• Andrews, D. G., J. R. Holton, and C. B. Leovy, 1987: Middle Atmosphere Dynamics. Academic Press, 489 pp.

  • Bannister, R. N., A. O’Neill, A. R. Gregory, and K. M. Nissen, 2004: The role of the south-east Asian monsoon and other seasonal features in creating the ‘tape-recorder’ signal in the Unified Model. Quart. J. Roy. Meteor. Soc., 130 , 15311554. doi:10.1256/qj.03.106.

    • Search Google Scholar
    • Export Citation
  • Butchart, N., and J. Austin, 1998: Middle atmosphere climatologies from the troposphere–stratosphere configuration of the UKMO’s Unified Model. J. Atmos. Sci., 55 , 27822809.

    • Search Google Scholar
    • Export Citation
  • Butchart, N., and Coauthors, 2006: Simulations of anthropogenic change in the strength of the Brewer–Dobson circulation. Climate Dyn., 27 , 727741. doi:10.1007/s00382-006-0162-4.

    • Search Google Scholar
    • Export Citation
  • Dall’Amico, M., L. J. Gray, K. H. Rosenlof, A. A. Scaife, K. P. Shine, and P. A. Stott, 2010: Stratospheric temperature trends: Impact of ozone variability and the QBO. Climate Dyn., 34 , 381398. doi:10.1007/s00382-009-0604-x.

    • Search Google Scholar
    • Export Citation
  • Edwards, J., and A. Slingo, 1996: Studies with a flexible new radiation code. 1. Choosing a configuration for a large-scale model. Quart. J. Roy. Meteor. Soc., 122 , 689719.

    • Search Google Scholar
    • Export Citation
  • Eyring, V., and Coauthors, 2006: Assessment of temperature, trace species, and ozone in chemistry–climate model simulations of the recent past. J. Geophys. Res., 111 , D22308. doi:10.1029/2006JD007327.

    • Search Google Scholar
    • Export Citation
  • Eyring, V., and Coauthors, 2008: Overview of the New CCMVal reference and sensitivity simulations in support of upcoming ozone and climate assessments and the planned SPARC CCMVal. SPARC Newsletter, No. 30, SPARC Office, Toronto, ON, Canada, 20–26.

    • Search Google Scholar
    • Export Citation
  • Garcia, R. R., D. R. Marsh, D. E. Kinnison, B. A. Boville, and F. Sassi, 2007: Simulation of secular trends in the middle atmosphere, 1950–2003. J. Geophys. Res., 112 , D09301. doi:10.1029/2006JD007485.

    • Search Google Scholar
    • Export Citation
  • Haynes, P. H., C. J. Marks, M. E. McIntyre, T. G. Shepherd, and K. P. Shine, 1991: On the “downward control” of extratropical diabatic circulations by eddy-induced mean zonal forces. J. Atmos. Sci., 48 , 651678.

    • Search Google Scholar
    • Export Citation
  • Ineson, S., and A. A. Scaife, 2008: The role of the stratosphere in the European climate response to El Niño. Nature Geosci., 2 , 3236. doi:10.1038/ngeo381.

    • Search Google Scholar
    • Export Citation
  • Kiehl, J. T., T. L. Schneider, R. W. Portmann, and S. Solomon, 1999: Climate forcing due to tropospheric and stratospheric ozone. J. Geophys. Res., 104 , (D24). 3123931254. doi:10.1029/1999JD900991.

    • Search Google Scholar
    • Export Citation
  • Lott, F., L. Fairhead, F. Hourdin, and P. Levan, 2005: The stratospheric version of LMDz: Dynamical climatologies, Arctic oscillation, and impact on the surface climate. Climate Dyn., 25 , 851868. doi:10.1007/s00382-005-0064-x.

    • Search Google Scholar
    • Export Citation
  • Manzini, E., M. A. Giorgetta, M. Esch, L. Kornblueh, and E. Roeckner, 2006: The influence of sea surface temperatures on the northern winter stratosphere: Ensemble simulations with the MAECHAM5 model. J. Climate, 19 , 38633881.

    • Search Google Scholar
    • Export Citation
  • Martin, G. M., M. A. Ringer, V. D. Pope, A. Jones, C. Dearden, and T. J. Hinton, 2006: The physical properties of the atmosphere in the new Hadley Centre Global Environmental Model (HadGEM1). Part I: Model description and global climatology. J. Climate, 19 , 12741301.

    • Search Google Scholar
    • Export Citation
  • McLandress, C., and T. G. Shepherd, 2009: Simulated anthropogenic changes in the Brewer–Dobson circulation, including its extension to high latitudes. J. Climate, 22 , 15161540.

    • Search Google Scholar
    • Export Citation
  • Monge-Sanz, B. M., M. P. Chipperfield, A. J. Simmons, and S. M. Uppala, 2007: Mean age of air and transport in a CTM: Comparison of different ECMWF analyses. Geophys. Res. Lett., 34 , L04801. doi:10.1029/2006GL028515.

    • Search Google Scholar
    • Export Citation
  • Morgenstern, O., P. Braesicke, F. M. O’Connor, A. C. Bushell, C. E. Johnson, S. M. Osprey, and J. A. Pyle, 2009: Evaluation of the new UKCA climate-composition model—Part 1: The stratosphere. Geosci. Model Dev., 2 , 4357.

    • Search Google Scholar
    • Export Citation
  • Mote, P., and Coauthors, 1996: An atmospheric tape recorder: The imprint of tropical tropopause temperatures on stratospheric water vapor. J. Geophys. Res., 101 , 39894006.

    • Search Google Scholar
    • Export Citation
  • Osprey, S. M., L. J. Gray, A. C. Bushnell, N. Butchart, S. C. Hardiman, and T. J. Hinton, 2009: The climatology of the middle atmosphere in a vertically extended version of the Met Office’s climate model. Part II: Variability. J. Atmos. Sci., submitted.

    • Search Google Scholar
    • Export Citation
  • Priestley, A., 1993: A quasi-conservative version of the semi-Lagrangian advection scheme. Mon. Wea. Rev., 121 , 621629.

  • Randel, W. J., and F. Wu, 1999: A stratospheric ozone trends data set for global modeling studies. Geophys. Res. Lett., 26 , 30893092.

    • Search Google Scholar
    • Export Citation
  • Randel, W. J., F. Wu, S. J. Oltmans, K. Rosenlof, and G. E. Nedoluha, 2004: Interannual changes of stratospheric water vapor and correlations with tropical tropopause temperatures. J. Atmos. Sci., 61 , 21332148.

    • Search Google Scholar
    • Export Citation
  • Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108 , 4407. doi:10.1029/2002JD002670.

    • Search Google Scholar
    • Export Citation
  • Rosenlof, K. H., 1995: Seasonal cycle of the residual mean meridional circulation in the stratosphere. J. Geophys. Res., 100 , (D3). 51735191.

    • Search Google Scholar
    • Export Citation
  • Rosenlof, K. H., and G. C. Reid, 2008: Trends in the temperature and water vapor content of the tropical lower stratosphere: Sea surface connection. J. Geophys. Res., 113 , D06107. doi:10.1029/2007JD009109.

    • Search Google Scholar
    • Export Citation
  • Scaife, A. A., and J. R. Knight, 2008: Ensemble simulations of the cold European winter of 2005–2006. Quart. J. Roy. Meteor. Soc., 134 , 16471659.

    • Search Google Scholar
    • Export Citation
  • Scaife, A. A., N. Butchart, C. D. Warner, D. Stainforth, W. Norton, and J. Austin, 2000: Realistic quasi-biennial oscillations in a simulation of the global climate. Geophys. Res. Lett., 27 , 34813484.

    • Search Google Scholar
    • Export Citation
  • Scaife, A. A., N. Butchart, C. D. Warner, and R. Swinbank, 2002: Impact of a spectral gravity wave parameterization on the stratosphere in the Met Office Unified Model. J. Atmos. Sci., 59 , 14731489.

    • Search Google Scholar
    • Export Citation
  • Scinocca, J. F., N. A. McFarlane, M. Lazare, J. Li, and D. Plummer, 2008: Technical note: The CCCma third generation AGCM and its extension into the middle atmosphere. Atmos. Chem. Phys., 8 , 70557074.

