Airborne measurements of gravity wave breaking at the tropopause

Size: px

Start display at page:

Download "Airborne measurements of gravity wave breaking at the tropopause"

Cody Briggs
5 years ago
Views:

1 GEOPHYSICAL RESEARCH LETTERS, VOL. 30, NO. 20, 2070, doi: /2003gl018207, 2003 Airborne measurements of gravity wave breaking at the tropopause James A. Whiteway, 1,6 Edward G. Pavelin, 2 Reinhold Busen, 3 Jorg Hacker, 4 and Simon Vosper 5 Received 19 July 2003; revised 27 August 2003; accepted 9 September 2003; published 30 October [1] A breaking atmospheric gravity wave was investigated with a combination of airborne in-situ dynamical measurements and ground-based VHF radar observations. The wave was generated by flow over mountains and it was observed to break near the tropopause. The measurements reveal that the wave was overturning at the tropopause and that the initial breakdown into turbulence involved the generation of smaller oscillations with a horizontal wavelength of around 500 m. There was also evidence that the turbulence associated with the wave breaking can cause substantial mixing in the tropopause region. INDEX TERMS: 0341 Atmospheric Composition and Structure: Middle atmosphere constituent transport and chemistry (3334); 3362 Meteorology and Atmospheric Dynamics: Stratosphere/ troposphere interactions; 3384 Meteorology and Atmospheric Dynamics: Waves and tides; 3379 Meteorology and Atmospheric Dynamics: Turbulence; 3334 Meteorology and Atmospheric Dynamics: Middle atmosphere dynamics (0341, 0342). Citation: Whiteway, J. A., E. G. Pavelin, R. Busen, J. Hacker, and S. Vosper, Airborne measurements of gravity wave breaking at the tropopause, Geophys. Res. Lett., 30(20), 2070, doi: /2003gl018207, Introduction [2] At the boundary between the troposphere and stratosphere there is mixing by air motions with scales ranging from thousands of kilometres to micrometers: from planetary waves, through gravity waves, to turbulence, and molecular diffusion. For example, breaking planetary waves regularly displace air pole-ward from the sub-tropical upper troposphere into the mid-latitude stratosphere [Bradshaw et al., 2002]. The amount of irreversible mixing that follows then depends on the location of turbulence in relation to the large-scale flow. It is well known that there is turbulent mixing around the tropopause caused by jet stream shear [Shapiro, 1980; Pavelin et al., 2002], cloud dynamics [Gulteppe and Starr, 1995], and gravity wave breaking [Lilly, 1971; Worthington, 1998; Pavelin et al., 2001; Pavelin and Whiteway, 2002], but there have been very few detailed in-situ measurements at these heights that are capable of resolving turbulent fluctuations. An accurate 1 Department of Physics, University of Wales, Aberystwyth, UK. 2 Department of Meteorology, University of Reading, UK. 3 Institute for Atmospheric Physics, DLR, Oberpfaffenhofen, Germany. 4 Flinders University, Adelaide, Australia. 5 Met Office, UK. 6 Now at Department of Earth and Atmospheric Science, York University, Toronto, Canada. Copyright 2003 by the American Geophysical Union /03/2003GL ASC 11-1 accounting for the role of mixing in determining the chemical composition at the tropopause requires experimental advancement in our knowledge of the sources and properties of the turbulence in this region. The goal of the research reported here was to investigate the role of atmospheric gravity waves in producing turbulence and mixing in the tropopause region. [3] An airborne measurement campaign was conducted during May/June 2000 to investigate mixing processes in the upper troposphere and lower stratosphere. A high altitude aircraft called the Egrett was used to carry instruments for measurements of dynamics and constituents. The focus of the campaign was divided equally between studies of large-scale transport and small-scale mixing. The flights to investigate small-scale mixing were conducted above Wales in order to obtain in-situ measurements of breaking gravity waves that were generated by flow over the Welsh mountains. Ground based VHF radar measurements at Aberystwyth (on the coast of Wales) provided a guide for the flight patterns and also complementary measurements. We report here on the measurements during a flight that encountered breaking mountain waves at the tropopause. 2. Measurements [4] The Egrett is a former reconnaissance aircraft that is now operated for atmospheric research by Airborne Research Australia at Flinders University. It has the unique ability to fly at heights of up to 15 km at a relatively slow airspeed of 100 m/s. This study makes use of measurements with a turbulence measurement system on-board the Egrett. Wind velocity was measured with two separate turbulence probes installed under either wing of the aircraft. One was a standard 5-hole Rosemount probe and the other was a recently developed BAT probe [Hacker and Crawford, 1999]. The BAT probe measurements are being presented in this paper. Correction for aircraft velocity and orientation made use of on-board GPS receivers and high-frequency accelerometers [Crawford and Dobosy, 1997]. The wind was measured accurately to within 10 cm/s in the horizontal and 15 cm/s in the vertical. Data were acquired at a rate of 55 samples per second, corresponding to a horizontal resolution of about 2 m. Such a capability for resolving both waves and turbulence within the stratosphere is unprecedented. This study also involves temperature measurements with a Rosemount PT100 probe and ozone measurements with an in-situ UV absorption photometer (Analytical Systems model TE-49C). [5] The Aberystwyth MST radar [Vaughan, 2002] consists of a array of 4-element Yagi antennas, covering an area of 110 meters square. The emitted VHF radio waves (46.5 MHz) are scattered back from atmo-

