• Andrews, D. G., , and McIntyre M. E. , 1976: Planetary waves in horizontal and vertical shear: The generalized Eliassen-Palm relation and the mean zonal acceleration. J. Atmos. Sci., 33 , 20312048.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Andrews, D. G., , Holton J. R. , , and Leovy C. B. , 1987: Middle Atmosphere Dynamics. Academic Press, 489 pp.

  • Baldwin, M. P., , and Dunkerton T. J. , 1999: Propagation of the Arctic Oscillation from the stratosphere to the troposphere. J. Geophys. Res., 104 , 3093730946.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Baldwin, M. P., , and Dunkerton T. J. , 2001: Stratospheric harbingers of anomalous weather regimes. Science, 294 , 581584.

  • Baldwin, M. P., , Cheng X. , , and Dunkerton T. , 1989: The stratospheric major warming of early December 1987. J. Atmos. Sci., 46 , 28632884.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Boyd, J., 1976: The noninteraction of waves with the zonally averaged flow on a spherical earth and the interrelationships of eddy fluxes of energy, heat and momentum. J. Atmos. Sci., 33 , 22852291.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Charlton, A. J., , and Polvani L. M. , 2007: A new look at stratospheric sudden warmings. Part I: Climatology and modeling benchmarks. J. Climate, 20 , 449469.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Coy, L., , Eckermann S. , , and Hoppel K. , 2009: Planetary wave breaking and tropospheric forcing as seen in the stratospheric sudden warming of 2006. J. Atmos. Sci., 66 , 495507.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cummings, J. A., 2005: Operational multivariate ocean data assimilation. Quart. J. Roy. Meteor. Soc., 131 , 35833604.

  • Gerber, E. P., , Orbe C. , , and Polvani L. M. , 2009: Stratospheric influence on the tropospheric circulation revealed by idealized ensemble forecasts. Geophys. Res. Lett., 36 , L24801. doi:10.1029/2009GL040913.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hogan, T. F., , and Rosmond T. E. , 1991: The description of the Navy Operational Global Atmospheric Prediction System’s spectral forecast model. Mon. Wea. Rev., 119 , 17861815.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jung, J-H., , Konor C. S. , , Mechoso C. R. , , and Arakawa A. , 2001: A study of the stratospheric major warming and subsequent flow recovery during the winter of 1979 with an isentropic vertical coordinate model. J. Atmos. Sci., 58 , 26302649.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., 2007: Balance of drag between the middle and lower atmospheres in a global atmospheric forecast model. J. Geophys. Res., 112 , D13104. doi:10.1029/2007JD008647.

    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., , and Arakawa A. , 1995: Improvement of orographic gravity wave parameterization using a mesoscale gravity-wave model. J. Atmos. Sci., 52 , 18751902.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., , and Doyle J. D. , 2005: Extension of an orographic drag parameterization scheme to incorporate orographic anisotropy and flow blocking. Quart. J. Roy. Meteor. Soc., 131 , 18931921.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., , Campbell W. F. , , and Swadley S. D. , 2010: Reduction of middle-atmospheric forecast bias through improvement in satellite radiance quality control. Wea. Forecasting, 25 , 681700.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kodera, K., 2006: Influence of stratospheric sudden warming on the equatorial troposphere. Geophys. Res. Lett., 33 , L06804. doi:10.1029/2005GL024510.

    • Search Google Scholar
    • Export Citation
  • Kodera, K., , and Chiba M. , 1995: Role of planetary waves in the stratosphere-troposphere coupled variability in the Northern Hemisphere winter. J. Geophys. Res., 100 , 1105511068.

    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., 2005: On the influence of the meridional circulation and surface pressure change on the Arctic Oscillation. J. Geophys. Res., 110 , D21107. doi:10.1029/2004JD005743.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., 2008: Effect of stratospheric sudden warming and vortex intensification on the tropospheric climate. J. Geophys. Res., 113 , D15110. doi:10.1029/2007JD009550.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., , and Kodera K. , 2004: Role of polar-night jet oscillation on the formation of the Arctic Oscillation in the Northern Hemisphere winter. J. Geophys. Res., 109 , D11112. doi:10.1029/2003JD004123.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., , and Kodera K. , 2007: Correction to “Role of polar-night jet oscillation on the formation of the Arctic Oscillation in the northern hemisphere winter”. J. Geophys. Res., 112 , D08104. doi:10.1029/2007JD008571.

    • Search Google Scholar
    • Export Citation
  • Limpasuvan, V., , Thompson D. W. J. , , and Hartmann D. L. , 2004: The life cycle of the Northern Hemisphere sudden stratospheric warmings. J. Climate, 17 , 25842596.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Manney, G. L., and Coauthors, 2009: Aura Microwave Limb Sounder observations of dynamics and transport during the record-breaking 2009 Arctic stratospheric major warming. Geophys. Res. Lett., 36 , L12815. doi:10.1029/2009GL038586.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Matsuno, T., 1971: A dynamic model of the stratospheric sudden warming. J. Atmos. Sci., 28 , 14791492.

  • Mechoso, C. R., , Yamazaki K. , , Kitoh A. , , and Arakawa A. , 1985: Numerical forecasts of stratospheric warming events during the winter of 1979. Mon. Wea. Rev., 113 , 10151029.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mukougawa, H., , and Hirooka T. , 2004: Predictability of stratospheric sudden warming: A case study for 1998/99 winter. Mon. Wea. Rev., 112 , 17641776.

    • Search Google Scholar
    • Export Citation
  • Mukougawa, H., , Sakai H. , , and Hirooka T. , 2005: High sensitivity to the initial condition for the prediction of stratospheric sudden warming. Geophys. Res. Lett., 32 , L17806. doi:10.1029/2005GL022909.

    • Search Google Scholar
    • Export Citation
  • Nakagawa, K. I., , and Yamazaki K. , 2006: What kind of stratospheric sudden warming propagates to the troposphere? Geophys. Res. Lett., 33 , L04801. doi:10.1029/2005GL024784.

    • Search Google Scholar
    • Export Citation
  • Naujokat, B., , Krüger K. , , Matthes K. , , Hoffmann J. , , Kunze M. , , and Labitzke K. , 2002: The early major warming in December 2001 – Exceptional? Geophys. Res. Lett., 29 , 2023. doi:10.1029/2002GL015316.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Palmer, T., 1981: Diagnostic study of a wavenumber-2 stratospheric sudden warming in a transformed Eulerian-mean formalism. J. Atmos. Sci., 38 , 844855.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Polvani, L. M., , and Waugh D. W. , 2004: Upward wave activity flux as a precursor to extreme stratospheric events and subsequent anomalous surface weather regimes. J. Climate, 17 , 35483554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., , and Strüfing R. , 1983: Numerical forecasts of stratospheric warming events using a model with a hybrid vertical coordinate. Quart. J. Roy. Meteor. Soc., 109 , 81111.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., , and Wallace J. M. , 1998: The Arctic oscillation signature in the wintertime geopotential height and temperature fields. Geophys. Res. Lett., 25 , 12971300.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Webster, S., , Brown A. R. , , Cameron D. R. , , and Jones C. P. , 2003: Improvements to the representation of orography in the Met Office Unified Model. Quart. J. Roy. Meteor. Soc., 129 , 19892010.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • View in gallery

    Zonal mean zonal winds from the Met Office analysis on (a) 10, (b) 20, (c) 24, and (d) 30 Jan 2009. A major SSW occurred on 24 Jan 2009.