    • Search Google Scholar
    • Export Citation
  • Sigmond, M., and J. F. Scinocca, 2010: The influence of the basic state on the Northern Hemisphere circulation response to climate change. J. Climate, 23 , 14341446.

    • Search Google Scholar
    • Export Citation
  • Sigmond, M., J. F. Scinocca, and P. J. Kushner, 2008: Impact of the stratosphere on tropospheric climate change. Geophys. Res. Lett., 35 , L12706. doi:10.1029/2008GL033573.

    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., A. Untch, C. Jakob, P. Kållberg, and P. Undén, 1999: Stratospheric water vapour and tropical tropopause temperatures in ECMWF analyses and multi-year simulations. Quart. J. Roy. Meteor. Soc., 125 , 353386. doi:10.1002/qj.49712555318.

    • Search Google Scholar
    • Export Citation
  • Solomon, S., 1999: Stratospheric ozone depletion: A review of concepts and history. Rev. Geophys., 37 , 275316.

  • Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Miller, Eds. 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.

    • Search Google Scholar
    • Export Citation
  • Untch, A., and A. J. Simmons, 1999: Increased stratospheric resolution in the ECMWF forecasting system. ECMWF Newsletter, No. 82, European Centre for Medium-Range Weather Forecasting, Reading, United Kingdom, 2–8. [Available online at http://www.ecmwf.int/publications/newsletters/pdf/82.pdf].

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 re-analysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012. doi:10.1256/qj.04.176.

  • Warner, C. D., and M. E. McIntyre, 1999: Toward an ultra-simple spectral gravity wave parameterization for general circulation models. Earth Planets Space, 51 , 475484.

    • Search Google Scholar
    • Export Citation
  • Zhong, W., S. M. Osprey, L. J. Gray, and J. D. Haigh, 2008: Influence of the prescribed solar spectrum on calculations of atmospheric temperature. Geophys. Res. Lett., 35 , L22813. doi:10.1029/2008GL035993.

    • Search Google Scholar
    • Export Citation
  • View in gallery

    Zonal mean zonal wind U (m s−1). DJF and JJA climatological means for N48, N96, and ERA-40. Black solid contours show positive values of U, black dashed contours show negative values of U, and white contours show the interannual std dev in U (m s−1).

  • View in gallery

    Temperature T (K). Black contours show climatological means for DJF and JJA for N48 and N96. Shaded regions show deviation from ERA-40; dark shading indicates a model cold bias and light shading indicates a model warm bias; contour interval is 2 K.

  • View in gallery

    Std dev of T (K) for DJF and JJA in winter high lats (60°–90°) at 10 hPa. Model ensemble members are plotted individually. Thicker lines are ERA-40 and UKMO analyses, shown for the NH.

  • View in gallery

    Climatological annual cycle in U at 10 hPa, for N48, N96, and ERA-40. Contours are as in Fig. 1.

  • View in gallery

    Climatological annual cycle in T at 10 and 70 hPa, for N48 and N96. Shading is as in Fig. 2.

  • View in gallery

    Annual cycle in the zonally asymmetric component of geopotential height (GPH;m), at 10 hPa: N48, N96, and ERA-40.

  • View in gallery

    Final warming times for N48 and N96, at 60°S and 60°N. Shading shows region within 1 interannual std dev of ERA-40.

  • View in gallery

    Profiles of resolved wave drag (DIVF) and orographic and nonorographic gravity wave drag (OGWD and NOGWD). Units are m s−1 day−1. Values are shown for N48 and for the difference N48 − N96. Positive values are shown with solid contours and negative values with dashed contours.

  • View in gallery

    (a) Annual mean mass streamfunction (kg s−1 m−1) at 70 hPa, calculated from residual vertical velocity w* and downward control; N48 ensemble mean. The contributions to the downward control estimate from individual wave drags are also shown. (b) Annual mean mass streamfunction (kg s−1 m−1) calculated from w*; N48 ensemble mean. (c) As in (b), but using downward control (DC). (d) Planetary wave component of (c). (e) Orographic GWD component of (c). (f) Nonorographic GWD component of (c).

  • View in gallery

    Vertical profile time series of monthly mean water vapor volume-mixing ratio (ppmv) between 5°S and 5°N, for (a) N48 and (b) N96.

  • View in gallery

    Climatological monthly mean specific humidity (10−7 kg kg−1) at 100 hPa for (a) N48 January, (b) N96 January, (c) N48 July, and (d) N96 July.

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 80 65 2
PDF Downloads 35 18 1

The Climatology of the Middle Atmosphere in a Vertically Extended Version of the Met Office’s Climate Model. Part I: Mean State

View More View Less
  • 1 Met Office Hadley Centre, Exeter, Devon, United Kingdom
  • | 2 National Centre for Atmospheric Science, University of Oxford, Oxford, United Kingdom
  • | 3 National Centre for Atmospheric Science, University of Reading, Reading, United Kingdom
  • | 4 Met Office, Exeter, Devon, United Kingdom
  • | 5 Met Office Hadley Centre, Exeter, Devon, United Kingdom
© Get Permissions
Full access

Abstract

The climatology of a stratosphere-resolving version of the Met Office’s climate model is studied and validated against ECMWF reanalysis data. Ensemble integrations are carried out at two different horizontal resolutions. Along with a realistic climatology and annual cycle in zonal mean zonal wind and temperature, several physical effects are noted in the model. The time of final warming of the winter polar vortex is found to descend monotonically in the Southern Hemisphere, as would be expected for purely radiative forcing. In the Northern Hemisphere, however, the time of final warming is driven largely by dynamical effects in the lower stratosphere and radiative effects in the upper stratosphere, leading to the earliest transition to westward winds being seen in the midstratosphere. A realistic annual cycle in stratospheric water vapor concentrations—the tropical “tape recorder”—is captured. Tropical variability in the zonal mean zonal wind is found to be in better agreement with the reanalysis for the model run at higher horizontal resolution because the simulated quasi-biennial oscillation has a more realistic amplitude. Unexpectedly, variability in the extratropics becomes less realistic under increased resolution because of reduced resolved wave drag and increased orographic gravity wave drag. Overall, the differences in climatology between the simulations at high and moderate horizontal resolution are found to be small.

Corresponding author address: Steven Hardiman, Met Office, FitzRoy Road, Exeter, Devon EX1 3PB, United Kingdom. Email: steven.hardiman@metoffice.gov.uk

Abstract

The climatology of a stratosphere-resolving version of the Met Office’s climate model is studied and validated against ECMWF reanalysis data. Ensemble integrations are carried out at two different horizontal resolutions. Along with a realistic climatology and annual cycle in zonal mean zonal wind and temperature, several physical effects are noted in the model. The time of final warming of the winter polar vortex is found to descend monotonically in the Southern Hemisphere, as would be expected for purely radiative forcing. In the Northern Hemisphere, however, the time of final warming is driven largely by dynamical effects in the lower stratosphere and radiative effects in the upper stratosphere, leading to the earliest transition to westward winds being seen in the midstratosphere. A realistic annual cycle in stratospheric water vapor concentrations—the tropical “tape recorder”—is captured. Tropical variability in the zonal mean zonal wind is found to be in better agreement with the reanalysis for the model run at higher horizontal resolution because the simulated quasi-biennial oscillation has a more realistic amplitude. Unexpectedly, variability in the extratropics becomes less realistic under increased resolution because of reduced resolved wave drag and increased orographic gravity wave drag. Overall, the differences in climatology between the simulations at high and moderate horizontal resolution are found to be small.

Corresponding author address: Steven Hardiman, Met Office, FitzRoy Road, Exeter, Devon EX1 3PB, United Kingdom. Email: steven.hardiman@metoffice.gov.uk

1. Introduction

Increased understanding of the importance of stratospheric influence for seasonal and decadal forecasting and climate prediction, along with advances in computational resources, means that general circulation models (GCMs) with increased vertical resolution and an upper boundary in the mesosphere are becoming more common (e.g., McLandress and Shepherd 2009; Manzini et al. 2006; Lott et al. 2005; Garcia et al. 2007). A key reason that stratosphere-resolving GCMs are needed is for detailed study of ozone recovery and climate change in the twenty-first century (Eyring et al. 2006).

