. (2014), contributions to such exchanges in the Pacific sector of the Southern Ocean are small. It is conceivable that such deep boundary layers and the close proximity of the top of the boundary layer to the tropopause, as in the present CAO case, are rather exceptional such that in this region CAOs indeed do not contribute significantly to the climatological deep stratosphere–troposphere exchange. Comparing the boundary layer structure in ERA-Interim with that in the high-resolution COSMO simulation indicates that the depth of the boundary layer within the CAO is significantly lower in ERA-Interim (not shown). Thus, it appears possible that these exchange events are simply not represented in relatively coarse-scale reanalysis datasets. However, also an overestimation of the boundary layer depth in the COSMO model cannot be ruled out and in situ observations would be needed to address this question conclusively.

5. Sensitivity to air–sea heat fluxes

To assess the relative importance of surface sensible and latent heat fluxes for the erosion of the cold air mass, we perform sensitivity experiments with separately switched off fluxes. These perturbations to the physics are applied immediately before the bulk of the cold air mass is advected across the sea ice edge—that is, at 0000 UTC 25 June—until the end of the simulation. We analyze the response at 1200 UTC 26 June.

a. Circulation response

To characterize circulation changes, we focus on SLP (left column of Fig. 11) and thus restrict ourselves to changes in the lower troposphere that are most relevant for the evolution of the cold air mass. Heating of a column of air results in divergence aloft owing to the expansion of the air, accompanied by a lowering of surface pressure (e.g., Holton and Hakim 2012). Accordingly, in the absence of surface sensible heating or a reduction of latent heat fluxes, higher surface pressure is observed in both sensitivity experiments compared to REF, with the largest differences in the dynamically active regions associated with mesocyclones M2 and M3. Mesocyclone M3 is well developed in both simulations with a slight upstream shift in the absence of sensible heat fluxes. The largest changes in SLP are found in the downstream part of the subregion, where especially mesocyclone M2 shows considerable differences in size and depth.

Fig. 11.
Fig. 11.

Changes in (a),(c) SLP and (b),(d) cold air mass at 1200 UTC 26 Jun, centered on mesocyclone M3 (cf. Fig. 5b), in sensitivity experiments with switched off surface fluxes of sensible (NOSHFL) and latent (NOLHFL) heat with respect to the reference run (REF). In addition, in (left) SLP in the respective sensitivity experiment (red) and in REF (gray) are shown in intervals of 5 hPa. Additionally SLP in intervals of 5 hPa (gray) and the 100-, 300-, and 500-hPa contours of the cold air mass (black) from the reference simulation (REF) are shown in (right). The thick black contour delineates the sea ice edge.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

As all simulations overall agree well with respect to the large-scale distribution of SLP, the differences in the cold airmass evolution discussed below can in the first place be attributed to the perturbed physical processes, whereas kinematic effects play a secondary role.

b. Sensitivity of cold airmass evolution

As the kinematic changes to the cold airmass evolution are of secondary importance, a positive cold airmass anomaly in the sensitivity experiments (Figs. 11b,d) indicates that the diabatic cold airmass erosion is generally reduced in the cases with no surface fluxes. For the interpretation of these differences it is instructive to additionally consider changes in vertically integrated water vapor, cloud water, and cloud ice (Fig. 12).

Fig. 12.
Fig. 12.

As in Figs. 11b and 11d, but for (a),(c) the percentage change in vertically integrated water vapor relative to REF and (b),(d) the absolute change in vertically integrated cloud water and ice. Vertical integrals are taken over the entire column.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

Without evaporation from the ocean surface, the column integrated water vapor is reduced by up to 75% (Fig. 12a), accompanied by strongly reduced cloud formation both in the areas influenced by the CAO tongues and associated with M3, especially in its warm sector and along the warm front (Fig. 12b). Consequently, the diabatic erosion of the cold air mass by M3 is almost absent and accordingly cold air mass remains larger by up to 250 hPa in the warm sector (Fig. 11b).

