A SUMMARY OF SOME OF THE SCIENTIFIC HIGHLIGHTS
Since the discovery of the ozone hole in the mid 1980s, great progress has been made in understanding the dynamical, chemical and transport processes that occur in the stratosphere. At the same time, the importance of the stratosphere as an integral part of the climate system has come to be more fully appreciated. Through exchanges of mass, momentum and energy, the stratosphere is strongly coupled to the climate system as a whole. Relevant processes are commonly non-linear (for instance, the dynamical coupling between the troposphere and stratosphere), and by no means fully understood. SPARC has been instrumental in promoting the science that has led to a wider appreciation of the importance of the stratosphere in climate. It has also been active in promoting greater integration between scientific disciplines involved in the broader World Climate Research Programme.
The Scientific Organising Committee of the 2nd SPARC General Assembly (SPARC 2000) aimed to structure the scientific meeting in a way that emphasised the importance of the stratosphere in the wider context of climate change. The five-day meeting comprised four sessions. Session 1 emphasised fundamental processes and interactions among processes. Session 2 discussed observations relevant to these processes and indicative of climate variability and trends. Session 3 focused on modelling stratospheric processes and climate variability, and was designed to present a synthesis of our current understanding. Session 4 was devoted to observations and modelling of solar ultraviolet radiation, a particularly timely topic during a year in which the Antarctic ozone hole covered the greatest area yet observed, extending for a time over the southern tip of South America.
Each of these sessions was accompanied by poster sessions, the scientific quality and breadth of which were outstanding. It is clear that the convenors of future SPARC General Assemblies must be vigilant in structuring the meeting to give equal emphasis to oral and poster sessions.
SPARC 2000 was attended by over 300 scientists from more than 30 countries. All Latin American countries were represented. Attendees will remember not only the high scientific quality of the meeting, but also the warmth of the reception given by our hosts in Argentina, their hard work over many months to make the meeting very enjoyable, and the generosity of the sponsors that enabled so many young scientists to attend.
The Chair of the Scientific Organising Committee, Alan ONeill, wishes to record here his thanks to the Scientific Conveners for their diligent work. He extends particular thanks, on their behalf, to the Chair of the Local Organising Committee, Pablo Canziani, for his unstinting efforts to ensure that the meeting was a success. A summary of some of the scientific highlights is given below.
SESSION 1: STRATOSPHERIC PROCESSES AND THEIR ROLE IN CLIMATE
(Convener: T. Shepherd)
Understanding the role of the stratosphere in the climate system interpreting indicators of climate change, developing credible models, providing a framework for diagnostic studies, and quantifying and interpreting changes in UV radiation all require a fundamental understanding of the basic physical processes involved. This was the rationale for Session 1, which to some extent underpinned the following sessions. A major focal point was the upper troposphere/lower stratosphere (UT/LS), also known as the dynamical "middleworld". The UT/LS is pivotal and complex: it is the interface between the troposphere and stratosphere, so it is where exchanges of material occur; radiative and chemical time scales are relatively long, so transport is very important; low temperatures imply a role for condensed matter in the chemical balance; and there is a sensitive radiative feedback from relatively short-lived greenhouse gases (water vapour and ozone). However, the entire stratosphere affects the UT/LS region. It does so dynamically through "downward control" and the diabatic circulation, chemically through transport (the Brewer-Dobson circulation) and through filtering of actinic fluxes, and radiatively through both short wave and long-wave fluxes.
The papers in Session 1 can be grouped into five main areas: chemistry, transport, clouds and water vapour, gravity waves, and climate variability and tropical oscillations. Because of the large number of papers, only the salient themes are highlighted.
An invited overview talk by A.R. Ravishankara (NOAA Aeronomy Lab) highlighted some of the complexities of the chemistry of the UT/LS region. He noted the unique aspects of ozone: its spatial inhomogeneity, its role in both UV and IR radiation, its production within the atmosphere, and the strong role of chemistry in its abundance. The complex role of NOx in the ozone budget, still not fully understood, was discussed; it was suggested that the summer polar stratosphere, being under photochemical control, is a good place to test our chemical understanding. Finally, the UT was noted to be chemically very fertile, and the possible role of chloral was raised. The role of NOx in the UT/LS region was also discussed in the paper by S. Meilinger (MPI, Mainz), given by T. Peter, which emphasised the need to distinguish carefully between the LS and the UT; e.g. for high NOx, NMHCs can have a large effect on ozone production just below the tropopause.