Increasing the resolution and upper boundary of a GCM has implications for radiation and gravity wave drag schemes, where changes will impact the model climatology and variability. This paper documents the climatology of a vertically extended version of the Met Office’s (UKMO’s) Unified Model (MetUM), the GCM used at the Met Office for climate and weather prediction. It is the first of two papers; the companion paper (Osprey et al. 2009, manuscript submitted to J. Atmos. Sci., hereafter Part II) will study the variability of this model.

There have been several studies that use this vertically extended version of the model (e.g., Ineson and Scaife 2008; Scaife and Knight 2008) but none has considered in detail the accuracy of the model climatology. Furthermore, this model coupled to the stratospheric chemistry component of the U.K. Chemistry and Aerosols (UKCA) composition model (Morgenstern et al. 2009) has been used for simulating twenty-first-century climate change and ozone recovery by carrying out the Chemistry–Climate Model Validation (CCMVal) reference simulations (Eyring et al. 2008). Because the UKCA coupled chemistry schemes are computationally expensive, when these schemes are included the model is run at a moderate horizontal resolution compared to that used in standard climate prediction runs. In this paper we look at the differences between climatologies from ensemble runs carried out at this moderate resolution (detailed in the next section) and at finer (doubled) resolution to investigate whether significant differences exist.

In what follows, section 2 describes the model in detail and the experimental setup used here, section 3 considers the model zonal mean climatology, section 4 details the model wave driving, and section 5 discusses the tropical “tape recorder” (TTR). Discussion and conclusions are given in section 6.

2. Model description and experimental setup

a. Model description

The climate version of the Met Office’s Unified Model is the Hadley Centre Global Environmental Model (HadGEM) package. HadGEM has a nonhydrostatic, fully compressible, deep atmosphere formulation. It uses a terrain-following, height-based vertical coordinate, semi-Lagrangian advection of all prognostic variables except density (Priestley 1993), an Arakawa-C grid in which the zonal and meridional wind components are staggered, and a Charney–Phillips vertical grid in which the momentum and thermodynamic variables are staggered. A vertically extended version of the atmosphere-only component of HadGEM1 (Martin et al. 2006), using the same physics packages in the troposphere, is used here. With a fully resolved stratosphere included, the major changes made to the vertically extended model are as follows:

  • The number of vertical levels is increased from 38 to 60. The extended model has 32 levels above 16 km with the top level at 84 km as opposed to 10 levels above 16 km with the top level at 39 km. Below 16 km the vertical levels remain the same. The distribution of vertical levels is such that the model resolution from an altitude of 16–30 km is 1.1–1.2 km, from 30–41 km is 1.3–1.8 km, and from 41–52 km is 2–3 km; the vertical resolution above this altitude is 3–6 km.

  • The atmospheric time step is reduced from 30 to 20 min.

  • Nonorographic gravity wave drag is included, using the Ultra Simple Spectral Parameterization (USSP) scheme (Warner and McIntyre 1999) implemented as in Scaife et al. (2002). This scheme calculates the forcing from a continuous spectrum of upward-propagating nonorographic gravity waves. Because the source spectrum used here launches net zero momentum flux (equal eastward and westward flux) and there is no flux through the top boundary (the remaining wave momentum is deposited at the model top), momentum is still conserved. Inclusion of the USSP scheme allows the model to simulate the quasi-biennial oscillation (QBO) of the tropical zonal mean zonal wind (Scaife et al. 2000).

  • The radiation scheme of Edwards and Slingo (1996) is modified. In particular, changes are made to the shortwave spectral irradiance and ozone absorption parameters (Zhong et al. 2008).

  • Methane oxidation is parameterized throughout the model domain, providing a source of water vapor in the stratosphere (Untch and Simmons 1999; Simmons et al. 1999).

b. Experimental setup

Time-varying sea surface temperatures and sea ice fields are prescribed using the Met Office Hadley Centre’s sea ice and sea surface temperature dataset (HadISST; Rayner et al. 2003). Time-evolving carbon dioxide and methane are used, and constant values of N2O, CFC11, and CFC12 are prescribed. No volcanic eruptions, solar cycle, or aerosols are included. A zonal and monthly mean ozone climatology, which includes a seasonal cycle, is prescribed. The ozone climatology was derived from a 1979–2003 time series consisting of merged data from a variety of satellite datasets. The dataset is described in more detail by Dall’Amico et al. (2010).

An ensemble of runs of four members is performed. Each member is 25 years in length, running from 1975 to 2000. A horizontal grid resolution of 2.5° latitude × 3.75° longitude, referred to as N48, is used. This is a moderate horizontal resolution (a T31 resolution spectral model is roughly equivalent to a 3.75° × 3.75° resolution, and T42 to 2.8° × 2.8°).

To test the robustness of the results to resolution, another two-member ensemble is run at twice that resolution, 1.25° latitude × 1.875° longitude, referred to as N96 (the resolution currently used for climate prediction with HadGEM). The members of this ensemble are 20 yr in length and start in 1980. The parameter settings for the gravity wave drag schemes at the two different resolutions are identical.

3. Zonal mean climatology

a. Seasonal means

Figure 1 shows climatological ensemble mean zonal mean zonal wind, U, for December–February (DJF) and June–August (JJA) at the two different resolutions, compared with the equivalent plot from the 40-yr European Centre for Medium-Range Weather Forecasts (ECMWF) Re-Analysis (ERA-40) dataset (Uppala et al. 2005).

Consider first the DJF climatology. The respective strengths at the southern and northern tropospheric jet cores are 33.0 and 42.6 m s−1 in N96, 34.2 and 41.1 m s−1 in N48, and 31.1 and 42.6 m s−1 in ERA-40. Thus, the N96 climatology is slightly closer to the reanalysis. The stratospheric polar night jet (PNJ), a winter hemisphere phenomenon, shows a separation from the subtropical jet that is slightly too large in the model. At 10 hPa (in DJF) the PNJ strength in the model is 20–30 m s−1 whereas in ERA-40 it is 30–40 m s−1. This difference will influence the propagation of planetary waves (e.g., Sigmond et al. 2008). However, the tilt of the PNJ with height is realistic. The summer stratospheric (westward) jet is also well represented in the model. Extratropical variability (of the PNJ) in N48, peaking at 11 m s−1, is seen to be slightly greater than that in the reanalysis, whereas in N96 it is slightly less, peaking at 7 m s−1. Here and in the rest of this paper, the measure of variability is the interannual standard deviation. A strength of the high-top MetUM model is the inclusion of a spectral gravity wave parameterization that allows it to produce a QBO (Scaife et al. 2000), a key component of variability in the tropics. In DJF the variability in N48 (Fig. 1a) peaks at 11 m s−1 in the tropics. In N96 (Fig. 1c) tropical variability peaks slightly higher at 15 m s−1, closer to the value seen in the reanalysis (Fig. 1e).

In JJA, the respective strengths at the southern and northern tropospheric jet cores are 41.4 and 19.1 m s−1 in N96, 40.1 and 19.2 m s−1 in N48, and 41.6 and 22.0 m s−1 in ERA-40. Thus, the N96 climatology is slightly closer to the reanalysis in the winter hemisphere, and the jet in the summer hemisphere is slightly weak in both resolutions. The PNJ is too weak in the model by 10 m s−1 at 10 hPa (as was seen in DJF). Further, the model jet does not tilt equatorward with height as in the reanalysis, so the jet center at 1 hPa is 60°S in the model as opposed to near 45°S in the reanalysis. However, many climate models show a Southern Hemisphere (SH) PNJ that is far too strong (Lott et al. 2005; Garcia et al. 2007), so the representation in the MetUM is reasonable in comparison. The summer stratospheric (westward) jet is far stronger than that in the reanalysis, reaching values of −50 m s−1 as opposed to −30 m s−1 at 1 hPa. (A strong summer jet is also seen in other GCMs; see, e.g., Fig. 3 of Lott et al. 2005; Fig. 1 of Garcia et al. 2007; Fig. 2 of Scinocca et al. 2008). The extratropical variability of the PNJ, at around 10 m s−1, agrees well with the reanalysis. Tropical variability due to the QBO is around 5 m s−1 greater in the model in JJA than in the reanalysis.