  • View in gallery

    As in Fig. 1, but for the zonal mean air temperatures.

  • View in gallery

    As in Fig. 1, but for the geopotential height at 10 hPa.

  • View in gallery

    The 14-day forecasts on 24 Jan 2009 of the zonal mean zonal winds with NOGAPS T239L42 forecast model (without data assimilation), which started on 10 Jan 2009, with orographic drag parameterization of (a) the control scheme by Webster et al. (2003), (b) the scheme by Kim and Arakawa (1995) and Kim and Doyle (2005) only, (c) both the Webster et al. scheme and the Kim and Arakawa and Kim and Doyle scheme, and (d) both schemes with the Webster et al. scheme limited to 100 hPa from the surface.

  • View in gallery

    As in Fig. 4, but for the zonal mean temperatures.

  • View in gallery

    As in Fig. 4, but for the geopotential height at 10 hPa.

  • View in gallery

    As in Fig. 4, but for the zonal mean wind tendency resulting from the zonal component of orographic drag, averaged over 10–31 Jan 2009. (a) Webster et al. (2003) drag from the experiment with the Webster et al. scheme only, (b) Kim drag from the experiment with the Kim and Arakawa and Kim and Doyle scheme only, (c) Webster et al. drag, and (d) Kim drag from the experiment with both the Webster et al. and Kim and Arakawa and Kim and Doyle schemes included, (e) Webster et al. drag, and (f) Kim drag from the experiment with both the Webster et al. and Kim and Arakawa and Kim and Doyle schemes included, but with the drag limited to 100 hPa from the surface for the Webster et al. scheme. Selected negative contours are drawn at −50, −100, −250, −500, and −1000 × 10−2 m s−1 day−1.

  • View in gallery

    The variations in forecast time of the (a) zonal mean zonal wind and (b) zonal mean temperature forecasts from UKMO analysis and various NOGAPS experiments at 10 hPa averaged over 65°–90°N. Additional experiments performed by varying the drag-limiting height (from the default value of 100 hPa) for the Webster et al. (2003) scheme and also varying the effective grid length that determines the magnitude of the reference-level GWD for the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme (thin curves).

  • View in gallery

    The time variations of the eddy heat fluxes averaged over 65°–90°N at (a) 100, (b) 10, and (c) 1 hPa.

  • View in gallery

    The differences in the zonally averaged and vertically integrated (from 10 to 100 hPa) torques resulting from grid-scale and subgrid-scale orography (mountain torque and orographic torque, respectively) between the modified run with the Webster et al. (2003) scheme (with the drag limited to 100 hPa) plus the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme and the control run with the Webster et al. scheme only.

  • View in gallery

    The time variations of the AO indices calculated over 20°–90°N at (a) 10 hPa, (b) 1000 hPa for 10–30 Jan 2009, and (c) at 1000 hPa for 10 Jan–21 Feb 2009.

  • View in gallery

    The time variations of the NAM indices calculated over 20°–90°N at all levels between the surface and 10 hPa for 10 Jan–21 Feb 2009 from the (a) UKMO analysis, (b) the NOGAPS control (CTL; Webster et al. 2003), and (c) NOGAPS modified [Webster et al. >100 hPa + Kim and Arakawa (1995) and Kim and Doyle (2005)] hindcast runs.

  • View in gallery

    As in Fig. 4, but for the time variations of the (a) zonal mean zonal wind at 10 hPa averaged over 65°–90°N, with forecasts started on 10, 15, and 20 Jan 2009 for each case.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 46 46 4
PDF Downloads 36 36 3

Hindcasting the January 2009 Arctic Sudden Stratospheric Warming and Its Influence on the Arctic Oscillation with Unified Parameterization of Orographic Drag in NOGAPS. Part I: Extended-Range Stand-Alone Forecast

View More View Less
  • 1 Marine Meteorology Division, Naval Research Laboratory, Monterey, California
© Get Permissions
Full access

Abstract

A very strong Arctic major sudden stratospheric warming (SSW) event occurred in late January 2009. The stratospheric temperature climbed abruptly and the zonal winds reversed direction, completely splitting the polar stratospheric vortex. A hindcast of this event is attempted by using the Navy Operational Global Atmospheric Prediction System (NOGAPS), which includes the full stratosphere with its top at around 65 km. As Part I of this study, extended-range (3 week) forecast experiments are performed using NOGAPS without the aid of data assimilation. A unified parameterization of orographic drag is designed by combining two parameterization schemes; one by Webster et al., and the other by Kim and Arakawa and Kim and Doyle. With the new unified orographic drag scheme implemented, NOGAPS is able to reproduce the salient features of this Arctic SSW event owing to enhanced planetary wave activity induced by more comprehensive subgrid-scale orographic drag processes. The impact of the SSW on the tropospheric circulation is also investigated in view of the Arctic Oscillation (AO) index, which calculated using 1000-hPa geopotential height. The NOGAPS with upgraded orographic drag physics better simulates the trend of the AO index as verified by the Met Office analysis, demonstrating its improved stratosphere–troposphere coupling. It is argued that the new model is more suitable for forecasting SSW events in the future and can serve as a tool for studying various stratospheric phenomena.

Corresponding author address: Dr. Young-Joon Kim, Stop 2, Marine Meteorology Division, Naval Research Laboratory, Monterey, CA 93943. Email: yj.kim@nrlmry.navy.mil

Abstract

A very strong Arctic major sudden stratospheric warming (SSW) event occurred in late January 2009. The stratospheric temperature climbed abruptly and the zonal winds reversed direction, completely splitting the polar stratospheric vortex. A hindcast of this event is attempted by using the Navy Operational Global Atmospheric Prediction System (NOGAPS), which includes the full stratosphere with its top at around 65 km. As Part I of this study, extended-range (3 week) forecast experiments are performed using NOGAPS without the aid of data assimilation. A unified parameterization of orographic drag is designed by combining two parameterization schemes; one by Webster et al., and the other by Kim and Arakawa and Kim and Doyle. With the new unified orographic drag scheme implemented, NOGAPS is able to reproduce the salient features of this Arctic SSW event owing to enhanced planetary wave activity induced by more comprehensive subgrid-scale orographic drag processes. The impact of the SSW on the tropospheric circulation is also investigated in view of the Arctic Oscillation (AO) index, which calculated using 1000-hPa geopotential height. The NOGAPS with upgraded orographic drag physics better simulates the trend of the AO index as verified by the Met Office analysis, demonstrating its improved stratosphere–troposphere coupling. It is argued that the new model is more suitable for forecasting SSW events in the future and can serve as a tool for studying various stratospheric phenomena.