The absence of surface sensible heat fluxes prevents the development of a deep boundary layer, limiting the atmospheric moisture content and reducing evaporation from the ocean surface as the cold air mass remains colder. Therefore, vertically integrated water vapor is reduced by 15% to 30% (Fig. 12c). Convection is more shallow, such that the well-mixed boundary layer becomes less deep in comparison to REF. The cloud mass is increased within the CAO tongues starting immediately off the sea ice edge (Fig. 12d), which is due to the fact that cloud formation sets in at lower specific humidity and closer to the sea surface. Because of the insulating effect of the cloud layer and the shallow boundary layer, this does not affect cold airmass erosion (Fig. 11d). In contrast, cold airmass loss is clearly reduced associated with M3. Especially the circulation at the warm front is weaker, resulting in less latent heat release, which is reflected in less cloud formation (Fig. 12d) and precipitation (Fig. 13a). In contrast to that, the cold front is less intense in REF owing to the frontolytic effect of sensible heat fluxes that heat the cold air more than the warm air. Accordingly, ascent ahead of the cold front is stronger in the absence of sensible heat fluxes and also cloud formation and precipitation are more intense. However, this does not compensate for the reduction of the cold airmass loss along the warm front and in a major fraction of the warm sector.

Fig. 13.
Fig. 13.

Impact of sensible heat fluxes (a) on precipitation in the presence of latent heat fluxes (NOSHFL − REF) and (b) on cold air mass in the absence of latent heat fluxes (NOLHFL − NOHFL). The fields are shown centered on M3 (cf. Fig. 5b) at 1200 UTC 26 Jun. Gray and black contours are as in Fig. 11.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

The strengthening of the cold airmass erosion in the dynamically active regions due to sensible heat fluxes is a synergistic effect, which requires the moistening by latent heat fluxes. In the absence of latent heat fluxes, sensible heat fluxes have little impact on cold airmass erosion, as shown in Fig. 13b. The synergistic effect of surface sensible and latent heat fluxes on cold airmass erosion is owed to two principal factors. First, sensible heat fluxes destabilize the cold air mass and thereby, for a given latent heating , the cross-isentropic mass flux becomes larger according to Eq. (4). This effect, however, relies on the presence of diabatic processes, which are mostly due to latent heat release, and therefore requires a moistening of the cold air mass by evaporation from the ocean surface. Second, more moisture can be picked up by evaporation from the ocean when the air mass is heated by sensible heat fluxes, raising the potential for latent heat release. Therefore, sensible heat fluxes in the presence of moist processes increase both terms determining the cross-isentropic mass flux—that is, and .

6. Synthesis

Based on the findings from this case study we can divide the evolution of the CAO into four stages, characterized by different dynamical mechanisms and diabatic processes. The physical processes active during stages 2–4, which cover the evolution of the CAO from the time when the cold air mass is advected across the sea ice edge onward, are schematically summarized in Fig. 14a. In addition, the related evolution of the intensity of the CAO as measured by is sketched in Fig. 14b.

Fig. 14.
Fig. 14.

(a) Lagrangian schematic of the relevant physical processes during CAO stages 2–4, whereby the horizontal axis represents time and the vertical axis is height. Shown are the height of the mixed layer (blue), the 280-K isentrope (black), clouds (gray), surface sensible and latent heat fluxes (red arrows), radiative cooling (blue arrow), and convective and turbulent mixing (circular arrows). Latent heat release associated with the mesocyclones essentially contributes to the erosion of the cold air mass, indicated by the large red arrow. The stages of the CAO development are discussed in detail in the text. (b) Schematic evolution of (green) in the CAO air mass.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

a. Cold airmass accumulation

The interplay of the large-scale flow in the mid- and the lower troposphere, as well as the blocking effect of the Transantarctic Mountains, are essential for the rapid buildup of a large amount of cold air mass in the interior RS. This incipient cold air mass is essentially the drainage outflow of air from West Antarctica, which is forced to descend onto the Ross Ice Shelf under the influence of an almost stationary midtropospheric cyclone, leading to persistent easterly winds on the elevated Marie Byrd Land, as well as a surface cyclone in the eastern RS, accompanied by southeasterly winds over the Ross Ice Shelf. The Transantarctic Mountains block the impinging flow, and concomitantly with the surface cyclone in the eastern RS, give rise to an intense Ross Ice Shelf airstream that advects the cold air along the mountain range to the north of the ice shelf, where it accumulates.