Consistently, N96 shows more variability than N48 in the tropical stratosphere because of a more realistic QBO amplitude (see Part II) and shows less variability in the extratropical stratosphere because of differences in planetary wave drag (see section 4).

Figure 2 shows the climatological mean temperature for the model in DJF and JJA, along with temperature biases (difference from ERA-40 climatological temperature). The cold bias in the tropical upper troposphere and warm bias in the tropical lower stratosphere is found also in the low-top version of the model (Martin et al. 2006). Note, however, that the ozone climatology used in that model (Kiehl et al. 1999; Randel and Wu 1999) is not the same as the ozone dataset used in the current high-top model (Dall’Amico et al. 2010). Both resolutions of the high-top model analyzed in this paper show a cold bias of 2–4 K throughout the troposphere and a warm (cold) bias of 2–6 K in the winter (summer) stratosphere poleward of 60°. The warm bias seen in the winter polar stratosphere is unusual—many models have a cold bias here (see Fig. 2 of Lott et al. 2005; Fig. 2 of Garcia et al. 2007). It is consistent with a slightly weak Southern Hemisphere PNJ, whereas most models show a PNJ that is too strong (see earlier in this section). The global mean temperature agrees well with the reanalysis, apart from a 2-K warm bias at 100 hPa (not shown), which suggests that the localized temperature biases seen are due to adiabatic processes. The quadrupole bias seen in the winter hemisphere stratosphere is consistent with the winter jet being too weak at 10 hPa, as was seen in Fig. 1. This, in turn, is consistent with too much wave drag at 10 hPa in the extratropics (see section 4). Warm biases on the order of 6–12 K are also seen above 10 hPa. However, the reanalysis temperatures above 10 hPa may themselves be inaccurate: large differences are seen between the ERA-40 and UKMO analyses above this height (Fig. 1 of Eyring et al. 2006). Below 10 hPa the model temperature biases do not exceed 6 K.

Figure 3 shows the interannual standard deviation in temperature at the winter pole, and at a height of 10 hPa, for DJF and JJA. Results are shown for individual ensemble members, and both ERA-40 and UKMO analyses are used in the Northern Hemisphere (NH). Figure 3a shows that modeled variability in DJF at northern high latitudes (60°–90°N) increases with height, as in the reanalyses, but is 1–2 K less than the variability in the reanalyses. As shown in Fig. 1, N96 is less variable than N48 in the extratropics, although this may be due to having only a two-member ensemble. The 10-hPa variability in DJF is shown as a function of latitude in Fig. 3b. In both the reanalyses and the model, the variability is maximum at the pole and decreases with latitude. However, the modeled variability, at around 4 K at 90°N and decreasing to around 1 K at 40°N, is little over half that seen in the reanalyses.

In the Southern Hemisphere, the reanalyses are noisy and inconsistent with each other and so are not shown. At southern high latitudes (60°–90°S) in JJA (Fig. 3c), modeled variability peaks below the stratopause, at around 6 hPa. A further difference to the structure in variability seen in the NH is that at 10 hPa in JJA (Fig. 3d), the variability peaks not at the pole but rather at 70°S. This is due to a stronger and more stable polar vortex in the SH. The magnitude of variability, around 4 K at 70°S and 1 K at 40°S, is very similar to that in the NH (see also Figs. 5 and 11 of Butchart and Austin 1998).

Overall, the structure of the variability in the model is similar to that in the reanalyses, but the magnitude of the variability is weaker. N48 shows more variability in the extratropics than N96. The ratio of the difference in the variability between ensemble members to the variance within an ensemble member is never larger than 1. Thus, using the F test, the differences between ensemble members are deemed not to be statistically significant at the 10% level.

b. Annual cycle

Figures 1 –3 consider DJF and JJA. Figure 4 shows the climatological annual cycle in U at 10 hPa for N48, N96, and ERA-40. The figure demonstrates that the winter jets peak slightly too early, especially in the N48 model. The strength of the SH winter jet peaks in July in N48 but not until August in the reanalysis and N96. The peak value of U in August in the reanalysis is 7 m s−1 above its July value, as opposed to 0.5 m s−1 in N96. Similarly, the strength of the NH winter jet peaks in December in the model but not until January in the reanalysis. This suggests that the mean wave drag in the model is slightly too large in the extratropics, causing the model dynamics to lag the radiative annual cycle, which peaks at the solstices, by less than is the case in the reanalysis. The magnitude of the annual cycle in the extratropics is slightly smaller in both hemispheres and both resolutions of the model, although that in the NH is much closer to the reanalysis in N96, being about 40 m s−1 as opposed to 30 m s−1 in N48. The seasonal cycle of westward winds in the tropics is also realistic although, as seen in Fig. 1, U is slightly too strong in the tropics in SH winter. Realistic tropical variability is seen in the model because of the ability to generate a QBO, with slightly greater variability in N96 than N48 as seen in Fig. 1.

Figure 5 shows the climatological annual cycle in temperature at 10 and 70 hPa for N48 and N96 and the deviation of the model temperature from the reanalysis (i.e., model biases). The pattern of model biases at 10 hPa (Figs. 5a,b) in the extratropics arises because of two separate features: temperature biases of +6 K or more at 80°S in August (when temperatures are colder than the annual average) and of −6 K at 80°S in January (when temperatures are warmer than the annual average) demonstrate a lack of intra-annual variability in extratropical temperatures. Consistent with Fig. 4, the coldest temperatures occur earlier in the model winter than in the reanalysis. This is demonstrated by the coldest temperatures at 60°S occurring in June when the peak warm bias at this latitude, indicating the coldest temperatures in the reanalysis, does not occur until July.

At 70 hPa (Figs. 5c,d) the same pattern of biases is seen, but an overall warm bias is also present, as was seen in Fig. 2. There is a warm bias in the tropics of just over 2 K. In the SH, from April to November, there is a warm bias poleward of 60°S of 2–6 K. Polar stratospheric clouds (PSCs), which play a key role in polar ozone depletion (Solomon 1999), form at temperatures less than 196 K, and the mean temperatures in this region are close to that value. Thus, the warm temperature bias suggests either that when simulating stratospheric chemistry this model will underestimate polar ozone depletion, or that the prescribed ozone climatology has too much ozone in this region. Determining whether the issue is with the model or with the prescribed ozone climatology would require a simulation with coupled stratospheric chemistry and is beyond the scope of this study.

c. Stationary waves

Whereas Figs. 4 and 5 display zonal mean fields, Fig. 6 shows the annual cycle in the monthly mean zonally asymmetric (stationary wave) components of geopotential height (GPH). The value shown is the standard deviation from the zonal mean of geopotential height at 10 hPa, at each latitude, around a circle of constant latitude. It thus includes components of all resolved wavenumbers from 1 upward. The figure shows that unlike the zonal mean fields, the stationary wave peaks at the same time in the model as in the reanalysis. The only exception to this is N96 in the SH where the peak is in September as opposed to October for the reanalysis. The magnitude of the modeled stationary wave agrees fairly well with the reanalysis although the following points should be made. Figure 1 showed that the winter jet in the stratosphere in JJA is approximately 10 m s−1 too weak in the model. Planetary waves cannot propagate through eastward winds that are too strong (Charney–Drazin criterion; Andrews et al. 1987); thus, an eastward jet that is too weak will lead to a greater vertical wave propagation in the model. Hence, in the reanalysis there is a local minimum in vertical wave propagation in July, when the jet is at its strongest, giving rise to a double peak structure in the stationary wave in the SH (Fig. 6c). Because of the model jets being slightly weaker, this feature is not present in the model, and the September–October peak in the model stationary wave is slightly too strong (by 50 m in N96 and 150 m in N48). In the NH extratropics, the magnitude of the stationary wave is slightly too strong in early winter in N48 and slightly too weak in late winter in both resolutions of the model (by about 50 m in each case). This is important since the intraseasonal variability in wave driving directly relates to the representation of stratospheric sudden warmings (SSWs) in the model (see Part II). The peak wave amplitudes are seen at the correct latitude in the model in both hemispheres, peaking at approximately 65°.