Corresponding author address: Dr. Young-Joon Kim, Stop 2, Marine Meteorology Division, Naval Research Laboratory, Monterey, CA 93943. Email: yj.kim@nrlmry.navy.mil

1. Introduction

Sudden stratospheric warming (SSW) is an abrupt disruption of the stratospheric winter circulation involving a rapid breakdown of the polar vortex. SSW is triggered by anomalous planetary (Rossby) wave activity propagating from the troposphere, and it is characterized by a rapid increase of the polar stratospheric temperature (about 40 K in a week) and a reversal of the stratospheric polar night jet. An SSW event is called “major” if the 10-hPa (or below in height) zonal mean temperature gradients reverse poleward of 60°N, as used by Manney et al. (2009), and “minor” if there is no reversal. It can also be classified into either a vortex “displacement” or “split” event (Charlton and Polvani 2007) usually based on 10-hPa geopotential height. According to these criteria, a major SSW vortex split event occurred on 24 January 2009 (Manney et al. 2009), the model hindcast of which will be the subject of this study.

The primary cause of SSWs is identified as the breaking of upward-propagating planetary waves originated from the troposphere (Andrews et al. 1987) that are dynamically forced by orography and land–sea temperature contrast. Matsuno (1971) introduced the basic mechanism of SSWs based upon a mechanistic model in terms of the interaction between large-amplitude planetary waves propagating from the troposphere and zonal-mean flows in the stratosphere. Successful simulation or prediction of SSW events is a good indicator of adequate representation of the stratosphere–troposphere dynamical coupling in the atmospheric model in terms of the model’s capability to capture both forcing and propagation.

The onset of the January 2009 major SSW event when it occurred was not simulated by the operational version of the Navy Operational Global Atmospheric Prediction System (NOGAPS; Hogan and Rosmond 1991), which now has the model top at around 65 km or 0.04 hPa. For this study, the model was integrated in the hindcast mode for an extended range (3 weeks or longer) without data assimilation. As an effort to improve the simulation of the SSW event, this study focuses on the parameterization of orographic processes (the subgrid-scale portion) in terms of orographic drag. We introduce a combined (unified) parameterization scheme based on two parameterizations—the scheme by Webster et al. (2003), and the scheme by Kim and Arakawa (1995) and Kim and Doyle (2005).

It is known that the effect of SSWs propagates down to the troposphere and surface within a few weeks and can last longer than 2 months (Baldwin and Dunkerton 2001). This effect is often measured by an index of the Arctic Oscillation (AO; Thompson and Wallace 1998), which describes the dominant pattern of nonseasonal sea level pressure variations north of 20°N. It has been reported that the surface pressure signal shifts to a negative pattern of the AO after SSW events (Limpasuvan et al. 2004). However, the negative phase of the AO, which signals a weak tropospheric polar vortex, does not necessarily occur after all SSW events (Nakagawa and Yamazaki 2006), that is, not all SSW events propagate into the troposphere. Nakagawa and Yamazaki (2006) classified SSW events into those that propagate into the troposphere and those that do not, depending, respectively, on whether the polar vortex splits or not. Gerber et al. (2009) observed that a sharp reversal of the 10-hPa westerly jet does not guarantee deep penetration through the stratosphere into the troposphere. The present study will determine into which category the 2009 January major SSW event fits by calculating the AO index (Thompson and Wallace 1998) for the SSW event.

SSWs have often been analyzed using long-term (e.g., 30 day) averaged fields. However, as Kuroda (2008) pointed out using his earlier studies of SSWs in association with the AO (e.g., Kuroda and Kodera 2004, 2007; Kuroda 2005), this may only address their slower variability and may not be suitable for shorter time scales. Kuroda (2008) used daily reanalysis data to analyze SSWs and their impact on the tropospheric climate. In the present study, we focus on the impact of the SSW on the stratospheric and tropospheric forecast in shorter time scales. Therefore, we use unfiltered daily data with an assumption that the time scales associated with the SSW event can be quite short, involving relatively fast interactions between the stratosphere and troposphere.

The primary goal of this study is to improve the NOGAPS simulation of SSW events in the extended-range hindcast mode without using data assimilation. We achieve that by improving the parameterization of orographic drag, that is, gravity wave drag (GWD) resulting from the breaking of upward-propagating subgrid-scale gravity waves plus blocked-layer drag (BLD) resulting from the blocking of horizontal flow by subgrid-scale orography (see Kim and Doyle 2005, for more information). We also investigate the downward control effect of SSWs on the troposphere with the aid of the AO index. Section 2 introduces the details of the experimental design. Section 3 presents and discusses the results. Section 4 concludes the study with a summary and further remarks. Hindcast experiments with the full NOGAPS, including the activated data assimilation, will be reported as Part II of this study in a forthcoming companion paper (Y.-J. Kim et al. 2010, unpublished manuscript).

2. Experimental setup

This study utilizes the forecast model component of the operational NOGAPS with its top at around 0.04 hPa (∼65 km) without the data assimilation component, in order to evaluate the genuine forecast capability of the model. The horizontal and vertical resolutions of the forecast model are T2391 (triangular spectral truncation at wavenumber 239, or ∼0.5° in latitude) and L42 (42 terrain-following levels), respectively. The operational model includes an orographic drag parameterization scheme based on Webster et al. (2003), which is a hybrid scheme between BLD at low levels and GWD at upper levels. The sum of these components is prescribed according to the linear hydrostatic single-barrier mountain-wave theory. This scheme steadily generates very large BLD near the surface and relatively small GWD at upper levels. Another orographic drag parameterization package is activated in the model for this study, parameterizing GWD based on Kim and Arakawa (1995) and BLD based on Kim and Doyle (2005), in which GWD and BLD are treated separately based on nonlinear nonhydrostatic multibarrier mountain-wave simulations. This scheme generates relatively small BLD at low levels and relatively large GWD at upper levels. This study will test each of these schemes, both separately and in combination, as an effort to better hindcast the January 2009 SSW event.

The model runs start at 0000 UTC 10 January 2009, 2 weeks before the onset of the major SSW event, and stop at 1800 UTC 31 January 2009, except for selected runs that continued until 21 February 2009. This startup date was chosen well before the SSW event in order to lessen the influence of the initial conditions. As atmospheric initial conditions we use 42-level NOGAPS test reanalysis fields, while as surface boundary conditions we use 12-h-analyzed sea surface temperature (SST) and ice data obtained from a univariate two-dimensional OI (optimal interpolation) scheme (Cummings 2005); the SST data are from Advanced Very High Resolution Radiometer (AVHRR) imagers onboard various satellites and in situ ship and fixed and drifting buoys, while the sea ice data are from Special Sensor Microwave Imager (SSM/I) and SSM/I/Sounder (SSMI/S) instruments on board the Defense Meteorological Satellite Program (DMSP) satellites. These analyzed boundary conditions provide the main, strong constraint on the extended-range hindcast.

The control experiment uses the operational Webster et al. (2003) orographic drag parameterization, and the test experiments use either the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme only, the Webster et al. scheme plus the Kim and Arakawa and Kim and Doyle scheme, or the Webster et al. scheme limited to 100 hPa (from surface) plus the Kim and Arakawa and Kim and Doyle scheme for the entire model atmosphere. The details of the experiments are summarized in Table 1. The experiments are compared with the Met Office (UKMO) analyses for verification (during the 2009 January SSW event NOGAPS covered only up to 50 km in height, which was raised to 65 km in September 2009).