2 ASC 11-2 WHITEWAY ET AL.: AIRBORNE MEASUREMENTS OF GRAVITY WAVE Figure 1. (a) Measurements of vertical wind by the Aberystwyth VHF radar. (b) The spectral width of the radar signal averaged between 12:30 and 01:00 UTC. (c) Vertical wind measured on the Egrett. Each flight leg is placed at its height relative to the vertical scale in (a). The topographic height below the Egrett track is shown in green at the bottom with the same relative vertical scale as in (a). The coast of Wales is at 4.1 longitude; the position of the Aberystwyth radar is indicated by the vertical dotted line at 4.0 longitude. Crosses in (a) indicate the time and height when the Egrett passed directly above the radar. spheric refractive index variations that, in the tropopause region, are associated with fine-scale temperature fluctuations. Analysis of the Doppler-shifted spectrum of the backscattered signal measured in various beam directions allows the 3-D wind vector to be determined. The horizontal wind speed in the tropopause region can be measured to an accuracy of within 1.8 m/s over an interval of 5 min. The vertical wind component can be measured to within 0.2 m/s. Also, the Doppler spectral width of the received signal is a measure of turbulence intensity [Pepler et al., 1998]. [6] On 11 May 2000 the radar observed strong mountain wave activity throughout the troposphere with breaking at the tropopause. The Egrett flight pattern was based on the early morning radar measurements and also on a numerical simulation of mountain waves that was derived from meteorological forecast data [e.g., Vosper and Worthington, 2002]. The flight consisted of six straight-line flight legs with vertical spacing of approximately 1 km between heights of 8.4 km and 13.3 km. All of the flight legs were parallel, oriented east-west, passing directly over the Cambrian Mountains, Aberystwyth, and Cardigan Bay. The orientation was chosen so that the flight direction was along the predicted wave vector, perpendicular to the wave phase fronts. The surface height beneath the Egrett flight path is depicted at the bottom of Figure 1. [7] Figure 1 shows the combined radar and Egrett measurements of the breaking gravity wave. In the troposphere, below a height of 11.5 km, the radar measurements exhibit modulations in the vertical wind with amplitude greater than 1 m/s and with phase changes occurring about twice per hour. This pattern is typical of what is regularly observed at Aberystwyth during periods of enhanced mountain wave activity [e.g., Worthington, 1998]. The modulations appear in vertical columns because the vertical wavelength of mountain waves is very long in the troposphere. Using the gravity wave dispersion relation, the vertical wavelength can be estimated from the component of the horizontal wind speed in the direction of wave Figure 2. (a) Vertical profile of eastward (u) and northward (v) horizontal wind components. (b) Vertical profiles of ozone mixing ratio (thick line) and potential temperature measured on the Egrett while ascending toward the west above Wales along the same track as in Figure 1. The dotted lines are at the heights of the flight legs shown in Figure 1.