Table 1 shows a breakdown of the stationary wave (azonal geopotential height at 10 hPa), in terms of percentage contribution to peak values, into different wavenumber components in both Northern and Southern Hemispheres. It also shows the month in which the individual wavenumber components peak. Contributions from wavenumbers 5 and higher are insignificant and thus not shown. In both the model and ERA-40, around 70%–80% of the stationary wave is due to wavenumber 1, 15%–20% is due to wavenumber 2, and 5%–10% is due to wavenumbers 3 and 4. The separation into different stationary wavenumber components is thus realistic in the model.

In the SH, the peak in wavenumber 1 occurs later in the year than for other wavenumbers, around October as opposed to August, and the double peak in the stationary wave seen in the reanalysis is due solely to wavenumber 1, which also peaks in June (not shown). The stationary wave in N48 shows a more realistic distribution across wavenumbers and better timing of peak values, but Fig. 6 shows that the magnitude of the stationary wave is closer to the reanalysis in N96. In the NH, the main point to note is that the peak amplitude in wavenumbers 3 and 4 is seen earlier in the model than in the reanalysis. This accounts for the stationary wave being slightly weak in late winter as noted above.

d. Final warming

Figure 7 shows the time of the final warming (averaged over all years) as a function of height at 60°S and 60°N. Monthly mean data are used, and the date is calculated by assuming that monthly mean data correspond to the 15th of each month and linearly interpolating to find when the zonal mean zonal wind undergoes a transition from eastward to westward.

For one ensemble member of the N96 run a curve is added to the graph that computes the time of the final warming using daily data. The time of the warming is found to be a few days later than that computed from monthly mean data in the SH, but overall the two methods agree well. The SH final warming times reported here are consistent with Scaife et al. (2000) and Eyring et al. (2006), both of which use monthly mean data. Using daily data in the NH, the zonal wind is found to change from eastward to westward multiple times, so the final warming date is harder to define, especially in the mid to lower stratosphere. Nevertheless, the shape of the vertical profile remains similar to and within one standard deviation of the reanalysis data. Therefore, monthly mean data are considered sufficient to define the time of final warming.

The downward progression in the zero wind line is clear in the SH (Figs. 7a,c), with N96 earlier than the reanalysis by about 10 days at all heights and N48 agreeing well with the reanalysis. In the NH (Figs. 7b,d) there is no such top-down progression of the zero wind line, with the transition from eastward to westward occurring first in the middle stratosphere (∼10 hPa). This very different character in the modeled profile of the zero wind line agrees well with that in the reanalysis, and although N96 is again early at some heights, it does fall within one interannual standard deviation of ERA-40 data in the NH. The earlier final warming times in the mid to lower stratosphere are almost certainly largely wave driven. The later final warming times in the upper stratosphere appear less affected by wave driving. This is consistent with a faster recovery, following a sudden warming, to strong vortex conditions in the upper stratosphere. There is substantially more variability in the final warming date in the NH than the SH, consistent with the generally more disturbed winters in the NH. The final warming in the SH occurs at the beginning of November at 1 hPa, and reaches 50 hPa by mid-December. In the NH the final warming occurs around April. Thus, the occurrence of the final warming with respect to the seasonal cycle is 1–2 months later in the SH than it is in the NH.

It should be mentioned that many coupled chemistry models show late final warming times in the Southern Hemisphere (Fig. 2 of Eyring et al. 2006). The final warming time may be influenced by the effects of ozone on temperature (discussed in section 3b of the current paper) and thus the early final warming times seen here may change when integrating with coupled stratospheric chemistry.

4. Wave driving

Momentum from the breaking of planetary and gravity waves acts to accelerate or decelerate the mean flow in the atmosphere. Through, for example, the effect of the mean flow on temperature (thermal-wind balance; Andrews et al. 1987, p.120), the onset of stratospheric sudden warmings (see Part II), and driving the Brewer–Dobson circulation (the seasonal-mean meridional mass transport in the stratosphere; Butchart et al. 2006), wave breaking has a large influence on the climatology of the atmosphere.

Figure 8 shows the climatological annual mean resolved wave drag (DIVF), parameterized orographic gravity wave drag (OGWD), and nonorographic gravity wave drag (NOGWD)1 in the stratosphere and mesosphere for N48 and the difference N48 − N96. All drags are scaled to units of m s−1 day−1. DIVF denotes the EP flux divergence, · F, the resolved wave driving in the transformed Eulerian mean (TEM) formalism. From Eqs. (3.5.2) and (3.5.3) of Andrews et al. (1987), the TEM zonal-mean momentum equation is
i1520-0469-67-5-1509-e1
where
i1520-0469-67-5-1509-e2
and the components of the EP flux are given by
i1520-0469-67-5-1509-e3
i1520-0469-67-5-1509-e4
Here spherical coordinates (λ, ϕ, Z) are used, overbars denote zonal means, the 3D velocity is (u, υ, w) with residual (wave driven) meridional circulation (υ*, w*), θ is potential temperature, f = 2Ω sinϕ is the Coriolis parameter (where Ω is the angular velocity of the earth), a is the radius of the earth, ρ0 is density, X denotes friction (or other nonconservative mechanical forcing), and subscripts denote derivatives.

The annual mean drags are dominated by the contribution from the winter hemisphere and are all negative in the extratropics. Up to a height of about 3 hPa DIVF dominates over OGWD and NOGWD everywhere except in the tropical stratosphere. The values of OGWD in the extratropical lower stratosphere are much smaller in the MetUM than, for instance, in the Canadian Middle Atmosphere Model (CMAM): Fig. 8c shows values of around −0.1 m s−1 day−1 at 50°N and 70 hPa, whereas Fig. 14a of McLandress and Shepherd (2009) shows values <−1.25 m s−1 day−1 in this region in the CMAM. OGWD in the MetUM was decreased in this region to dramatically reduce stratospheric errors over and downstream of Tibet. Before this decrease, some large drags over the Himalayas, the Rockies, and the Andes led to unrealistic localized wind minima over these mountainous areas. Fortunately, the decrease in OGWD did not reintroduce a cold polar vortex in the NH winter stratosphere (S. Webster 2009, personal communication). Values of OGWD in the extratropical lower stratosphere are currently not well constrained by observations, but it is important to know which values are appropriate as they affect conclusions on the contribution of OGWD to tropical upwelling via the downward control principle (see Fig. 9).

The magnitude of OGWD in the extratropics in N96 is around 1.3 times larger than in N48 (Fig. 8d) because the vertical flux into the stratosphere of horizontal momentum from orographic gravity waves is 1.3 times larger in N96 than N48. Recall that the parameter settings of the OGWD scheme are identical in the N48 and N96 model setups. The momentum flux on the lower boundary is found to be similar between the two resolutions; the difference arises, therefore, because more momentum is dissipated in the troposphere in N48. The increase in OGWD in the lower stratosphere in N96 is compensated for by a decrease in resolved wave drag (DIVF; Fig. 8b), which is consistent with a change in the wave refractive index, n02:
i1520-0469-67-5-1509-e5
[Eq. (5.3.8) from Andrews et al. 1987], where N is the buoyancy frequency, q is the quasigeostrophic potential vorticity, and H is a scale height (7 km). The wave refractive index, as computed from climatological annual mean U and T, shows a local minimum centered at 35° and 100 hPa in N48 that is absent in N96 (not shown). The increase in refractive index in this region in N96 implies that vertically propagating resolved waves originating in the midlatitudes will be directed more equatorward toward the subtropical lower stratosphere, reducing resolved wave drag in the extratropical stratosphere with respect to N48 (Fig. 8b). This agrees with the result, shown in Part II, that fewer stratospheric sudden warmings are simulated in N96 than in N48. The greater NOGWD in the stratosphere in N48 (Fig. 8f) is due to a greater wave flux into the stratosphere (not shown), due in turn to changes in U in the troposphere.