The AO index is calculated based on Thompson and Wallace (1998) as a projection of the daily 1000-hPa geopotential height anomalies poleward of 20°N onto the loading pattern of AO defined as the leading mode of the long-term geopotential height data. This method is used by the National Oceanic and Atmospheric Administration (NOAA)/Climate Prediction Center. The first empirical orthogonal function (EOF) of the 1000-hPa geopotential height is calculated utilizing the 1979–2000 National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) reanalysis data. We calculate the AO index using 1000-hPa heights from the UKMO analysis or the NOGAPS runs. The mixed use of observation and model datasets in calculating the index may introduce some inconsistency. Our sample test, however, indicates that the choice of NOGAPS or NCEP–NCAR analysis data virtually has no influence on the value of the AO index at 1000 hPa, at least for this particular case. A similar procedure is used to calculate the index at other model levels; the model geopotential height anomaly is projected onto the first EOF of the monthly geopotential height from the NCEP–NCAR reanalysis data.

3. Results and discussion

Figure 1 reveals the evolution of the stratospheric Arctic polar night jet in January 2009 as depicted in the UKMO analysis. The usual, strong stratospheric polar night jet in the Northern Hemisphere is present initially (10 January 2009; Fig. 1a), but it starts being decelerated (20 January 2009; Fig. 1b), reverses entirely to easterlies (24 January 2009; Fig. 1c), and then starts recovering back to westerlies (30 January 2009; Fig. 1d). Figure 2 displays the corresponding temperatures. The warming of the Arctic stratosphere is clearly seen on 24 January 2009 (Fig. 2c). It is noted that abnormal warming is found also in the Arctic mesosphere (around 1 hPa; Fig. 2b), earlier than that in the stratosphere (around 3 hPa; Fig. 2c), but this is not a subject of this study. Figure 3 shows the evolution of the Arctic polar vortex in terms of the geopotential height at 10 hPa, which is completely split on 24 January 2009 (Fig. 3c), consistently with the report of Manney et al. (2009).

The control extended-range hindcast using the operational NOGAPS with the Webster et al. (2003) orographic drag scheme detects a sign of the 24 January SSW signal above 1 hPa (Fig. 4a) in terms of the zonal mean zonal winds, but the reversal does not extend down to around the 10-hPa level, as found in the UKMO analysis (Fig. 1c); instead, a strong spurious polar westerly jet is present centered around 10 hPa. When the Kim scheme (Kim and Arakawa 1995; Kim and Doyle 2005) is used alone (Fig. 4b), the region of the wind reversal extends somewhat farther down in the middle latitudes compared with the control case, and the spurious westerly jet is located lower and is weaker (i.e., less erratic), but the magnitudes of the easterlies are quite weak. When the two schemes are used together, the extent of the wind reversal is improved (although still rather weak) and the spurious polar westerly jet almost disappears (Fig. 4c). Further improvement is achieved when the drag from the Webster et al. scheme is limited to 100 hPa from surface (Fig. 4d) in that the magnitudes of reversed winds are stronger and much closer to the analysis shown in Fig. 1c. The corresponding temperatures (Fig. 5) also show that the SSW event is best simulated with the last configuration (Fig. 5d) compared with the analysis (Fig. 2c). The improvement of the SSW event with the combined drag parameterization is also seen in terms of the polar vortex split. Figure 6 compares the simulated 10-hPa geopotential height and reveals that the combined parameterization with the drag height limit, unlike other cases (Figs. 6a–c), successfully simulates the vortex split (Fig. 6d).

Figure 7 compares different configurations of the parameterized orographic drag in terms of the tendency of the zonal mean zonal wind by NOGAPS. Figure 7a displays orographic drag (BLD near the surface and the residual GWD above) from the Webster et al. (2003) scheme, and Fig. 7b displays GWD at upper levels and BLD at low levels, separately calculated by the two parts of the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme, when each scheme is used alone. The negative minimum value from the Kim and Arakawa and Kim and Doyle scheme (Fig. 7b) is about 2 times larger than that from the Webster et al. scheme (Fig. 7a), and the average value is almost 6 times larger in magnitude. Figures 7c,d display the drag from each of the two schemes when they are used together: Fig. 7c is the drag from the Webster et al. scheme, while Fig. 7d is that from the Kim and Arakawa and Kim and Doyle scheme. The minimum value from the Webster et al. scheme greatly decreases (nearly to the half) when the Webster et al. scheme is used together with the Kim and Arakawa and Kim and Doyle scheme (cf. Figs. 7a,c) and the minimum value from the Kim and Arakawa and Kim and Doyle scheme also decreases significantly (cf. Figs. 7b,d). When the two schemes are used together, if the drag from the Webster et al. scheme is limited to 100 hPa, then the minimum value from the Kim and Arakawa and Kim and Doyle scheme increases back (cf. Fig. 7b,f) because the Webster et al. drag is significantly reduced because of the height limit (Figs. 7e,f). This redistribution of drag magnitude is due to the model’s self-adjustment process, which preserves the total angular momentum as discussed by Kim (2007).

We now look into the time evolution of the SSW event. Figure 8 compares among the experiments the temporal variations of the zonal mean zonal winds (Fig. 8a) and temperature (Fig. 8b) at 10 hPa averaged over 65°–90°N, which is used as a criteria for a major SSW. In comparison with the UKMO analysis, the control NOGAPS run with the Webster et al. (2003) scheme fails to reproduce the SSW-induced wind change on 24 January 2009 (Fig. 8a). The runs including the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme successfully simulate the onset of the SSW, but unlike the analysis, they tend to return to non-SSW states after the onset. The magnitude of the averaged wind after the SSW onset depends on whether the Kim and Arakawa and Kim and Doyle scheme is used together with the Webster et al. scheme, with or without employing the drag height limit. This may imply the limit in the model predictability of about 2 weeks, or sensitivity to other factors. The sharp rise of some curves after the SSW onset is partly due to the fact that the wind reversal marginally extends to 10 hPa from above (cf. Fig. 1c with Figs. 4b–d). The thin curves on Fig. 8a are from additional experiments, performed by varying the drag-limiting height (from the default value of 100 hPa) for the Webster et al. scheme and the effective grid length that determines the magnitude of the reference-level GWD for the Kim and Arakawa and Kim and Doyle scheme (Kim and Doyle 2005). (It is noticeable that some of these thin curves are overall closer to the analysis than the selected cases, but their corresponding 10-hPa vortex splits were not as good as the selected case.) Regardless of those parameter values, whose variations provide some stochasticity of the experiments (their ensemble mean is similar to the combined case without the drag limit), it is clear that the combined scheme experiments distinctively capture the SSW signal in terms of the averaged high-latitude 10-hPa zonal mean zonal winds. As for the temperature (Fig. 8b), the control run severely underestimates the magnitude and onset time of the warming while other experiments—with the Kim and Arakawa and Kim and Doyle scheme alone or two schemes combined—simulate them relatively well. The run with the Kim and Arakawa and Kim and Doyle scheme plus the Webster et al. scheme limited to 100 hPa shows a slower return to a non-warming state after the SSW onset.