b. Export of the cold air mass across the sea ice edge

The export of the cold air mass from the interior RS across the sea ice edge coincides with the formation of an upper-level PV trough. These two processes are dynamically linked, as isentropic surfaces dome below an upper-level PV trough, associated with an enhanced depth of the cold air mass. Kinematically, the onset of the rapid transport of the cold air mass across the sea ice edge is driven by the alignment of the southerly winds associated with the midtropospheric cyclone and the synoptic surface cyclone in the eastern RS, leading to a vertically deep transport. The cold air mass is additionally fuelled by the burstlike inflow of cold air from the sea ice covered part of the south Indian Ocean.

c. Cutoff dome formation and deepening of the boundary layer

Concomitant with the cyclonic wave breaking of the upper-level PV trough, the cold airmass flux from the interior RS decreases. Therefore, the dome of cold air becomes detached from the high-latitude reservoir of cold air. Sensible heat fluxes and latent heat release warm the cold air mass, whereby they weaken the intensity of the CAO (i.e., decreases) and cause a deepening of the boundary layer. In parts of the CAO, the top of the boundary layer consequently reaches up to the low-lying tropopause within the upper-level PV trough, facilitating the deep exchange of mass between the boundary layer and the stratosphere due to turbulent mixing or radiative processes.

Once the top of the boundary layer has risen up to the height of the 280-K isentropic surface, the entrainment of potentially warmer air by turbulent mixing in principle can erode the cold air mass. However, cloud-top radiative cooling and the dissolution of clouds due to the entrainment of dryer air tend to insulate the cold air mass against such turbulent erosion.

d. Erosion of the cold air mass

The bulk of the erosion of the cold air mass is driven by diabatic processes associated with a train of mesocyclones. These mesocyclones have genesis along the outer edge of the CAO, where the slope of the isentropic surfaces is particularly large, and therefore the baroclinic growth rate is high. Initially, the cyclonic circulation associated with the mesocyclones causes a fragmentation of the CAO air mass into several tongues of cold air and the incursion of warmer air from the north. Subsequent latent heat release in the warm sectors of the mesocyclones rapidly erodes the cold air mass. Thereby, the latent heat provided by evaporation from the ocean surface is the essential contributor to the erosion. Sensible heat fluxes alone do not significantly accelerate the erosion, but jointly with latent heat fluxes they unfold a significant amplification of the cold airmass loss. This synergistic impact of the fluxes is due to enhanced moisture availability, increasing latent heat release, and a reduction of static stability.

In relatively coarse-resolution models mesocyclones are not well resolved. For instance, Condron et al. (2006) found that in the ECMWF ERA-40 dataset, with an equivalent grid spacing of approximately 125 km, many mesocyclones with a diameter of less than 500 km are missing. Concluding from our results, this would imply that the erosion of the cold air mass in marine CAOs is too slow, leading to an overestimation of the intensity and lifetime of CAOs in such models. This view is supported by a comparison of the cold air mass at 1200 UTC 26 June 2010 in REF (Fig. 15a) with that in a simulation at a strongly degraded resolution of 1.0° × 1.0° (approximately 110 km; Fig. 15b). In the coarse-resolution simulation the mesocyclone M3 essentially fails to develop such that no splitting of the cold air mass into two tongues occurs, concomitant with a much-weaker diabatic erosion. Consequently, the cold air mass is substantially overestimated, increasing the spatial extent of the CAO and prolonging its lifetime.

Fig. 15.
Fig. 15.

Cold air mass (shading), SLP (gray contours, in intervals of 5 hPa), and the sea ice edge (black, thick line) at 1200 UTC 26 Jun 2010 for simulations at a horizontal resolution of (a) 0.05° × 0.05° and (b) 1.0° × 1.0°.