3 WHITEWAY ET AL.: AIRBORNE MEASUREMENTS OF GRAVITY WAVE ASC 11-3 propagation multiplied by the buoyancy period [Nappo, 2002]. In this case a suitable approximation to the wave propagation direction is eastward (facing upwind at a height of 1 km) and thus the westward wind component is used for estimating the vertical wavelength. The vertical profile of wind velocity measured by the radar is shown in Figure 2. Between the heights of 10 km and 12 km the wind speed decreased from 20 m/s to 5 m/s, and this would have caused the vertical wavelength to decrease from 6 km to 2.7 km. It will be shown below that this decrease in vertical wavelength would cause the wave to break by overturning just below the tropopause in the 11 to 12 km height range. The turbulence generated by the wave breaking is detected as an enhancement of the radar spectral width in the km height range in Figure 1b. The Richardson number derived from the background temperature and wind profiles in this region does not drop below unity, so the turbulence is not generated by shear in the background wind. [8] The Egrett measurements of vertical wind in Figure 1c clearly illustrate the horizontal and vertical structure of the waves and turbulence in the tropopause region. The focus of this study is on the distinct mountain wave above the Aberystwyth radar site ( 4 longitude). The wave amplitude grows with height, as it must in response to the decreasing atmospheric density, reaching a maximum in flight leg D at a height of 10.6 km. The wave is starting to overturn and break into turbulence within flight leg C at a height of 11.4 km. One km above, in flight leg B at a height of 12.3 km, there is only turbulence remaining where the mountain wave would have propagated if it did not break. A separate wave of differing wavelength is detected up-wind in flight leg B. There is substantial turbulence in Flight leg E (height 9.6 km), which is also associated with the wave system, but this is not given further attention in this letter. 3. Wave Breaking by Overturning [9] Figure 3 shows measurements along flight leg C where the wave is starting to break. The vertical displacement in Figure 3b is computed by integrating the vertical wind along the horizontal wind direction (westward). The variations in temperature and horizontal wind along this flight track are consistent with vertical advection of the background wind and potential temperature gradients. The expected temperature deviation determined from the vertical displacement is shown in Figure 3c and this is a close match to the measured temperature. [10] It can be demonstrated that the wave is overturning by showing that it is inducing a convectively unstable temperature gradient. Determination of the temperature gradient induced by the wave requires knowledge of the vertical wavelength. In flight level C this can be estimated as 4.5 km using a background wind speed of 10 m/s (Figure 2). With a peak-to-peak temperature perturbation amplitude of 5.5 C (Figure 3c), the maximum waveinduced gradient is 3.8 C/km (A2p/l, where l is wavelength and A is amplitude). Combined with the background temperature gradient of 5.3 C/km, the maximum negative temperature gradient is 9.1 C/km. This is very near to the threshold of 9.8 C/km for convective instability, equivalent to overturning the density stratification. The wave is Figure 3. Measurements along flight leg C (height 11.3 km) of (a) vertical wind, (b) calculated vertical air displacement, (c) the potential temperature, and (d) eastward wind. The dashed line in (c) is the potential temperature estimated by using the calculated vertical displacement. The dotted lines in (a) contain the region that is shown in Figure 4. thus starting to break at flight level C. At a height of 12 km the vertical wavelength would be reduced to 2.7 km. If the wave maintained its amplitude then it would certainly be overturning below flight level B. [11] Overturning occurs where the wave induced horizontal wind perturbation is greater than the wave phase speed. For a stationary mountain wave, overturning happens when the perturbation wind is equal to or greater than the background wind. The wave induced horizontal wind perturbation in Figure 3d has a maximum value of about 6 m/s at 3.9 longitude. It is seen in Figure 2 that the background wind decreases to 6 m/s just above flight level C at a height of 12 km. We can then assume that the wave is overturning just below 12 km since the measurements also demonstrate that the wave is dissipated into turbulence between flight levels C and B. 4. Transition to Turbulence [12] An intriguing aspect of these measurements is the emergence of secondary smaller scale oscillations that occur at the point of wave breaking around 3.9 longitude in flight level C. The wind and temperature measurements in this region are shown in Figure 4. A high pass filter has been applied in order to isolate the small-scale oscillations. These oscillations have a horizontal wavelength of around 500 m and it is likely that they are associated with the onset of instability in the wave field. As seen in Figure 3, the oscillations occur where the wave is beginning to overturn: just downstream from the largest