The difference between these experiments can be examined further in view of the vertical propagation of planetary waves represented by the eddy heat flux, which is proportional to the vertical component of the Eliassen–Palm flux (Andrews and McIntyre 1976; Boyd 1976). Mechoso et al. (1985) discussed an important role of the wave-mean flow interaction in the upper troposphere and lower stratosphere during SSW events in promoting upward propagation of planetary waves. Figure 9 compares the eddy heat fluxes at 100 (Fig. 9a), 10 (Fig. 9b), and 1 hPa (Fig. 9c) averaged over the high latitudes. Note that unlike, for example, Polvani and Waugh (2004), who showed averages of the prior 40 days of the data, we present instant daily values, and thus our results fluctuate largely in time. Figure 9a reveals that the eddy heat fluxes at 100 hPa, which is assumed as the approximate wave source level, peak around 20 January preceding the SSW event that occurred on 24 January. This rapid increase of upward planetary wave activity before the onset of the SSW is found in all runs (although it is rather weak in the control run). This agrees with previous studies (e.g., Palmer 1981; Baldwin et al. 1989; Kodera and Chiba 1995; Naujokat et al. 2002; Polvani and Waugh 2004). At the SSW onset on 24 January, the analysis shows a secondary peak, which indicates that the planetary wave activity increases again when the SSW occurs at higher levels (e.g., 10 hPa). After the SSW onset, the model heat fluxes deviate significantly from the analysis and varies widely, presumably resulting from the complex interactions of upward- and downward-propagating (reflected) waves and their downward impact on the troposphere. This interaction is not properly represented in the model, and may demonstrate lack of the model’s predictability at this extended time range.

At 10 hPa (Fig. 9b), the UKMO analysis shows that the heat flux peaks around 22 January and rapidly decreases toward negative values around 25–26 January. Comparing the peak fluxes at 100 (Fig. 9a) and 10 hPa (Fig. 9b), it seems that it takes a couple of days for the enhanced planetary waves to propagate from 100 to 10 hPa while maintaining the wave activity pattern between the two levels. The SSW onset in terms of rapid decrease of the flux at 10 hPa is captured in all of the experiments (Fig. 9b), although the control run with the Webster et al. (2003) scheme and also the combined run with the drag limit show their maxima on 23 January. The flux from the analysis changes its sign during 25–26 January, but the fluxes from the combined runs do that 1 or 2 days earlier or later than the analysis counterpart and the flux from the control run decreases after the SSW onset too slowly and never becomes negative. The variation after the SSW onset is quite complicated at 10 hPa as well, because of the complex interactions between the waves propagated from below and those that break near 10 hPa.

Furthermore, at 1 hPa (Fig. 9c), which is above the SSW region, the flux is severely overestimated by the control run and is moderately underestimated by other runs. The peak flux values lag by 2 or 3 days in all of the experiments. Under the assumption of stationary westerlies, the overestimated control run flux is consistent with the underestimated flux at lower levels (Figs. 9a,b), implying that the planetary waves, which did not break at the SSW level of 10 hPa (Fig. 9b), continue to propagate upward. Other runs suggest that the planetary wave breaking was somewhat excessive at lower levels or some of the waves are reflected down, resulting in underestimated flux at 1 hPa. In reality, however, it is more likely that the overestimated control run flux (Fig. 9c) is primarily due to the fact that planetary waves can propagate upward more actively through the stronger predicted westerlies in the stratosphere for the control run (Fig. 4a).

The 10-hPa heat fluxes (Fig. 9b) before the SSW onset (18–23 January) are somewhat underestimated by the NOGAPS runs (especially the control run). This implies that the planetary wave activity (propagation and breaking) between 100 and 10 hPa is somehow improperly represented in NOGAPS, especially in the control run. Because the only difference among these runs is the orographic drag parameterization, it is reasonable to assume that the parameterized drag (resulting from subgrid-scale wave breaking and/or flow-blocking mechanisms) is responsible for this discrepancy. We thus investigate next the change in grid-scale part of the wave spectrum (planetary waves) induced by the changes in parameterized subgrid-scale (i.e., gravity) waves.

The impact of orographic drag parameterization can be further investigated by comparing the model momentum balance among the drag mechanisms, that is, orographic GWD, convective GWD, mountain drag, friction drag, form drag, and BLD, etc. (Kim 2007). Here, we limit our vertical domain to 100–10 hPa and eliminate the drag imposed on or near the surface. Figure 10 shows the differences in the zonally averaged and vertically integrated torque between the modified run with the Webster et al. (2003) scheme (with the drag limited to 100 hPa) plus the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme, and the control run with the Webster et al. scheme only. In this pressure (height) range, the modified run excludes BLD because of the drag height limit (lower in height than 100 hPa; see Fig. 7e), while the control run by default includes no GWD from the Kim and Arakawa and Kim and Doyle scheme. The result is noteworthy in that the relatively small change in subgrid-scale orographic drag (or its angular version, torque) causes a significant change in the grid-scale mountain drag (torque; see Kim 2007 for relevant discussions). This result can be interpreted that increased (improved) orographic drag results in enhanced mountain drag in the Northern Hemisphere (compensated by reduced mountain drag in the tropics; see Fig. 10), which results in enhanced planetary wave activity in the Northern Hemisphere. This process involving the interaction between grid-scale (resolved planetary wave activity) and subgrid-scale (parameterized orographic drag) processes eventually helped the model better hindcast the SSW event.

It is well documented that a large change in the polar vortex strength can propagate into the troposphere within a few weeks (Baldwin and Dunkerton 1999, 2001). This effect is progressively becoming more important in weather forecasting as well as in climate simulation, because the model tops are gradually raised to include more of the middle atmosphere and the temporal forecast ranges are being extended. Baldwin and Dunkerton (1999, 2001) demonstrated the downward propagation of the AO [more generally referred to as northern annular mode (NAM)] signature from the 10-hPa level to the surface on a time scale of about 3 weeks. Polvani and Waugh (2004) discussed that AO index calculated at 10 hPa is a very good proxy for the strength of the polar night jet in the stratosphere. Figure 11a shows our calculated AO index for 10 hPa, which indeed has a strong resemblance to the high-latitude averaged 10-hPa zonal mean zonal winds (Fig. 8a), confirming the results of Polvani and Waugh (2004). It is of interest whether the SSW occurrence corresponds to a large negative deviation in surface AO index time series, and how well our model simulates this process.

We thus investigate the influence of the SSW on the surface in terms of the AO index for 1000-hPa geopotential heights. The AO index calculated using the UKMO analysis shown in Fig. 11b indicates that the index fluctuates over the hindcast period with the local minima on 15, 21–22, and 30 January, while gradually approaching zero. Our model hindcasts are very consistent with one another and closely match the analysis until around 16 January when they start diverging. The model runs basically follow the fluctuation trend/cycle of the analysis, but the index values do not seem to decrease toward being negative over time, as in the analysis. The combined case [with the Webster et al. (2003) scheme limited to 100 hPa], however, seems to follow the fluctuation of the analyzed index most closely. To check the further trend of the AO index, we extended selected runs by another 21 days and compared with the UKMO analysis counterpart (Fig. 11c). The AO index from the analysis indeed becomes negative beyond the first 21 days [consistent with the 3-week time reported by Baldwin and Dunkerton (1999, 2001)]. This result also confirms those of Kuroda (2008) that AO becomes negative following SSW events. The hindcast using the combined scheme with the drag limit roughly follows the index from the analysis and reaches negative values, whereas the control run does not follow it and never becomes negative. Furthermore, Fig. 12 compares the AO (NAM) indices corresponding to Fig. 11c, but calculated for all levels (up to 10 hPa, where climatology data are available). Unlike the control run (Fig. 12b), the modified run (Fig. 12c) overall reproduces the negative values of the index calculated from the analysis (Fig. 12a) well, which start developing near the top of the domain and are persistent throughout the 42-day run period. An exception is an area of erroneous positive values, centered around 40 hPa and 1 February 2009 (Fig. 12c), probably associated with the complex nature of the vertical wave propagation after the SSW onset, as discussed above, which the model fails to represent properly. These results, however, also clearly demonstrate the benefit of the combined parameterization in terms of the stratosphere–troposphere coupling.