Citation: Monthly Weather Review 144, 1; 10.1175/MWR-D-15-0268.1

7. Conclusions

In this numerical process study, the mechanisms governing the large-scale evolution of an intense, marine CAO that occurred in the RS and ABS were quantitatively investigated in an isentropic framework. To this end, a simulation of the event at high resolution (~5.5 km) was performed using the numerical weather prediction model COSMO. Conceptually, four stages in the development of the CAO were identified, during which its evolution is shaped by different dynamical processes. These stages are the following:

  1. the accumulation of cold air mass over the Ross Ice Shelf and in the interior Ross Sea,
  2. the export of cold air mass across the sea ice edge below an incipient upper-level PV trough,
  3. the formation of a cutoff dome of cold air mass over the open ocean when the supply of cold air mass ceases, and
  4. the decay of the cold air mass, induced primarily by latent heating in the warm sectors of mesocyclones that develop on the baroclinic outer edge of the CAO.

A key finding of our study is that the extent, intensity, and lifetime of this CAO strongly benefits from the configuration of the upper-level flow, in particular the formation and cyclonic wave breaking of an upper-level PV trough. Throughout stage (i) the easterly winds over the elevated terrain of West Antarctica associated with a midtropospheric cyclone and the southeasterly low-level winds over the Ross Ice Shelf, imposed by a surface cyclone in the eastern RS, act together to drain continental air. This interplay of upper- and lower-level flow and the blocking effect of the Transantarctic Mountains account for the accumulation of a major part of the cold air mass constituting this CAO. The subsequent advection of the cold air mass toward and across the sea ice edge in stage (ii) is intimately related to the formation of an upper-level PV trough and the doming of isentropic surfaces below. Furthermore, the cyclonic breaking of the upper-level PV trough shelters the cold air mass by deflecting the jet stream around it and allows for the formation of a cutoff dome of cold air in stage (iii).

Once the cold air mass is located over open ocean, the boundary layer grows rapidly with distance from the sea ice edge as a result of the energy supplied by surface sensible and latent heat fluxes. In this particular case, it grows to extreme depths in the later stages of the CAO evolution, reaching up to the height of the tropopause, which under the influence of the upper-level PV trough reaches down to almost 500 hPa. This potentially allows for the deep exchange of mass between the boundary layer and the stratosphere. To what extent CAOs in the Pacific sector of the Southern Ocean contribute climatologically to deep troposphere–stratosphere transport, however, remains an important question for further investigations. The thorough quantification of the exchange of mass in this specific case—for instance, using numerical tracers, as well as evaluating the frequency of CAOs with extremely deep boundary layers—will provide first steps forward to settling this question.

A major conclusion from our analysis is that the decay of the CAO is essentially promoted by the circulation and diabatic processes associated with mesocyclones, whereas away from the mesocyclones radiative cooling at cloud tops and latent cooling due to the entrainment of dryer air at the top of the boundary layer insulate the cold air mass against turbulent erosion. First, the mesocyclones cause the fragmentation of the CAO into separate tongues, accompanied by the incursion of potentially warmer air. Second and most importantly, they induce the loss of cold air mass by latent heat release in their warm sectors, ramping up the decay of the CAO. As revealed in sensitivity experiments with suppressed surface fluxes of sensible or latent heat, the moistening of the cold air mass by latent heat fluxes from the ocean surface and the subsequent release of the latent heat is a prerequisite for the rapid erosion of the cold air mass. Sensible heat fluxes act to support this erosion process.

The finding that the decay of the CAO is ramped up by mesocyclones likely has important implications for the representation of CAOs in coarse-resolution global circulation models. If mesocyclones are not resolved, the erosion of the cold air mass in CAOs might be underestimated. Biases in CAO lifetime or intensity could lead to errors in the spatial distribution and intensity of mean air–sea heat fluxes in such models, affecting the ocean meridional overturning circulation, with potential implications on a global scale. Generally, our results emphasize the fundamental role of mesocyclones for the air–sea heat flux forcing of the ocean. In addition to their direct impact (Condron and Renfrew 2013), our findings suggest that mesocyclones have an indirect impact via their influence on the life cycle of CAOs.

Acknowledgments

We are indebted to Heini Wernli (ETH Zürich) for thoughtful discussions and thank three anonymous reviewers for their very constructive comments and suggestions that greatly helped to improve this manuscript. L. P. is supported by ETH Research Grant CH2-01 11-1. MeteoSwiss and the ECMWF are acknowledged for providing access to the ECMWF analyses. The open-source software package R (http://www.r-project.org/) has been used to create some of the figures in this study.