A measure of the strength of the Brewer–Dobson circulation, important for the distribution of chemical species throughout the stratosphere and its effect on the thermal structure of the stratosphere, is the residual (wave driven) upwelling mass flux from troposphere to stratosphere in the tropics. Figure 9a shows the annual mean mass streamfunction, calculated from the residual vertical velocity w*, and from wave driving via the downward control principle (Haynes et al. 1991), at 70 hPa. This demonstrates the strength of the tropical upwelling (difference in maxima and minima of the streamfunction; Rosenlof 1995) and the width of the tropical pipe (latitudinal distance between the maxima and minima of the streamfunction). The upwelling is weaker in the model than in reanalysis data but by less than one interannual standard deviation (not shown). However, Monge-Sanz et al. (2007) show that age of air is too young when derived from the ERA-40 and UKMO analyses, suggestive of tropical upwelling being too strong in the analyses. The downward control streamfunction in Fig. 9a is dominated by the contribution from DIVF. Figures for N96 are very similar to those for N48 and hence are not shown. Figures 9b and 9c show the streamfunctions (as calculated from w* and downward control) as a function of latitude and height. The downward control streamfunction approximates well the streamfunction given from w*, the circulation from equator to pole in the troposphere and throughout the stratosphere being evident in both hemispheres. The only discrepancy is that downward control fails to capture the region of upwelling in the NH polar stratosphere. The streamfunction is again dominated by the DIVF component (Fig. 9d). Gravity waves drive a much weaker circulation that exists only in the stratosphere and extends much higher into the stratosphere than that forced by resolved waves (Figs. 9e,f). This would not be the case if OGWD were stronger in the extratropical lower stratosphere, as discussed earlier in this section. In all these panels it is the winter wave driving that dominates the stratospheric part of the streamfunction. This is as expected from the Charney–Drazin criterion (Andrews et al. 1987) as the zonal wind is largely westward in the summer stratosphere.

5. Tropical tape recorder

Transient concentrations of water vapor in the lower stratosphere are a useful diagnostic of the strength of the tropical tape recorder (Mote et al. 1996). The TTR is thought to be controlled by the seasonal change in temperature at the tropical tropopause and the strength of the Brewer–Dobson circulation. The tropical tropopause temperature affects the extent to which water vapor condenses out of the air before it reaches the stratosphere. The Brewer–Dobson circulation, discussed in the previous section, influences the rate of troposphere–stratosphere exchange of air and the transport of stratospheric air.

Figure 10 shows the time evolution of water vapor profiles (specific humidity), between 5°S and 5°N, for both N48 and N96 ensembles. The time–altitude gradients of the plots are similar to observations from the HALOE satellite instrument (Rosenlof and Reid 2008), the tape recorder signal taking 1–1.5 years to ascend from 100 to 10 hPa, implying reasonable simulation of the strength of the Brewer–Dobson circulation. Noticeable interannual variability is also evident, consistent with observations (Randel et al. 2004). A realistic annual cycle is reproduced at all heights in the model as seen in Halogen Occultation Experiment (HALOE) observations (not shown), linked with the annual cycle in cold-point tropopause temperature (Bannister et al. 2004; Rosenlof and Reid 2008).

N96 has higher specific humidity in the stratosphere than N48 by more than 0.4 ppmv. To investigate this difference, Fig. 11 shows January and July climatologies of specific humidity at 100 hPa for N48 and N96 ensembles. In the tropics, relatively low values are seen in the NH winter (January; Figs. 11a,b), linked to anomalously low temperatures at the tropopause. These cold-point temperatures (not shown) are restricted to a broad region centered over the tropical Pacific. During NH summer (July; Figs. 11c,d) also, an anomalously low water vapor concentration can be found over the tropical Pacific, but it is now restricted to equatorial and southern latitudes. Anomalously high water vapor concentration is found in northern latitudes over the Pacific and Central America. The tropical tropopause temperatures in both seasons are very similar in N48 and N96 (not shown); for this reason, the tropical values of specific humidity are similar in N48 and N96.

In the subtropics and extratropics, however, N96 shows higher specific humidity at 100 hPa than N48. In January the extratropical specific humidity is 2.2–2.4 × 10−6 kg kg−1 in N96 and 2.0–2.2 × 10−6 kg kg−1 in N48. In July it is 2.2–2.6 × 10−6 kg kg−1 in N96 and 2.0–2.4 × 10−6 kg kg−1 in N48. The fact that the differences in specific humidity between N96 and N48 are found in the subtropics and extratropics suggests that stratospheric water vapor is higher in N96 not simply because of increased water vapor transport through the tropical pipe but also, perhaps primarily, through increased horizontal mixing of moist tropospheric air across the tropopause in the subtropics.

6. Discussion and conclusions

In this paper, the climatology of a vertically extended version of the MetUM has been evaluated and for the most part found to agree well with the reanalysis data. Ensemble integrations were carried out at two different horizontal resolutions, moderate (2.5° latitude × 3.75° longitude, referred to as N48) and high (1.25° latitude × 1.875° longitude, referred to as N96), to consider the differences in climatology between these resolutions.

The strength and position of the climatological zonal mean zonal wind tropospheric jets are realistic. However the summer hemisphere stratospheric jet in June–August (JJA) is too strong by up to 20 m s−1 at 1 hPa in the model (as with other GCMs; see Lott et al. 2005; Garcia et al. 2007; Scinocca et al. 2008). N96 shows greater interannual variability in the zonal wind in the tropics, consistent with better-resolved tropical waves leading to a more realistic QBO amplitude. The tropical variability in N48 is weaker than that in N96 and the reanalysis by 5 m s−1. However, N48 shows greater and more realistic variability in the extratropics, consistent with the greater extratropical resolved wave drag seen in N48. The model temperature has a cold bias of 2–4 K in the troposphere but otherwise agrees well with the reanalysis. Overall, the interannual standard deviation in model temperature is weaker than that in the reanalysis by around 1 K, with N48 showing more realistic variability than N96.

The strength of the Southern Hemisphere winter jet is found to peak early in the model. Also, the coldest 10-hPa temperatures at 60°S occur earlier in the model than in the reanalysis. A similar result holds in the Northern Hemisphere. This is consistent with the wave drag in the model being slightly too strong, causing the dynamical response to lag the radiative cycle by less than is shown in the reanalysis. Furthermore, the extreme values of zonal wind and temperature in the extratropics seen during the year in the model are less than those in the reanalysis by as much as 10 m s−1 and 6 K, respectively, again consistent with the model wave drag being too strong. The magnitude of the azonal (stationary wave) component of geopotential height peaks in June and October in the Southern Hemisphere in the reanalysis, showing a minimum in July. This minimum is not captured by the model since the modeled polar night jet in Southern Hemisphere winter is slightly weak. A weaker jet leads to greater vertical propagation of planetary waves and thus increased stationary wave amplitude. This difference in vertical wave propagation is potentially important for correctly simulating stratospheric sudden warmings. In general, N48 shows more realistic timing in the seasonal cycle of azonal geopotential height, but N96 shows more realistic magnitude of the stationary wave.

The timing of the final stratospheric warming in both hemispheres is well captured by the model, especially at the N48 resolution. N96 is found to give final warming times about 10 days earlier than the reanalysis, but the times are still within one interannual standard deviation of the reanalysis times. The final warming times calculated from monthly mean fields are found to agree well with those calculated from daily fields, especially in the SH where the final warming occurs first at high altitudes and propagates downward. In the Northern Hemisphere, the earliest final warming time is found at 10 hPa. The presence of stratospheric sudden warmings in the Northern Hemisphere influences this structure because of the polar vortex reforming at higher altitudes where the radiative time scales are shorter. The vertical structure of the Northern Hemisphere final warming times is substantially wave driven in the lower stratosphere and is much further from a purely radiative effect than in the Southern Hemisphere.

The flux of orographic gravity waves into the extratropical stratosphere is 30% larger in N96 than N48 because more momentum from these waves is dissipated in the troposphere in N48. This leads to around 30% greater orographic gravity wave drag (OGWD) in the stratosphere in N96. OGWD in the MetUM is intentionally somewhat weaker in the lower stratosphere than is the case in the CMAM (McLandress and Shepherd 2009). Obtaining observations of OGWD is important because currently there is uncertainty in the contribution of OGWD to wave-driven upwelling and thus to the strength of the stratospheric meridional circulation (the Brewer–Dobson circulation). Conversely, there is greater resolved wave drag in the extratropical stratosphere in N48. Small differences in the background wind distribution lead to differences in wave refractive index between N48 and N96, causing resolved waves to propagate more equatorward in N96. This is consistent with more stratospheric sudden warmings being simulated at N48 (see Part II).