4. Summary and further remarks

This study reports our efforts to improve the hindcast of the Arctic major sudden stratospheric warming (SSW) event that occurred in late January 2009 and its surface influence using the Navy Operational Global Atmospheric Prediction System (NOGAPS) with its top at around 65 km. A unified parameterization scheme of orographic drag has been designed by combining two schemes: one based on Webster et al. (2003) and the other based on Kim and Arakawa (1995) and Kim and Doyle (2005). With the unified orographic drag scheme, NOGAPS better reproduces the essential features of the 2009 January Arctic SSW event as verified by the Met Office analysis. The impact of the SSW event on the tropospheric circulation has also been investigated in view of the Arctic Oscillation (AO) index. The results demonstrate that the unified parameterization improves the downward propagation of the SSW signal as well. The present study confirms that the January 2009 SSW was a vortex-splitting event (Fig. 6) that reached the surface (relatively rapidly) as represented by negative AO index (Figs. 11c and 12). We believe that NOGAPS with the unified orographic drag parameterization is better suited for extended-range forecasts for which the stratosphere–troposphere coupling is important.

To improve the orographic drag parameterization used in NOGAPS, attempts had been made earlier to calibrate each parameterization scheme before considering their combination or unification. The orographic drag scheme by Webster et al. (2003) assumes that the total drag is determined by the linear hydrostatic mountain-wave theory for an isolated single mountain. The scheme calculates BLD near the surface based on an ad hoc assumption and GWD is assumed to be the residual of the BLD from the drag specified by the theory so that the sum of BLD and GWD is constant. This makes the calibration of the GWD part strongly dependent on that of the BLD part. Because the BLD part of the scheme had been extensively calibrated for operational short-range tropospheric weather forecast, its change strongly affects the forecast skill whether or not it improves the upper-level GWD and corresponding upper-level forecast. On the other hand, the Kim and Arakawa and Kim and Doyle scheme uses separate algorithms for BLD (Kim and Doyle 2005) and GWD (Kim and Arakawa 1995), which do not depend on each other, although they naturally affect each other by changing the background flow. Therefore, the two parts of the Kim and Arakawa and Kim and Doyle scheme can be separately calibrated to improve the lower troposphere and stratosphere, although they are dynamically and physically coupled. (Figure 7d shows the parameterized BLD near the surface and the parameterized GWD above the BLD area.) It has been found to be beneficial to include these two drag components from both the Webster et al. and Kim and Arakawa and Kim and Doyle schemes at least for improving the hindcast of the 2009 January SSW event discussed in this study. It was rather interesting that the two schemes collectively produced an optimal amount of drag, especially at upper levels, while each scheme could not be calibrated to provide adequate drag without degradation in some other regions and/or levels. The apparent reason is that the responses of the two schemes to the same atmospheric condition are somehow different and their sum is nonlinear. It needs to be investigated further how exactly this combination of the two schemes results in alleviation of the deficiencies of each scheme and an overall improvement of the SSW hindcast.

Although not explicitly, the results of this study address the predictability of SSWs. Simmons and Strüfing (1983) explored the predictability of SSW by performing 10-day forecasts of the 1979 winter SSW using the European Centre for Medium-Range Weather Forecast (ECMWF) operational model and found both predictability of a wavenumber-2 warming event up to 10 days ahead as well as a preceding development of wavenumber 1 and the subsequent decay of the wavenumber-2 perturbation. Mechoso et al. (1985) also performed 10-day forecasts of the 1979 SSW using the University of California Los Angeles (UCLA) general circulation model (GCM) and reported that SSWs are predictable several days in advance. Mechoso et al. reported that success of SSW forecasts depends on initial conditions; the predictability of the SSWs that were initialized 2 days apart was quite different from each other. To explore this sensitivity to initial conditions, we performed additional experiments by starting the forecast 5 and 10 days later than the default time of 10 January 2009. The results (Fig. 13) revealed that when the forecast was started on 15 January, the SSW was not simulated accurately in all cases (in terms of the stratospheric jet reversal). However, the combined scheme experiments still showed far better forecasts than the control. When the forecast was started on 20 January (i.e., 4 days before the SSW onset), all of the runs showed excellent predictability. Because the predictability of the SSW event in our hindcast is better with a 10 January start than a 15 January start, it is not merely the forecast range, but more likely the characteristics of the initial and/or boundary conditions and their impact on the wave generation and propagation that determine the predictability. Mechoso et al. (1985) argued that relatively small errors in the predicted tropospheric zonal mean wind can produce large differences in the characteristics of upward wave propagation affecting stratospheric forecast. Mukougawa et al. (2005) found that a distinct tropospheric zonal mean zonal wind profile during the onset stage of the SSW is significantly related to the subsequent SSW. Coy et al. (2009) also pointed out the importance of an accurate forecast of the upper-tropospheric disturbances for forecasting a major SSW event. Investigation of this sensitivity of the January 2009 SSW to initial conditions is the subject of further study.

Mukougawa and Hirooka (2004) investigated the predictability of the December 1998 SSW using an operational extended-range (1-month) numerical prediction model and reported that the event is predictable about 1 month in advance. They attributed the difference in the predictability of SSWs to the type of the SSW: whether the SSW is caused by wavenumber-1 (as in the 1979 SSW case) or wavenumber-2 (as in the 1998 SSW case) planetary waves. They discussed that an SSW caused by the amplification of wavenumber-2 planetary waves is associated with a shorter time scale than the wavenumber-1 case, and thus will affect shorter-range prediction. Mechoso et al. (1985) presented results from the UCLA GCM forecasts showing that minor warming is associated with a large amplification of wavenumber-1 wave, whereas major warming is associated with that of the wavenumber-2 wave. Jung et al. (2001) found that the amplified wave 1 is responsible for the acceleration of the mean flow in the recovery stage of the SSW. Our results, which can be deduced from Fig. 3, also reveal strong development of wavenumber 2 in the stratosphere. Based on the findings of the previous studies and the results of the present study, therefore, the January 2009 SSW was due to the amplification of wavenumber-2 waves.