4 ASC 11-4 WHITEWAY ET AL.: AIRBORNE MEASUREMENTS OF GRAVITY WAVE upwind from the observed wave breaking, but as the conditions were favourable, it is likely that there was wave breaking east of Aberystwyth as well. [14] The gravity wave was breaking in between flight levels C and B and fortunately the radar detected the resulting turbulence. The radar spectral width, w t, reached a peak value of 0.75 m/s. By applying the work of Weinstock [1978, 1981] and Pepler et al. [1998], the maximum value of the eddy diffusion coefficient for heat, K H, can be estimated (roughly) from the radar vertical spectral width and buoyancy frequency (N s 1 ), as K H 0.05w t 2 /N 2.0 m 2 /s. A similar value of eddy diffusivity was previously obtained from Egrett measurements in turbulence generated by shear above the tropopause [Pavelin et al., 2002]. In that case it was clearly observed that the associated mixing caused substantial changes in the abundance of ozone and water vapour. We can thus conclude here that the turbulence generated by gravity wave breaking can also cause significant mixing in the tropopause region. Figure 4. The perturbations of (a) vertical, (b) eastward, (c) northward wind components, and (d) temperature in the wave breaking region in the vicinity of 3.9 longitude in flight leg C (as indicated in Figure 3a) where there are small scale oscillations. A high pass filter has been applied in order to emphasize the small-scale oscillations. The vertical wind is shown for comparison in each panel as the dotted line. downward displacement and the largest temperature increase. Fluctuations in the northward wind component, in addition to the eastward and vertical components, may be caused by vertical advection of the background shear or could be directly associated with coherent structures. In either case, three-dimensional motions are emerging en route to turbulence. Recent numerical simulations of atmospheric gravity wave breaking have found that the first 3-D motions to emerge in overturning are counter rotating vortices that are elongated along the direction of shear [Fritts et al., 1996; Dornbrack, 1998]. The scale and location of the 3-D oscillations observed here are consistent with the simulations. Separate flights on two other days during the campaign encountered similar small-scale oscillations associated with gravity wave breaking. 5. Mixing [13] If the gravity wave breaking induced strong vertical mixing then we would expect to see evidence in the vertical profiles of ozone and potential temperature. There would be no vertical gradient within an ideal mixed layer. This would be seen as a step in the vertical profile of ozone in which the mixing ratio is constant with height. The vertical profiles of ozone and potential temperature that were measured on the Egrett along the ascent are shown in Figure 2. The ozone mixing ratio and potential temperature both exhibit steps with very small gradient between heights of 11 km and 12.3 km, where the wave was breaking. This is consistent with the hypothesis that the breaking gravity wave was causing substantial mixing. The Egrett actually ascended 6. Conclusions [15] The main goal of the measurement campaign was to obtain high-resolution in-situ measurements of gravity wave breaking and to determine if this caused significant mixing in the tropopause region. Both of these objectives are addressed here. It was observed that wave breaking by overturning was initiated through generation of secondary oscillations with a wavelength of about 500 m. The turbulence generated by the wave breaking was measured to have an intensity that is sufficient to cause significant mixing and evidence of this was present in the measured vertical profiles of temperature and ozone. [16] Acknowledgments. This research was funded by the Upper Troposphere/Lower Stratosphere programme of the UK Natural Environment Research Council (NERC). Meteorological analyses for flight planning were provided by the European Centre for Medium Range Weather Forecasting. The Egrett aircraft is owned and operated by Airborne Research Australia (ARA), of Flinders University in Adelaide. References Bradshaw, N. G., G. Vaughan, R. Busen, S. Garcelon, R. Jones, T. Gardiner, and J. Hacker, Tracer filamentation generated by small-scale Rossby wave breaking in the lower stratosphere, J. Geophys. Res., 107(D23), 4689, doi: /2002jd002086, Crawford, T., and R. Dobosy, Pieces to a puzzle: Air-surface exchange and climate, GPS World, November, Dornbrack, A., Turbulent mixing by breaking gravity waves, J. Fluid Mech., 375, , Fritts, D. C., J. F. Garten, and O. Andreassen, Wave breaking and transition to turbulence in stratified shear flows, J. Atmos. Sci., 53, , Gulteppe, I., and D. O C. Starr, Dynamical structure and turbulence in cirrus clouds: Aircraft observations during FIRE, J. Atmos. Sci., 52, , Hacker, J. M., and T. Crawford, The Bat-Probe: The ultimate tool to measure turbulence from any kind of aircraft (or sailplane), Technical Soaring, 13(2), 43 46, April Lilly, D. K., Observations of mountain-induced turbulence, J. Geophys. Res., 76, , Nappo, C. J., An introduction to atmospheric gravity waves, International Geophysics Series, Vol. 85, Academic, Pavelin, E. G., J. A. Whiteway, and G. Vaughan, Observation of gravity wave generation and breaking in the lowermost stratosphere, J. Geophys. Res., 106, , Pavelin, E., J. A. Whiteway, R. Busen, and J. Hacker, Airborne observations of turbulence, mixing, and gravity waves in the tropopause region, J. Geophys. Res., 107(D10), 4084, doi: /2001jd000775, 2002.