The Brewer–Dobson circulation has been diagnosed by evaluating the mass streamfunction from residual vertical velocity w* and from downward control. In the MetUM, resolved waves dominate in the downward control calculation, with negligible contribution from parameterized gravity waves. The downward control diagnostic compares well with w* in general but fails to capture the upwelling in the NH stratosphere. Streamfunctions for N48 and N96 are very similar in pattern, showing convergence in the behavior of this diagnostic.

A tropical “tape recorder” signal is simulated in the MetUM. Its ascent with time is in good agreement with water vapor observations from the HALOE satellite instrument, taking 1–1.5 years to ascend from 100 to 10 hPa, and the annual cycle about the cold-point tropopause temperature is realistic. N96 shows greater specific humidity in the stratosphere than N48, by more than 0.4 ppmv. This may result primarily from more lateral mixing of moist tropospheric air into the stratosphere in N96 rather than from the vertical transport from troposphere to stratosphere through the tropical pipe.

In conclusion, the climatology of the vertically extended version of the MetUM considered here is found to match the reanalysis closely. There are some differences between N48 and N96, showing that convergence has not been achieved by the N48 resolution, but N48 is found to be as close to the reanalysis as N96 in the diagnostics considered here. Thus, within the scope of the diagnostics considered in this paper, N48 is thought to be a reasonable choice of resolution for stratospheric climate modeling with the MetUM. This is an important consideration when running computationally expensive simulations, such as those with coupled chemistry. However, it must be emphasized that higher-resolution simulations may have improved tropospheric representation (Martin et al. 2006; Solomon et al. 2007, chapter 8) and both the basic zonal mean zonal wind climatology and the resolved waves propagating from the troposphere to the stratosphere can influence the stratosphere–troposphere coupled response to climate change (e.g., Sigmond and Scinocca 2010). The possible effects of increased horizontal resolution on simulating the impacts of climate change using interactive chemistry and/or a coupled ocean–sea ice model are worthy of future study.

Acknowledgments

This work was supported by the Joint DECC and Defra Integrated Climate Programme, DECC/Defra (GA01101), and partially by the NERC National Centre for Atmospheric Science (NCAS) climate directorate. ECMWF ERA-40 data used in this study have been obtained from the ECMWF data server. The authors thank Gill Martin for input to development of the high-top HadGEM model.

REFERENCES

  • Andrews, D. G., J. R. Holton, and C. B. Leovy, 1987: Middle Atmosphere Dynamics. Academic Press, 489 pp.

  • Bannister, R. N., A. O’Neill, A. R. Gregory, and K. M. Nissen, 2004: The role of the south-east Asian monsoon and other seasonal features in creating the ‘tape-recorder’ signal in the Unified Model. Quart. J. Roy. Meteor. Soc., 130 , 15311554. doi:10.1256/qj.03.106.

    • Search Google Scholar
    • Export Citation
  • Butchart, N., and J. Austin, 1998: Middle atmosphere climatologies from the troposphere–stratosphere configuration of the UKMO’s Unified Model. J. Atmos. Sci., 55 , 27822809.

    • Search Google Scholar
    • Export Citation
  • Butchart, N., and Coauthors, 2006: Simulations of anthropogenic change in the strength of the Brewer–Dobson circulation. Climate Dyn., 27 , 727741. doi:10.1007/s00382-006-0162-4.

    • Search Google Scholar
    • Export Citation
  • Dall’Amico, M., L. J. Gray, K. H. Rosenlof, A. A. Scaife, K. P. Shine, and P. A. Stott, 2010: Stratospheric temperature trends: Impact of ozone variability and the QBO. Climate Dyn., 34 , 381398. doi:10.1007/s00382-009-0604-x.

    • Search Google Scholar
    • Export Citation
  • Edwards, J., and A. Slingo, 1996: Studies with a flexible new radiation code. 1. Choosing a configuration for a large-scale model. Quart. J. Roy. Meteor. Soc., 122 , 689719.

    • Search Google Scholar
    • Export Citation
  • Eyring, V., and Coauthors, 2006: Assessment of temperature, trace species, and ozone in chemistry–climate model simulations of the recent past. J. Geophys. Res., 111 , D22308. doi:10.1029/2006JD007327.

    • Search Google Scholar
    • Export Citation
  • Eyring, V., and Coauthors, 2008: Overview of the New CCMVal reference and sensitivity simulations in support of upcoming ozone and climate assessments and the planned SPARC CCMVal. SPARC Newsletter, No. 30, SPARC Office, Toronto, ON, Canada, 20–26.

    • Search Google Scholar
    • Export Citation
  • Garcia, R. R., D. R. Marsh, D. E. Kinnison, B. A. Boville, and F. Sassi, 2007: Simulation of secular trends in the middle atmosphere, 1950–2003. J. Geophys. Res., 112 , D09301. doi:10.1029/2006JD007485.

    • Search Google Scholar
    • Export Citation
  • Haynes, P. H., C. J. Marks, M. E. McIntyre, T. G. Shepherd, and K. P. Shine, 1991: On the “downward control” of extratropical diabatic circulations by eddy-induced mean zonal forces. J. Atmos. Sci., 48 , 651678.

    • Search Google Scholar
    • Export Citation
  • Ineson, S., and A. A. Scaife, 2008: The role of the stratosphere in the European climate response to El Niño. Nature Geosci., 2 , 3236. doi:10.1038/ngeo381.

    • Search Google Scholar
    • Export Citation
  • Kiehl, J. T., T. L. Schneider, R. W. Portmann, and S. Solomon, 1999: Climate forcing due to tropospheric and stratospheric ozone. J. Geophys. Res., 104 , (D24). 3123931254. doi:10.1029/1999JD900991.

    • Search Google Scholar
    • Export Citation
  • Lott, F., L. Fairhead, F. Hourdin, and P. Levan, 2005: The stratospheric version of LMDz: Dynamical climatologies, Arctic oscillation, and impact on the surface climate. Climate Dyn., 25 , 851868. doi:10.1007/s00382-005-0064-x.

    • Search Google Scholar
    • Export Citation
  • Manzini, E., M. A. Giorgetta, M. Esch, L. Kornblueh, and E. Roeckner, 2006: The influence of sea surface temperatures on the northern winter stratosphere: Ensemble simulations with the MAECHAM5 model. J. Climate, 19 , 38633881.

    • Search Google Scholar
    • Export Citation
  • Martin, G. M., M. A. Ringer, V. D. Pope, A. Jones, C. Dearden, and T. J. Hinton, 2006: The physical properties of the atmosphere in the new Hadley Centre Global Environmental Model (HadGEM1). Part I: Model description and global climatology. J. Climate, 19 , 12741301.

    • Search Google Scholar
    • Export Citation
  • McLandress, C., and T. G. Shepherd, 2009: Simulated anthropogenic changes in the Brewer–Dobson circulation, including its extension to high latitudes. J. Climate, 22 , 15161540.

    • Search Google Scholar
    • Export Citation
  • Monge-Sanz, B. M., M. P. Chipperfield, A. J. Simmons, and S. M. Uppala, 2007: Mean age of air and transport in a CTM: Comparison of different ECMWF analyses. Geophys. Res. Lett., 34 , L04801. doi:10.1029/2006GL028515.

    • Search Google Scholar
    • Export Citation
  • Morgenstern, O., P. Braesicke, F. M. O’Connor, A. C. Bushell, C. E. Johnson, S. M. Osprey, and J. A. Pyle, 2009: Evaluation of the new UKCA climate-composition model—Part 1: The stratosphere. Geosci. Model Dev., 2 , 4357.

    • Search Google Scholar
    • Export Citation
  • Mote, P., and Coauthors, 1996: An atmospheric tape recorder: The imprint of tropical tropopause temperatures on stratospheric water vapor. J. Geophys. Res., 101 , 39894006.