Kodera (2006) discussed the influence of SSWs on equatorial tropospheric convective activity. His composite analysis of the SSWs from 1979 to 2001 revealed that the meridional circulation changes associated with SSWs lower the temperature in the equatorial lower stratosphere and upper troposphere, causing a seesaw pattern of convective activity in the troposphere. Kuroda (2008), using composite daily observation data, found that tropospheric climate is significantly changed in association with SSWs from the polar area to the tropical area. He argued that such a structure is maintained by the Eulerian meridional circulation, forced by the eddy forcing in the polar troposphere and stratosphere. The meridional circulation triggered by the upward-propagating planetary wave in the tropics induces positive feedback between the convection and meridional circulation, which creates tropical circulation. Coy et al. (2009) discussed the poleward propagation of negative potential vorticity anomaly from the tropics, associated with the January 2006 major SSW. Figure 10 indicates a fairly large fluctuation (reduction) in the mountain drag in the tropics, in response to the change (gain) in the orographic drag in the high and middle latitudes. This may be a consequence of the SSW-induced meridional circulation connecting the tropics and subtropics. We investigated the meridional circulation patterns in our experiments and found some changes in the tropics (not shown), but they were not systematic enough to support or deny the aforementioned arguments. Investigation of other SSW cases could verify this remote influence of SSWs.

Finally, as we stated earlier, this study is Part I of our efforts to improve the hindcast of SSWs, which utilized stand-alone global atmospheric forecast model. We plan to repeat the experiments with the full NOGAPS system, including the data assimilation. Kim et al. (2010) argued that an improvement in the middle-atmospheric physics should accompany an equivalent improvement in data assimilation via the quality control of satellite data (in particular, radiances). In NOGAPS, the difference between the observations and model background (i.e., innovations) is currently limited by prescribed values, before being fed into the data assimilation procedures. Therefore, any deficiency in the forecast model (i.e., in the physics such as middle-atmospheric GWD) can create a significant deviation of unrealistic model background from realistic observations beyond the prescribed limit (i.e., too large innovations). This causes good observation data to be rejected during the quality control process and the eventual degradation of the forecast skill. Kim et al. (2010) suggested the use of improved middle-atmospheric orographic drag parameterization together with the improved quality control process and the determination of optimal innovation limit values that are adequate for such extreme events as SSWs. This work will be a subject of Part II of this study (Y.-J. Kim et al. 2010, unpublished manuscript).

Acknowledgments

The authors appreciate the support by the Office of Naval Research under ONR Program Element 0601153N. They thank T. Hogan for processing the initial and boundary conditions for the runs, T. Whitcomb for transferring and processing the UKMO analysis data, J. Cummings for providing and explaining the SST and sea-ice data, and M. Peng for comments. They thank D. Boyd at UKMO for retrieving the UKMO data from the archive. They also acknowledge the constructive comments from the anonymous reviewers that helped improve the manuscript. The computing time was provided in large part by the Navy DSRC (DoD Supercomputing Resource Center).

REFERENCES

  • Andrews, D. G., , and McIntyre M. E. , 1976: Planetary waves in horizontal and vertical shear: The generalized Eliassen-Palm relation and the mean zonal acceleration. J. Atmos. Sci., 33 , 20312048.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Andrews, D. G., , Holton J. R. , , and Leovy C. B. , 1987: Middle Atmosphere Dynamics. Academic Press, 489 pp.

  • Baldwin, M. P., , and Dunkerton T. J. , 1999: Propagation of the Arctic Oscillation from the stratosphere to the troposphere. J. Geophys. Res., 104 , 3093730946.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Baldwin, M. P., , and Dunkerton T. J. , 2001: Stratospheric harbingers of anomalous weather regimes. Science, 294 , 581584.

  • Baldwin, M. P., , Cheng X. , , and Dunkerton T. , 1989: The stratospheric major warming of early December 1987. J. Atmos. Sci., 46 , 28632884.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Boyd, J., 1976: The noninteraction of waves with the zonally averaged flow on a spherical earth and the interrelationships of eddy fluxes of energy, heat and momentum. J. Atmos. Sci., 33 , 22852291.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Charlton, A. J., , and Polvani L. M. , 2007: A new look at stratospheric sudden warmings. Part I: Climatology and modeling benchmarks. J. Climate, 20 , 449469.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Coy, L., , Eckermann S. , , and Hoppel K. , 2009: Planetary wave breaking and tropospheric forcing as seen in the stratospheric sudden warming of 2006. J. Atmos. Sci., 66 , 495507.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Cummings, J. A., 2005: Operational multivariate ocean data assimilation. Quart. J. Roy. Meteor. Soc., 131 , 35833604.

  • Gerber, E. P., , Orbe C. , , and Polvani L. M. , 2009: Stratospheric influence on the tropospheric circulation revealed by idealized ensemble forecasts. Geophys. Res. Lett., 36 , L24801. doi:10.1029/2009GL040913.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Hogan, T. F., , and Rosmond T. E. , 1991: The description of the Navy Operational Global Atmospheric Prediction System’s spectral forecast model. Mon. Wea. Rev., 119 , 17861815.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Jung, J-H., , Konor C. S. , , Mechoso C. R. , , and Arakawa A. , 2001: A study of the stratospheric major warming and subsequent flow recovery during the winter of 1979 with an isentropic vertical coordinate model. J. Atmos. Sci., 58 , 26302649.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., 2007: Balance of drag between the middle and lower atmospheres in a global atmospheric forecast model. J. Geophys. Res., 112 , D13104. doi:10.1029/2007JD008647.

    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., , and Arakawa A. , 1995: Improvement of orographic gravity wave parameterization using a mesoscale gravity-wave model. J. Atmos. Sci., 52 , 18751902.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., , and Doyle J. D. , 2005: Extension of an orographic drag parameterization scheme to incorporate orographic anisotropy and flow blocking. Quart. J. Roy. Meteor. Soc., 131 , 18931921.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kim, Y-J., , Campbell W. F. , , and Swadley S. D. , 2010: Reduction of middle-atmospheric forecast bias through improvement in satellite radiance quality control. Wea. Forecasting, 25 , 681700.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kodera, K., 2006: Influence of stratospheric sudden warming on the equatorial troposphere. Geophys. Res. Lett., 33 , L06804. doi:10.1029/2005GL024510.

    • Search Google Scholar
    • Export Citation
  • Kodera, K., , and Chiba M. , 1995: Role of planetary waves in the stratosphere-troposphere coupled variability in the Northern Hemisphere winter. J. Geophys. Res., 100 , 1105511068.

    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., 2005: On the influence of the meridional circulation and surface pressure change on the Arctic Oscillation. J. Geophys. Res., 110 , D21107. doi:10.1029/2004JD005743.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., 2008: Effect of stratospheric sudden warming and vortex intensification on the tropospheric climate. J. Geophys. Res., 113 , D15110. doi:10.1029/2007JD009550.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., , and Kodera K. , 2004: Role of polar-night jet oscillation on the formation of the Arctic Oscillation in the Northern Hemisphere winter. J. Geophys. Res., 109 , D11112. doi:10.1029/2003JD004123.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kuroda, Y., , and Kodera K. , 2007: Correction to “Role of polar-night jet oscillation on the formation of the Arctic Oscillation in the northern hemisphere winter”. J. Geophys. Res., 112 , D08104. doi:10.1029/2007JD008571.