REFERENCES

  • Baldauf, M., , A. Seifert, , J. Förstner, , D. Majewski, , M. Raschendorfer, , and T. Reinhardt, 2011: Operational convective-scale numerical weather prediction with the COSMO model: Description and sensitivities. Mon. Wea. Rev., 139, 38873905, doi:10.1175/MWR-D-10-05013.1.

    • Search Google Scholar
    • Export Citation
  • Bodas-Salcedo, A., , K. D. Williams, , P. R. Field, , and A. P. Lock, 2012: The surface downwelling solar radiation surplus over the Southern Ocean in the Met Office model: The role of midlatitude cyclone clouds. J. Climate, 25, 74677486, doi:10.1175/JCLI-D-11-00702.1.

    • Search Google Scholar
    • Export Citation
  • Bodas-Salcedo, A., and et al. , 2014: Origins of the solar radiation biases over the Southern Ocean in CFMIP2 models. J. Climate, 27, 4156, doi:10.1175/JCLI-D-13-00169.1.

    • Search Google Scholar
    • Export Citation
  • Bracegirdle, T., , and E. Kolstad, 2010: Climatology and variability of Southern Hemisphere marine cold-air outbreaks. Tellus, 62A, 202208, doi:10.1111/j.1600-0870.2009.00431.x.

    • Search Google Scholar
    • Export Citation
  • Bromwich, D. H., , J. F. Carrasco, , and C. R. Stearns, 1992: Satellite observations of katabatic-wind propagation for great distances across the Ross Ice Shelf. Mon. Wea. Rev., 120, 19401949, doi:10.1175/1520-0493(1992)120<1940:SOOKWP>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brümmer, B., 1996: Boundary-layer modification in wintertime cold-air outbreaks from the Arctic sea ice. Bound.-Layer Meteor., 80, 109125, doi:10.1007/BF00119014.

    • Search Google Scholar
    • Export Citation
  • Brümmer, B., 1997: Boundary layer mass, water, and heat budgets in wintertime cold-air outbreaks from the Arctic sea ice. Mon. Wea. Rev., 125, 18241837, doi:10.1175/1520-0493(1997)125<1824:BLMWAH>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Brümmer, B., 1999: Roll and cell convection in wintertime Arctic cold-air outbreaks. J. Atmos. Sci., 56, 26132636, doi:10.1175/1520-0469(1999)056<2613:RACCIW>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Businger, S., , and R. J. Reed, 1989: Cyclogenesis in cold air masses. Wea. Forecasting, 4, 133156, doi:10.1175/1520-0434(1989)004<0133:CICAM>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Carleton, A. M., , and Y. Song, 1997: Synoptic climatology, and intrahemispheric associations, of cold air mesocyclones in the Australasian sector. J. Geophys. Res., 102, 13 87313 887, doi:10.1029/96JD03357.

    • Search Google Scholar
    • Export Citation
  • Coggins, J. H. J., , A. J. McDonald, , and B. Jolly, 2014: Synoptic climatology of the Ross Ice Shelf and Ross Sea region of Antarctica: k-means clustering and validation. Int. J. Climatol., 34, 23302348, doi:10.1002/joc.3842.

    • Search Google Scholar
    • Export Citation
  • Condron, A., , and I. A. Renfrew, 2013: The impact of polar mesoscale storms on northeast Atlantic Ocean circulation. Nat. Geosci., 6, 3437, doi:10.1038/ngeo1661.

    • Search Google Scholar
    • Export Citation
  • Condron, A., , G. R. Bigg, , and I. A. Renfrew, 2006: Polar mesoscale cyclones in the northeast Atlantic: Comparing climatologies from ERA-40 and satellite imagery. Mon. Wea. Rev., 134, 15181533, doi:10.1175/MWR3136.1.

    • Search Google Scholar
    • Export Citation
  • Cook, P. A., , and I. A. Renfrew, 2015: Aircraft-based observations of air-sea turbulent fluxes around the British Isles. Quart. J. Roy. Meteor. Soc., 141, 139152, doi:10.1002/qj.2345.