The global modelling of chemical climate using general circulation models has emerged as a major evolving area of research. While several years ago only a few groups were active, now there is a wealth of activity in various countries. Results were presented on the Japanese CCSR/NIES model (K. Sudo), the NCAR model (F. Sassi, D. Kinnison), the Canadian MAM model (A. Jonsson), the French MOCAGE model (H. Teyssedre), the German KASIMA model (T. Reddmann), and the Utrecht model (B. Bregman).
Late-winter/spring polar ozone depletion continues to be a focus of investigation. M. Rex (Wegener Inst., Potsdam) gave an update on the latest results from the MATCH program, which identifies chemical ozone loss in the Arctic on specific air parcels through the combined use of trajectory modelling and ozonesondes. The coverage is sufficiently dense that an estimate of the net ozone loss over the Arctic can be obtained for the late-winter/spring season. Antarctic ozone loss was estimated in a paper by B. Connor (NIWA, Lauder). With regard to polar processes relevant to ozone depletion, the role of mesoscale processes in Arctic PSC formation was discussed by M. Mueller (Free Univ., Berlin) using lidar measurements, while denitrification was identified in the Arctic by H. Oelhaf (FZK Karlsruhe) using a combination of MIPAS-B measurements and modelling, and in the Antarctic by R. de Zafra (SUNY, Stony Brook) using mm-wave spectroscopic measurements of various species.
An invited talk by P. Haynes (Cambridge Univ.) presented the notion that the mid-latitude tropopause is a (lateral) mixing barrier, which emerges naturally in idealised studies of baroclinic instability. In this view, the extratropical tropopause is the transition level between altitudes which experience widespread mixing (the tropopause) and those which are constrained by this mixing barrier (the lowermost stratosphere). The subject of mixing and transport across the extratropical tropopause was addressed in several other papers: by E. Shuckburgh (Cambridge Univ.) and R. Scott (Laboratoire d'Aérologie, Toulouse) using analysed winds, by A. Zahn and H. Fischer (both MPI, Mainz) using measurements from the CARIBIC and STREAM aircraft campaigns, respectively, and by Y. Rao (IITM, Pune) and Y. Tomikawa (Tokyo Univ.) using radar data.
Another mixing barrier exists on either side of the tropical stratosphere, and there has been much interest in recent years in quantifying the upwelling in the "tropical pipe" and the associated mixing rates. M. Volk (Frankfurt Univ.) provided latest estimates of the mixing rates using Geophysica data from the APE-THESEO campaign, which were complemented by a study using both lidar measurements and modelling by H. Bencherif (Réunion Univ.). The representation of the tropical tape recorder in general circulation models was found to be highly sensitive to the choice of numerical advection scheme in studies by K. Nissen and V. West (both Edinburgh Univ.) and S. J. Lin (NASA/GSFC). In a somewhat different but related study addressing the origin of the tropical upwelling, R. Garcia argued that a transient calculation was required in order to account for the seasonal cycle of upwelling.
Yet another mixing barrier exists on the edge of the wintertime stratospheric polar vortex, and several studies were devoted to quantifying mixing across the vortex edge and within the neighbouring surf zone. This remains an important issue for quantifying polar ozone loss, especially in the Arctic. The papers included several studies with the French MIMOSA model by F. Fierli, A. Hauchecorne and M. Marchand (Service d'Aéronomie, Paris) and one by the Berlin model by K. Krueger (Free Univ., Berlin). A variety of measurement platforms are available in the Arctic; papers were presented using ILAS satellite data by H. Nakajima (NIES, Tsukuba), L. Pan (NCAR), and W. Choi (Seoul National Univ.), by D. Gibson-Wilde (CRA, Boulder) using ALOMAR ground-based data, and by C. Basdevant (LMD, Paris) using trajectory methods.
