S P A R C Stratospheric Processes And their Role in Climate A project of the World Climate Research Programme Home Initiatives Organisation Publications Meetings Acronyms and Abbreviations Useful Links

The Darwin Area Wave Experiment (DAWEX)

Kevin Hamilton, University of Hawaii, Honolulu, Hawaii, USA (kph@soest.hawaii.edu)

(For a better resolution of the images, please click on the plot)

Participants in the Workshop on Analysis of DAWEX Results, Honolulu, Dec. 3-5, 2002 1st row: S. Suzuki, A. Liu, T. Tsuda, K. Hamilton, J. Alexander 2nd row: K. Shiokawa, M. Taylor, P. May, A. McKinnon, B. Vincent, J. Hecht, R. Walterscheid, G. Stenchikov

Introduction

The study of high-frequency gravity wave (GW) motions in the middle atmosphere (MA) has been a leading area of research in both meteorology and aeronomy. This interest has arisen since GW motions often dominate the variance observed in the upper mesospheric and thermospheric wind field, and since the vertical transport of mean flow momentum by GW is thought to play a key role in maintaining global-scale circulation throughout the MA. A key issue for modelling stratospheric climate and chemistry is parameterizing the unresolved GW effects on the resolved flow. A major difficulty in formulating credible GW parameterizations is the lack of empirical data on the details of the wave spectrum in the MA and its relation to sources such as moist convection, flow over topography and jet stream instabilities.

Discussions in the community beginning with the December 1996 SPARC SSG meeting in Adelaide raised the idea of a field experiment in the Darwin, Australia area to study the GW excited by deep convection in the pre-monsoon period. Particular interest was expressed in studying the so-called "Hector" convection that appears regularly in the afternoon over the Tiwi Islands located north of Darwin. In the Austral winter, convection over Northern Australia is rather limited. However, in the late pre-monsoon, typically from November to mid-December, intense convection over the Tiwi Islands, known locally as "Hector", develops virtually every afternoon. It is generally initiated by interaction of convectively driven cold pools and sea breeze fronts and builds to a bundle of intense thunderstorms over the western portion of the Islands, reaching the tropopause in mid-afternoon. The Hector thunderstorms represent some of the most intense and penetrative convection observed anywhere, with updrafts as strong as 40 m/s and cloud tops sometimes above 20 km (e.g. Keenan et al., 2000).

The SPARC GW Initiative co-chairs formalized the DAWEX proposal in 2000 (Hamilton and Vincent, 2000), and the field campaign took place during October-December 2001. Preliminary results of the experiment were recently reviewed at a workshop in Honolulu (USA) from December 3-5, 2002.

The Scope of the Experiment

DAWEX involved contributions from scientists from at least ten different institutions in Australia, Japan and the USA. The field campaign was organised around three intensive observing periods (IOPs): October 13-18, November 15-20 and December 11-16, but many instruments were deployed before and after the IOPs.

The convection over the Tiwi Islands and elsewhere in the Darwin area was observed in detail by the Australian Bureau of Meteorology Research Centre C-band polarized Doppler (CPOL) radar located near Darwin, while the occurrence of convection over a larger area was monitored by weather radar and geostationary satellite IR imagery.

Wave responses in the vicinity of the mesopause were observed by a total of five airglow imagers deployed during the experiment. Figure 1 shows the estimated useful fields-of-view near 90 km altitude for each of the imagers. A boundary-layer radar that profiled the three components of the wind up to about 7 km height was installed at Pirlangimpi on the Tiwi Islands. At Katherine, about 400 km south and east of the Tiwi Islands, a medium-frequency radar was installed to measure horizontal winds near the mesopause.

Figure 1. Locations of some observational sites in the DAWEX field campaign. The triangle shows Pirlangimpi on the Tiwi Islands. The five closed contours show the estimated useful fields-of-view near 90 km for the airglow imagers at Wyndham (15.3S, 128.1E), Darwin (12.3S, 130.5E), Katherine (14.3S, 132.2E), Alice Springs (23.5S, 133.5E) and Adelaide (34.5S, 138.3E).

Three-hourly balloon soundings of the wind and temperature were made during each of the IOPs at Pirlangimpi, Darwin (130 km south) and Katherine. The soundings were made with larger balloons than used routinely; in practice over 75% of the IOP soundings reached 25 km and over 65% reached 30 km.

Some Preliminary Results

The character of the convection differed among the three IOPs as the prevailing circulation changed from pre-monsoonal to monsoonal. The October IOP had three days with modest Hectors and two with only disorganised convection. In the November IOP there were three consecutive days (15,16 and 17) with very strong Hectors as well as more extensive areas of convection. In the December IOP a more monsoonal pattern with widespread convection over the mainland prevailed. Figure 2 shows a CPOL radar section through the intense Hector.

Figure 2. Radar reflectivity measured in a vertical scan from the CPOL radar looking approximately northward from near Darwin at 15:48 local time on November 15, 2001. This shows a well-developed Hector storm with radar echoes up to near 20 km. Courtesy Peter May.

The airglow imagers yielded a rich trove of data indicating the horizontal wavelengths and horizontal phase velocities of waves propagating through the field of view of each imager. One overall tendency observed at all the imagers is dominant southward propagation. This suggests that the dominant source for waves in this season is convective activity in Northern Australia or the Maritime Continent. More detailed analysis of the behaviour of waves at each station and its relation to the weather each day is now being performed.

The radiosonde observations revealed some systematic wavelike variations at both short time scales and longer time scales. One prominent component near the tropopause in the October and December IOPs is a large-scale wave (coherent at Darwin and Katherine) with period near 84 hours (Figure 3). Overall there is tendency for small vertical scale variations to be strong near the tropopause and also above 25 km, but to be suppressed in the 20-25 km height range. All these findings are unexpected and await satisfactory explanations.

Lessons for Future Campaigns

There was considerable enthusiasm at the Honolulu workshop for another field campaign focussed on GW generation by convection in Northern Australia. A new experiment would benefit greatly by extending the observational capabilities available for DAWEX. Valuable additions would be provided by Rayleigh lidars to obtain information of wave fluctuations in the 30-70 km height range, and also by deployment of airglow imagers on aircraft to allow a broader geographical coverage. An experiment that could extend over a long period of time would also be useful, as it would allow sampling of days with a wide variety of weather situations. It was noted, in particular, that the Austral spring season of 2002 featured at least one day with a strong Hector storm that was rather isolated from any other significant convection, a situation that did not occur so clearly during DAWEX.

Figure 3. Profiles of meridional wind measured at roughly three-hourly intervals at Darwin and Katherine during the October 2001 DAWEX IOP. Successive profiles are displaced rightward on the plot. Courtesy Toshitaka Tsuda.

References

Hamilton, K. and R.A. Vincent, Experiment will examine gravity waves in the middle atmosphere. Eos, 81, 517, 2000.

Keenan, T. et al., The Maritime Continent Thunderstorm Experiment (MCTEX): Overview and some results. Bull. Amer. Met. Soc., 81, 2433-2455, 2000.