    • Search Google Scholar
    • Export Citation
  • Osprey, S. M., L. J. Gray, A. C. Bushnell, N. Butchart, S. C. Hardiman, and T. J. Hinton, 2009: The climatology of the middle atmosphere in a vertically extended version of the Met Office’s climate model. Part II: Variability. J. Atmos. Sci., submitted.

    • Search Google Scholar
    • Export Citation
  • Priestley, A., 1993: A quasi-conservative version of the semi-Lagrangian advection scheme. Mon. Wea. Rev., 121 , 621629.

  • Randel, W. J., and F. Wu, 1999: A stratospheric ozone trends data set for global modeling studies. Geophys. Res. Lett., 26 , 30893092.

    • Search Google Scholar
    • Export Citation
  • Randel, W. J., F. Wu, S. J. Oltmans, K. Rosenlof, and G. E. Nedoluha, 2004: Interannual changes of stratospheric water vapor and correlations with tropical tropopause temperatures. J. Atmos. Sci., 61 , 21332148.

    • Search Google Scholar
    • Export Citation
  • Rayner, N. A., D. E. Parker, E. B. Horton, C. K. Folland, L. V. Alexander, D. P. Rowell, E. C. Kent, and A. Kaplan, 2003: Global analyses of sea surface temperature, sea ice, and night marine air temperature since the late nineteenth century. J. Geophys. Res., 108 , 4407. doi:10.1029/2002JD002670.

    • Search Google Scholar
    • Export Citation
  • Rosenlof, K. H., 1995: Seasonal cycle of the residual mean meridional circulation in the stratosphere. J. Geophys. Res., 100 , (D3). 51735191.

    • Search Google Scholar
    • Export Citation
  • Rosenlof, K. H., and G. C. Reid, 2008: Trends in the temperature and water vapor content of the tropical lower stratosphere: Sea surface connection. J. Geophys. Res., 113 , D06107. doi:10.1029/2007JD009109.

    • Search Google Scholar
    • Export Citation
  • Scaife, A. A., and J. R. Knight, 2008: Ensemble simulations of the cold European winter of 2005–2006. Quart. J. Roy. Meteor. Soc., 134 , 16471659.

    • Search Google Scholar
    • Export Citation
  • Scaife, A. A., N. Butchart, C. D. Warner, D. Stainforth, W. Norton, and J. Austin, 2000: Realistic quasi-biennial oscillations in a simulation of the global climate. Geophys. Res. Lett., 27 , 34813484.

    • Search Google Scholar
    • Export Citation
  • Scaife, A. A., N. Butchart, C. D. Warner, and R. Swinbank, 2002: Impact of a spectral gravity wave parameterization on the stratosphere in the Met Office Unified Model. J. Atmos. Sci., 59 , 14731489.

    • Search Google Scholar
    • Export Citation
  • Scinocca, J. F., N. A. McFarlane, M. Lazare, J. Li, and D. Plummer, 2008: Technical note: The CCCma third generation AGCM and its extension into the middle atmosphere. Atmos. Chem. Phys., 8 , 70557074.

    • Search Google Scholar
    • Export Citation
  • Sigmond, M., and J. F. Scinocca, 2010: The influence of the basic state on the Northern Hemisphere circulation response to climate change. J. Climate, 23 , 14341446.

    • Search Google Scholar
    • Export Citation
  • Sigmond, M., J. F. Scinocca, and P. J. Kushner, 2008: Impact of the stratosphere on tropospheric climate change. Geophys. Res. Lett., 35 , L12706. doi:10.1029/2008GL033573.

    • Search Google Scholar
    • Export Citation
  • Simmons, A. J., A. Untch, C. Jakob, P. Kållberg, and P. Undén, 1999: Stratospheric water vapour and tropical tropopause temperatures in ECMWF analyses and multi-year simulations. Quart. J. Roy. Meteor. Soc., 125 , 353386. doi:10.1002/qj.49712555318.

    • Search Google Scholar
    • Export Citation
  • Solomon, S., 1999: Stratospheric ozone depletion: A review of concepts and history. Rev. Geophys., 37 , 275316.

  • Solomon, S., D. Qin, M. Manning, Z. Chen, M. Marquis, K. B. Averyt, M. Tignor, and H. L. Miller, Eds. 2007: Climate Change 2007: The Physical Science Basis. Cambridge University Press, 996 pp.

    • Search Google Scholar
    • Export Citation
  • Untch, A., and A. J. Simmons, 1999: Increased stratospheric resolution in the ECMWF forecasting system. ECMWF Newsletter, No. 82, European Centre for Medium-Range Weather Forecasting, Reading, United Kingdom, 2–8. [Available online at http://www.ecmwf.int/publications/newsletters/pdf/82.pdf].

    • Search Google Scholar
    • Export Citation
  • Uppala, S. M., and Coauthors, 2005: The ERA-40 re-analysis. Quart. J. Roy. Meteor. Soc., 131 , 29613012. doi:10.1256/qj.04.176.

  • Warner, C. D., and M. E. McIntyre, 1999: Toward an ultra-simple spectral gravity wave parameterization for general circulation models. Earth Planets Space, 51 , 475484.

    • Search Google Scholar
    • Export Citation
  • Zhong, W., S. M. Osprey, L. J. Gray, and J. D. Haigh, 2008: Influence of the prescribed solar spectrum on calculations of atmospheric temperature. Geophys. Res. Lett., 35 , L22813. doi:10.1029/2008GL035993.

    • Search Google Scholar
    • Export Citation

Fig. 1.
Fig. 1.

Zonal mean zonal wind U (m s−1). DJF and JJA climatological means for N48, N96, and ERA-40. Black solid contours show positive values of U, black dashed contours show negative values of U, and white contours show the interannual std dev in U (m s−1).

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 2.
Fig. 2.

Temperature T (K). Black contours show climatological means for DJF and JJA for N48 and N96. Shaded regions show deviation from ERA-40; dark shading indicates a model cold bias and light shading indicates a model warm bias; contour interval is 2 K.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 3.
Fig. 3.

Std dev of T (K) for DJF and JJA in winter high lats (60°–90°) at 10 hPa. Model ensemble members are plotted individually. Thicker lines are ERA-40 and UKMO analyses, shown for the NH.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 4.
Fig. 4.

Climatological annual cycle in U at 10 hPa, for N48, N96, and ERA-40. Contours are as in Fig. 1.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 5.
Fig. 5.

Climatological annual cycle in T at 10 and 70 hPa, for N48 and N96. Shading is as in Fig. 2.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 6.
Fig. 6.

Annual cycle in the zonally asymmetric component of geopotential height (GPH;m), at 10 hPa: N48, N96, and ERA-40.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 7.
Fig. 7.

Final warming times for N48 and N96, at 60°S and 60°N. Shading shows region within 1 interannual std dev of ERA-40.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 8.
Fig. 8.

Profiles of resolved wave drag (DIVF) and orographic and nonorographic gravity wave drag (OGWD and NOGWD). Units are m s−1 day−1. Values are shown for N48 and for the difference N48 − N96. Positive values are shown with solid contours and negative values with dashed contours.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 9.
Fig. 9.

(a) Annual mean mass streamfunction (kg s−1 m−1) at 70 hPa, calculated from residual vertical velocity w* and downward control; N48 ensemble mean. The contributions to the downward control estimate from individual wave drags are also shown. (b) Annual mean mass streamfunction (kg s−1 m−1) calculated from w*; N48 ensemble mean. (c) As in (b), but using downward control (DC). (d) Planetary wave component of (c). (e) Orographic GWD component of (c). (f) Nonorographic GWD component of (c).

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 10.
Fig. 10.

Vertical profile time series of monthly mean water vapor volume-mixing ratio (ppmv) between 5°S and 5°N, for (a) N48 and (b) N96.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Fig. 11.
Fig. 11.

Climatological monthly mean specific humidity (10−7 kg kg−1) at 100 hPa for (a) N48 January, (b) N96 January, (c) N48 July, and (d) N96 July.

Citation: Journal of the Atmospheric Sciences 67, 5; 10.1175/2009JAS3337.1

Table 1.

Wavenumber contributions to peak amplitude of 10-hPa azonal component of geopotential height.

Table 1.

1

Note that despite the use of the usual term “gravity wave drag,” gravity waves can act to both accelerate and decelerate the flow.

Save