    • Search Google Scholar
    • Export Citation
  • Dee, D., and et al. , 2011: The ERA-Interim reanalysis: Configuration and performance of the data assimilation system. Quart. J. Roy. Meteor. Soc., 137, 553597, doi:10.1002/qj.828.

    • Search Google Scholar
    • Export Citation
  • Doms, G., and et al. , 2011: A description of the nonhydrostatic regional COSMO model. Part II: Physical parameterization. Consortium for Small-Scale Modelling Tech. Rep. LM_F90 4.20, 154 pp. [Available online at http://www.cosmo-model.org/content/model/documentation/core/cosmoPhysParamtr.pdf.]

  • Fairall, C. W., , E. F. Bradley, , D. P. Rogers, , J. B. Edson, , and G. S. Young, 1996: Bulk parameterization of air-sea fluxes for tropical ocean-global atmosphere coupled-ocean atmosphere response experiment. J. Geophys. Res., 101, 37473764, doi:10.1029/95JC03205.

    • Search Google Scholar
    • Export Citation
  • Fairall, C. W., , E. F. Bradley, , J. E. Hare, , A. A. Grachev, , and J. B. Edson, 2003: Bulk parameterization of air–sea fluxes: Updates and verification for the COARE algorithm. J. Climate, 16, 571591, doi:10.1175/1520-0442(2003)016<0571:BPOASF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Forbes, G. S., , and W. D. Lottes, 1985: Classification of mesoscale vortices in polar airstreams and the influence of the large-scale environment on their evolutions. Tellus, 37A, 132155, doi:10.1111/j.1600-0870.1985.tb00276.x.

    • Search Google Scholar
    • Export Citation
  • Green, J. S. A., 1960: A problem in baroclinic stability. Quart. J. Roy. Meteor. Soc., 86, 237251, doi:10.1002/qj.49708636813.

  • Green, J. S. A., 1979: Topics in dynamical meteorology: 8. Trough-ridge systems as slantwise convection. Weather, 34, 210, doi:10.1002/j.1477-8696.1979.tb03366.x.

    • Search Google Scholar
    • Export Citation
  • Grossman, R. L., , and A. K. Betts, 1990: Air–sea interaction during an extreme cold air outbreak from the eastern coast of the United States. Mon. Wea. Rev., 118, 324342, doi:10.1175/1520-0493(1990)118<0324:AIDAEC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Gryschka, M., , J. Fricke, , and S. Raasch, 2014: On the impact of forced roll convection on vertical turbulent transport in cold air outbreaks. J. Geophys. Res. Atmos., 119, 12 51312 532, doi:10.1002/2014JD022160.

    • Search Google Scholar
    • Export Citation
  • Hartmann, J., , C. Kottmeier, , and S. Raasch, 1997: Roll vortices and boundary-layer development during a cold air outbreak. Bound.-Layer Meteor., 84, 4565, doi:10.1023/A:1000392931768.

    • Search Google Scholar
    • Export Citation
  • Holton, J. R., , and G. J. Hakim, 2012: An Introduction to Dynamic Meteorology. Academic Press, 532 pp.

  • Hoskins, B. J., , M. E. McIntyre, , and A. W. Robertson, 1985: On the use and significance of isentropic potential vorticity maps. Quart. J. Roy. Meteor. Soc., 111, 877946, doi:10.1002/qj.49711147002.

    • Search Google Scholar
    • Export Citation
  • Irving, D., , I. Simmonds, , and K. Keay, 2010: Mesoscale cyclone activity over the ice-free Southern Ocean: 1999–2008. J. Climate, 23, 54045420, doi:10.1175/2010JCLI3628.1.

    • Search Google Scholar
    • Export Citation
  • Isachsen, P. E., , M. Drivdal, , S. Eastwood, , Y. Gusdal, , G. Noer, , and Ø. Saetra, 2013: Observations of the ocean response to cold air outbreaks and polar lows over the Nordic Seas. Geophys. Res. Lett., 40, 36673671, doi:10.1002/grl.50705.

    • Search Google Scholar
    • Export Citation
  • Iwasaki, T., , T. Shoji, , Y. Kanno, , M. Sawada, , M. Ujiie, , and K. Takaya, 2014: Isentropic analysis of polar cold airmass streams in the Northern Hemispheric winter. J. Atmos. Sci., 71, 22302243, doi:10.1175/JAS-D-13-058.1.