An invited talk by T. Peter (ETH, Zurich) addressed the question of the composition and origin of the subvisible cirrus clouds located at about 17.5 km altitude in the tropics. They are seen only in certain backscatter wavelengths, and sit several km above, and are apparently unconnected with, the thick cirrus associated with convection. T. Peter argued that it seems difficult to account for their existence unless a rather large upwelling of about 6 mm/s, which is an order of magnitude larger than the large-scale upwelling inferred on radiative ground, can be hypothesised. Some questioners wondered whether inertia-gravity waves might possibly be responsible, and this possibility was in fact discussed further in a paper by F. Hasebe (Ibaraki Univ.). The APE-THESEO campaign that observed these subvisible cirrus clouds was presented in a paper by L. Stefanutti (CNR IROE, Firenze).
The issue of tropical dehydration proved to be a lively topic. X. Zhou (Washington Univ.) attempted to associate variability in the cold point temperature with the QBO and ENSO. S. Sherwood (NASA/GSFC) examined the role of convective overshoots, noting the existence of a transition zone between the troposphere and stratosphere wherein different kinds of air coexist. A. Gettelman (NCAR) noted that the diurnal cycle in tropopause temperature (of about 1.5 K) over land, with a much smaller variation over the ocean, implicates the role of convection, but that convection above the tropopause is very rare, perhaps occurring less than 1% of the time across the tropics.
D. Kley (Juelich) presented the findings of the SPARC Water Vapour Assessment Report (WAVAS), which was recently completed and which focuses on the upper troposphere and lower stratosphere (see report in this newsletter). The report discusses the various measurement techniques, the data quality, limitations of data sets and advice on how to combine them (e.g. for trend studies), and the current understanding of the distribution and variability of water vapour. Copies of the report are available from the SPARC Office.
An invited talk by T. Tsuda (RASC, Kyoto) described the new GPS/MET data set which is providing an unprecedented global view of gravity-wave activity in the lower stratosphere. The data show stronger gravity-wave activity in the tropics than in the extratropics, presumably associated with stronger convective activity, and interest naturally focuses on quantifying the connection between convection and gravity-wave excitation. This quantification is of great importance for climate modelling. Several papers investigated this connection: T. Kerzenmacher (Wales Univ.) using radiosonde analyses, and Z. Eitzen (Colorado State Univ.) and Z. Chen (LASG, Beijing) using idealised model simulations. N. McFarlane (CCCma, Victoria) reported results from the Canadian middle atmosphere GCM showing that the nature of the convective adjustment scheme had a strong effect on the gravity waves (including equatorial waves) generated in the model.
While GPS/MET is providing a global view from satellites, SPARC has been co-ordinating the development of a global gravity-wave climatology from high-resolution radiosonde data. Even though this has the obvious spatial biases, it does provide excellent vertical resolution. R. Vincent (Adelaide Univ.) gave an overview paper on the SPARC radiosonde initiative, while T. Birner (DLR, Oberpfaffenhofen) reported on German results, M. Geller (SUNY, SB) on US results, and I. Son (Yonsei Univ.) on Korean results. These studies are beginning to show links with source mechanisms. In related papers, E. Pavelin (Wales Univ.) studied gravity waves in a field experiment over Aberystwyth, while K. Sato (Kyoto Univ.) identified long-lived layered structures in radiosonde observations over Japan.
Gravity waves are important in the middle atmosphere because of their effect on mixing and momentum transport. D. Fritts (CRA, Boulder) reported on high-resolution numerical simulations directed at quantifying the generation of turbulence by breaking gravity waves. J. Alexander (also CRA, Boulder) derived constraints on the gravity-wave forcing of the atmospheric circulation based on UARS satellite observations to infer the radiative heating and cooling, while H. Chun (Yonsei Univ.) did a similar thing based on closing the momentum budget in the UKMO analyses. The important question of parameterising the effects of unresolved gravity waves in climate models was addressed in papers by R. Tailleux (LMD, Paris) and M. Charron (MPI, Hamburg).