    • Search Google Scholar
    • Export Citation
  • Limpasuvan, V., , Thompson D. W. J. , , and Hartmann D. L. , 2004: The life cycle of the Northern Hemisphere sudden stratospheric warmings. J. Climate, 17 , 25842596.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Manney, G. L., and Coauthors, 2009: Aura Microwave Limb Sounder observations of dynamics and transport during the record-breaking 2009 Arctic stratospheric major warming. Geophys. Res. Lett., 36 , L12815. doi:10.1029/2009GL038586.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Matsuno, T., 1971: A dynamic model of the stratospheric sudden warming. J. Atmos. Sci., 28 , 14791492.

  • Mechoso, C. R., , Yamazaki K. , , Kitoh A. , , and Arakawa A. , 1985: Numerical forecasts of stratospheric warming events during the winter of 1979. Mon. Wea. Rev., 113 , 10151029.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Mukougawa, H., , and Hirooka T. , 2004: Predictability of stratospheric sudden warming: A case study for 1998/99 winter. Mon. Wea. Rev., 112 , 17641776.

    • Search Google Scholar
    • Export Citation
  • Mukougawa, H., , Sakai H. , , and Hirooka T. , 2005: High sensitivity to the initial condition for the prediction of stratospheric sudden warming. Geophys. Res. Lett., 32 , L17806. doi:10.1029/2005GL022909.

    • Search Google Scholar
    • Export Citation
  • Nakagawa, K. I., , and Yamazaki K. , 2006: What kind of stratospheric sudden warming propagates to the troposphere? Geophys. Res. Lett., 33 , L04801. doi:10.1029/2005GL024784.

    • Search Google Scholar
    • Export Citation
  • Naujokat, B., , Krüger K. , , Matthes K. , , Hoffmann J. , , Kunze M. , , and Labitzke K. , 2002: The early major warming in December 2001 – Exceptional? Geophys. Res. Lett., 29 , 2023. doi:10.1029/2002GL015316.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Palmer, T., 1981: Diagnostic study of a wavenumber-2 stratospheric sudden warming in a transformed Eulerian-mean formalism. J. Atmos. Sci., 38 , 844855.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Polvani, L. M., , and Waugh D. W. , 2004: Upward wave activity flux as a precursor to extreme stratospheric events and subsequent anomalous surface weather regimes. J. Climate, 17 , 35483554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., , and Strüfing R. , 1983: Numerical forecasts of stratospheric warming events using a model with a hybrid vertical coordinate. Quart. J. Roy. Meteor. Soc., 109 , 81111.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Thompson, D. W. J., , and Wallace J. M. , 1998: The Arctic oscillation signature in the wintertime geopotential height and temperature fields. Geophys. Res. Lett., 25 , 12971300.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Webster, S., , Brown A. R. , , Cameron D. R. , , and Jones C. P. , 2003: Improvements to the representation of orography in the Met Office Unified Model. Quart. J. Roy. Meteor. Soc., 129 , 19892010.

    • Crossref
    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Zonal mean zonal winds from the Met Office analysis on (a) 10, (b) 20, (c) 24, and (d) 30 Jan 2009. A major SSW occurred on 24 Jan 2009.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 2.
Fig. 2.

As in Fig. 1, but for the zonal mean air temperatures.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 3.
Fig. 3.

As in Fig. 1, but for the geopotential height at 10 hPa.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 4.
Fig. 4.

The 14-day forecasts on 24 Jan 2009 of the zonal mean zonal winds with NOGAPS T239L42 forecast model (without data assimilation), which started on 10 Jan 2009, with orographic drag parameterization of (a) the control scheme by Webster et al. (2003), (b) the scheme by Kim and Arakawa (1995) and Kim and Doyle (2005) only, (c) both the Webster et al. scheme and the Kim and Arakawa and Kim and Doyle scheme, and (d) both schemes with the Webster et al. scheme limited to 100 hPa from the surface.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 5.
Fig. 5.

As in Fig. 4, but for the zonal mean temperatures.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 6.
Fig. 6.

As in Fig. 4, but for the geopotential height at 10 hPa.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 7.
Fig. 7.

As in Fig. 4, but for the zonal mean wind tendency resulting from the zonal component of orographic drag, averaged over 10–31 Jan 2009. (a) Webster et al. (2003) drag from the experiment with the Webster et al. scheme only, (b) Kim drag from the experiment with the Kim and Arakawa and Kim and Doyle scheme only, (c) Webster et al. drag, and (d) Kim drag from the experiment with both the Webster et al. and Kim and Arakawa and Kim and Doyle schemes included, (e) Webster et al. drag, and (f) Kim drag from the experiment with both the Webster et al. and Kim and Arakawa and Kim and Doyle schemes included, but with the drag limited to 100 hPa from the surface for the Webster et al. scheme. Selected negative contours are drawn at −50, −100, −250, −500, and −1000 × 10−2 m s−1 day−1.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 8.
Fig. 8.

The variations in forecast time of the (a) zonal mean zonal wind and (b) zonal mean temperature forecasts from UKMO analysis and various NOGAPS experiments at 10 hPa averaged over 65°–90°N. Additional experiments performed by varying the drag-limiting height (from the default value of 100 hPa) for the Webster et al. (2003) scheme and also varying the effective grid length that determines the magnitude of the reference-level GWD for the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme (thin curves).

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 9.
Fig. 9.

The time variations of the eddy heat fluxes averaged over 65°–90°N at (a) 100, (b) 10, and (c) 1 hPa.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 10.
Fig. 10.

The differences in the zonally averaged and vertically integrated (from 10 to 100 hPa) torques resulting from grid-scale and subgrid-scale orography (mountain torque and orographic torque, respectively) between the modified run with the Webster et al. (2003) scheme (with the drag limited to 100 hPa) plus the Kim and Arakawa (1995) and Kim and Doyle (2005) scheme and the control run with the Webster et al. scheme only.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 11.
Fig. 11.

The time variations of the AO indices calculated over 20°–90°N at (a) 10 hPa, (b) 1000 hPa for 10–30 Jan 2009, and (c) at 1000 hPa for 10 Jan–21 Feb 2009.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 12.
Fig. 12.

The time variations of the NAM indices calculated over 20°–90°N at all levels between the surface and 10 hPa for 10 Jan–21 Feb 2009 from the (a) UKMO analysis, (b) the NOGAPS control (CTL; Webster et al. 2003), and (c) NOGAPS modified [Webster et al. >100 hPa + Kim and Arakawa (1995) and Kim and Doyle (2005)] hindcast runs.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Fig. 13.
Fig. 13.

As in Fig. 4, but for the time variations of the (a) zonal mean zonal wind at 10 hPa averaged over 65°–90°N, with forecasts started on 10, 15, and 20 Jan 2009 for each case.

Citation: Weather and Forecasting 25, 6; 10.1175/2010WAF2222421.1

Table 1.

Details of the NOGAPS SSW hindcast experiments. The Webster et al. (2003) orographic drag parameterization includes both GWD and BLD, whereas the Kim and Arakawa (1995) and Kim and Doyle (2005) parameterization separately includes GWD (Kim and Arakawa 1995) and BLD (Kim and Doyle 2005).

Table 1.

1

The horizontal resolution of the operational model has been increased to T319 on 18 May 2010.

Save