    • Search Google Scholar
    • Export Citation
  • Jensen, T. G., , T. J. Campbell, , R. A. Allard, , R. J. Small, , and T. A. Smith, 2011: Turbulent heat fluxes during an intense cold-air outbreak over the Kuroshio Extension Region: Results from a high-resolution coupled atmosphere–ocean model. Ocean Dyn., 61, 657674, doi:10.1007/s10236-011-0380-0.

    • Search Google Scholar
    • Export Citation
  • Kolstad, E. W., 2011: A global climatology of favourable conditions for polar lows. Quart. J. Roy. Meteor. Soc., 137, 17491761, doi:10.1002/qj.888.

    • Search Google Scholar
    • Export Citation
  • Kolstad, E. W., , and T. J. Bracegirdle, 2008: Marine cold-air outbreaks in the future: An assessment of IPCC AR4 model results for the Northern Hemisphere. Climate Dyn., 30, 871885, doi:10.1007/s00382-007-0331-0.

    • Search Google Scholar
    • Export Citation
  • Liu, Z., , and D. H. Bromwich, 1997: Dynamics of the katabatic wind confluence zone near Siple Coast, West Antarctica. J. Appl. Meteor., 36, 97118, doi:10.1175/1520-0450(1997)036<0097:DOTKWC>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Louis, J., 1979: A parametric model of vertical eddy fluxes in the atmosphere. Bound.-Layer Meteor., 17, 187202, doi:10.1007/BF00117978.

    • Search Google Scholar
    • Export Citation
  • Lüpkes, C., , and K. H. Schlünzen, 1996: Modelling the Arctic convective boundary-layer with different turbulence parameterizations. Bound.-Layer Meteor., 79, 107130, doi:10.1007/BF00120077.

    • Search Google Scholar
    • Export Citation
  • Nigro, M. A., , and J. J. Cassano, 2014: Identification of surface wind patterns over the Ross Ice Shelf, Antarctica, using self-organizing maps. Mon. Wea. Rev., 142, 23612378, doi:10.1175/MWR-D-13-00382.1.

    • Search Google Scholar
    • Export Citation
  • O’Connor, W. P., , D. H. Bromwich, , and J. F. Carrasco, 1994: Cyclonically forced barrier winds along the Transantarctic Mountains near Ross Island. Mon. Wea. Rev., 122, 137150, doi:10.1175/1520-0493(1994)122<0137:CFBWAT>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Pagowski, M., , and G. W. K. Moore, 2001: A numerical study of an extreme cold-air outbreak over the Labrador Sea: Sea ice, air–sea interaction, and development of polar lows. Mon. Wea. Rev., 129, 4772, doi:10.1175/1520-0493(2001)129<0047:ANSOAE>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Papritz, L., , S. Pfahl, , H. Sodemann, , and H. Wernli, 2015: A climatology of cold air outbreaks and their impact on air–sea heat fluxes in the high-latitude South Pacific. J. Climate, 28, 342364, doi:10.1175/JCLI-D-14-00482.1.

    • Search Google Scholar
    • Export Citation
  • Parish, T. R., , and J. J. Cassano, 2003a: Diagnosis of the katabatic wind influence on the wintertime Antarctic surface wind field from numerical simulations. Mon. Wea. Rev., 131, 11281139, doi:10.1175/1520-0493(2003)131<1128:DOTKWI>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Parish, T. R., , and J. J. Cassano, 2003b: The role of katabatic winds on the Antarctic surface wind regime. Mon. Wea. Rev., 131, 317333, doi:10.1175/1520-0493(2003)131<0317:TROKWO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Parish, T. R., , J. J. Cassano, , and M. W. Seefeldt, 2006: Characteristics of the Ross Ice Shelf air stream as depicted in Antarctic Mesoscale Prediction System simulations. J. Geophys. Res., 111, D12109, doi:10.1029/2005JD006185.

    • Search Google Scholar
    • Export Citation
  • Rasmussen, E., , and J. Turner, 2003: Polar Lows: Mesoscale Weather Systems in the Polar Regions. Cambridge University Press, 612 pp.