It was at the first SPARC General Assembly in Melbourne in 1996 that the first results were announced concerning the simulation of a QBO-like oscillation in a (simplified) high-resolution GCM (by M. Takahashi). Since then there has been a lot of progress, with QBO-like oscillations first simulated in more realistic but still high-resolution GCMs, and most recently in coarse-resolution GCMs. The missing ingredient seems to have been momentum transport by small-scale gravity waves, which can either be resolved (at very high resolution) or parameterised. The latter possibility is crucial if GCMs are to include this most important mode of atmospheric variability, and yet run long enough to produce climate simulations. The focus now is on understanding why models produce QBO-like oscillations, and whether they do so for the right reasons. There were three papers on the presence or absence of such oscillations in coarse-resolution models: A. Scaife (UKMO) using the UKMO unified model, C. McLandress (Toronto Univ.) using the Canadian middle atmosphere model, and M. Giorgetta (MPI, Hamburg) using the MAECHAM model. McLandress noted that the period of the oscillation depended on the finite differencing used to determine momentum flux divergence in the parameterisation scheme, suggesting that much more work is needed to assess the reliability of these kinds of results.
A large component of climate variability in the stratosphere concerns ozone and temperature in mid-latitudes and in polar regions. Understanding this variability, and its possible link with the troposphere, is crucial for delineating anthropogenic ozone loss and climate change. There were several papers on this topic; e.g. R. Bernardi (Univ. Republica, Uruguay) linked changes in the SH vortex to SST anomalies, P. Canziani (Buenos Aires Univ.) linked total ozone anomalies (or 'mini-holes') to synoptic variability, and L. Hingane (IITM, Pune) linked the evolution of the mini-holes to the monsoon circulation.
SESSION 2: STRATOSPHERIC INDICATORS OF CLIMATE CHANGE
(Convener: V. Ramaswamy)
This session focussed on recent research concerning stratospheric indicators of climate change. The subjects covered included variations and changes in trace species such as methane and other well-mixed gases, ozone, water vapour, aerosols, polar stratospheric and cirrus clouds; temperature; tropopause features; stratospheric circulation; and forcing/ variability of stratospheric ozone and climate. Both the oral talks and posters on display yielded new insights besides covering a wide array of technical issues related to the above topics.
The invited presentation (W. Randel) pointed out the prominent as well as unusual features in the observations of the stratosphere during the 1990s, e.g., the variation in the Brewer-Dobson circulation and eddy wave fluxes, the accompanying variations in methane and water vapour, and trends in temperatures especially in the northern polar regions during springtime. Besides influences due to solar cycle and volcanoes, there is also a significant coupling with the climate of the troposphere.
The availability of long (~a decade or more) time series of several relevant stratospheric parameters, as measured by satellites and ground based instruments, have initiated analyses studies seeking to understand causes of the associated phenomena. Besides ozone which was the focus of several papers, recent observations of stratospheric water vapour have generated a lot of interest, in part because of the importance of the apparent changes for stratospheric temperature trends/climate and stratospheric chemistry, and in part because the basic causes are not well understood.
Findings of the recently completed SPARC water vapour assessment (see also Session 1 report) suggest significant increases in the lower stratospheric water vapour content over the past few decades.
Investigations of the long-term (~2 decades or more) temperature trends in the global stratosphere based on measurements from a number of platforms reveal a general cooling, but with distinct seasonal and latitudinal variations; however, there is a wide divergence in the estimates for the mesosphere. There are some problematic issues regarding the currently (and the only one) deployed AMSU satellite instruments' ability to detect climate change in the stratosphere. Stratospheric aerosol concentrations appear to be at their lowest values now compared to the late 1970s when such global observations were begun. The utility of long-term lidar measurements at a specific site to diagnose the climatology of locally present cirrus clouds was demonstrated. The tropical upper tropospheric and tropopause regions are generating substantial research interest. Radiosonde measurements indicate correlation between surface and stratospheric temperatures, with a cooling of the lower stratosphere during El Niño events, although the sonde data quality at the lower stratospheric altitudes have to be accounted for with some care. Intraseasonal and longer-term variations in the tropical tropopause are inferred from observations and analysed fields.