5 WHITEWAY ET AL.: AIRBORNE MEASUREMENTS OF GRAVITY WAVE ASC 11-5 Pavelin, E., and J. A. Whiteway, Gravity wave interactions around the jet stream, Geophys. Res. Lett., 29(21), 2024, doi: /2002gl015783, Pepler, S. J., G. Vaughan, and D. A. Hooper, Detection of turbulence around jet streams using VHF radar, Q. J. R. Meteorol. Soc., 124, , Shapiro, M. A., Turbulent mixing within tropopause folds as a mechanism for exchange of chemical constituents between the stratosphere and troposphere, J. Atmos. Sci., 37, , Vaughan, G., The UK MST Radar, Weather, 57, 69 73, Vosper, S. B., and R. M. Worthington, VHF radar measurements and model simulations of mountain waves over Wales, Q. J. R. Meteorol. Soc., 128, , Worthington, R. M., Tropopausal turbulence caused by the breaking of mountain waves, J. Atmos. Solar Terr. Phys., 60, , Weinstock, J., Vertical turbulent diffusion in a stably stratified fluid, J. Atmos. Sci., 35, , Weinstock, J., Energy dissipation rates of turbulence in the stable free atmosphere, J. Atmos. Sci., 38, , R. Busen, Deutsches Zentrum fuer Luft und Raumfahrt (DLR), Institut fuer Physik der Atmosphaere, Oberpfaffenhofen, D Wessling, Germany. (reinhold.busen@dlr.de) J. Hacker, Airborne Research Australia, Flinders University, P.O. Box 335, Salisbury South, 5106, Australia. (jorg.hacker@airborneresearch. com.au) E. G. Pavelin, Department of Meteorology, University of Reading, P.O. Box 243, Reading RG6 6BB, UK. (e.pavelin@reading.ac.uk) S. Vosper, Met Office UK, FitzRoy Road, Exeter, Devon EX1 3PB, UK. (simon.vosper@metoffice.com) J. A. Whiteway, Department of Earth and Atmospheric Science, York University, 4700 Keele Street, Toronto, ON M3J 1P3, Canada. (whiteway@ yorku.ca)

Gravity wave breaking, secondary wave generation, and mixing above deep convection in a three-dimensional cloud model

GEOPHYSICAL RESEARCH LETTERS, VOL. 33,, doi:10.1029/2006gl027988, 2006 Gravity wave breaking, secondary wave generation, and mixing above deep convection in a three-dimensional cloud model Todd P. Lane