  • Renfrew, I. A., , and G. W. K. Moore, 1999: An extreme cold-air outbreak over the Labrador Sea: Roll vortices and air–sea interaction. Mon. Wea. Rev., 127, 23792394, doi:10.1175/1520-0493(1999)127<2379:AECAOO>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Schröder, D., , G. Heinemann, , and S. Willmes, 2011: The impact of a thermodynamic sea-ice module in the COSMO numerical weather prediction model on simulations for the Laptev Sea, Siberian Arctic. Polar Res., 30, 6334, doi:10.3402/polar.v30i0.6334.

    • Search Google Scholar
    • Export Citation
  • Shoji, T., , Y. Kanno, , T. Iwasaki, , and K. Takaya, 2014: An isentropic analysis of the temporal evolution of East Asian cold air outbreaks. J. Climate, 27, 93379348, doi:10.1175/JCLI-D-14-00307.1.

    • Search Google Scholar
    • Export Citation
  • Škerlak, B., , M. Sprenger, , and H. Wernli, 2014: A global climatology of stratosphere–troposphere exchange using the ERA-Interim data set from 1979 to 2011. Atmos. Chem. Phys., 14, 913937, doi:10.5194/acp-14-913-2014.

    • Search Google Scholar
    • Export Citation
  • Skyllingstad, E. D., , and J. B. Edson, 2009: Large-eddy simulation of moist convection during a cold air outbreak over the Gulf Stream. J. Atmos. Sci., 66, 12741293, doi:10.1175/2008JAS2755.1.

    • Search Google Scholar
    • Export Citation
  • Steppeler, J., , G. Doms, , U. Schättler, , H. W. Bitzer, , A. Gassmann, , U. Damrath, , and G. Gregoric, 2003: Meso-gamma scale forecasts using the nonhydrostatic model LM. Meteor. Atmos. Phys., 82, 7596, doi:10.1007/s00703-001-0592-9.

    • Search Google Scholar
    • Export Citation
  • Talley, L. D., 2008: Freshwater transport estimates and the global overturning circulation: Shallow, deep and throughflow components. Prog. Oceanogr., 78, 257303, doi:10.1016/j.pocean.2008.05.001.

    • Search Google Scholar
    • Export Citation
  • Terpstra, A., , T. Spengler, , and R. W. Moore, 2015: Idealised simulations of polar low development in an Arctic moist-baroclinic environment. Quart. J. Roy. Meteor. Soc., 141, 19871996, doi:10.1002/qj.2507.

    • Search Google Scholar
    • Export Citation
  • Tiedtke, M., 1989: A comprehensive mass flux scheme for cumulus parameterization in large-scale models. Mon. Wea. Rev., 117, 17791800, doi:10.1175/1520-0493(1989)117<1779:ACMFSF>2.0.CO;2.

    • Search Google Scholar
    • Export Citation
  • Vihma, T., , and B. Brümmer, 2002: Observations and modelling of the on-ice and off-ice air flow over the Northern Baltic Sea. Bound.-Layer Meteor., 103, 127, doi:10.1023/A:1014566530774.

    • Search Google Scholar
    • Export Citation
  • Wacker, U., , K. V. Jayaraman Potty, , C. Lüpkes, , J. Hartmann, , and M. Raschendorfer, 2005: A case study on a polar cold air outbreak over Fram Strait using a mesoscale weather prediction model. Bound.-Layer Meteor., 117, 301336, doi:10.1007/s10546-005-2189-1.

    • Search Google Scholar
    • Export Citation
  • Williams, K. D., and et al. , 2013: The Transpose-AMIP II experiment and its application to the understanding of Southern Ocean cloud biases in climate models. J. Climate, 26, 32583274, doi:10.1175/JCLI-D-12-00429.1.

    • Search Google Scholar
    • Export Citation
  • Yuan, X., , J. Patoux, , and C. Li, 2009: Satellite-based midlatitude cyclone statistics over the Southern Ocean: 2. Tracks and surface fluxes. J. Geophys. Res., 114, D04106, doi:10.1029/2008JD010874.

    • Search Google Scholar
    • Export Citation
Save