Diagnostic investigations, based on observations and models, have explored the possible links between stratospheric climate, ozone and tropospheric features. Solar variations and their possible modulation of the QBO could be enabling a solar component in the long-term variability of the stratosphere. The plausible connection between interannual changes in stratospheric circulation, tropospheric structure and total ozone may bear on the causes of the observed stratospheric ozone and temperature trends in the northern mid-latitudes. The modes of atmospheric variability, in particular involving planetary wave propagation, could have substantial implications for climate change and detection.
While the oral presentations exposed several important contemporary issues concerning stratospheric indicators of climate change, the poster sessions provided the setting for amplifications on several of the above issues. In addition, the poster sessions reported important outcomes or anticipated results from several campaigns or new measurement techniques or new diagnostic methods. The details showcased the in-depth scrutiny and research underlying the diagnostic analyses of observations e.g., different kinds of problems affecting interpretation of observed phenomena, sampling problems and instrument uncertainties, temporal discontinuity difficulties, inconsistencies between various observational platforms, global versus regional features, and accounting for the natural internal variability of the stratosphere.
Nonetheless, the ability to synthesise the numerous observations available now, and the resulting understanding of the variations in stratospheric features, have advanced considerably since the First SPARC General Assembly and have opened up a host of new research opportunities.
SESSION 3: MODELLING AND DIAGNOSIS OF STRATOSPHERIC EFFECTS ON CLIMATE
(Co-conveners: S. Yoden and V. Ramaswamy)
This session was focussed on the synthesis of stratospheric processes in the context of climate interpretation and prediction. Issues concerning stratospheric climate and effects of stratospheric processes on the lower atmosphere were discussed based on both numerical modelling and diagnostic studies involving observations. It was divided into four sub-sessions; (3-1) Climatology, (3-2) Internal variations in S-T coupled system, (3-3) Responses to forcings, and (3-4) Trends. Regular periodic annual cycle or internal variations of the stratosphere-troposphere coupled system were discussed in the first two sub-sessions, while responses to "external" forcings were discussed in the rest.
In the first sub-session of climatology, M. Chipperfield (Leeds Univ.) gave an invited talk on the current status of three-dimensional global chemical transport models (CTMs) and their use to understand the interaction of ozone depletion and climate. R. Kawa (NASA/GSFC) showed the capability of such CTMs by the comparison with atmospheric measurements on the winter of 1999-2000 in the Northern Hemisphere. We also had active discussions on the interactions between chemistry and climate in the focused discussion session in which several new results with different hierarchy of numerical models (2-D or 3-D, CTM or fully coupled chemistry-climate (dynamics) model) were presented. Several other papers in this sub-session dealt with the GRIPS(GCM Reality Intercomparison Project for SPARC) initiative. For instance, T. Horinouchi (Kyoto Univ., presented by S. Pawson of NASA/GSFC) gave a clear dependence of the generation of vertically propagating waves in the tropics on the cumulus parameterisation scheme in each GCM.
The subject of the second sub-session was the internal variations in the stratosphere-troposphere coupled system, and the hot issue was the Arctic oscillation (AO), or the annular mode. More than ten papers were submitted on this subject. M. Baldwin (NWRA) described the observed deep, longitudinally symmetric patterns of low-frequency variability and the downward propagation of the AO signature, while A. O'Neill (Reading Univ.) pointed out the importance of the three-dimensional structure in the AO. Not only observational studies but several types of numerical studies were presented on the AO-like internal variations. In the invited talk of this sub-session, D. Shindell (NASA/GISS) showed that the AO patterns are frequently obtained in responses to greenhouse gas, ozone, solar and volcanic forcing in climate change simulations. Other internal variations with intraseasonal and interannual time scales were also discussed intensively in this sub-session.
The third sub-session of responses to forcings were divided into three: (1) QBO and ENSO, which are, in a sense, "external forcings" to the extratropical stratosphere, (2) solar forcing with the 11-year cycle, and (3) volcanic aerosols. Some fully coupled chemistry-general circulation models were used to investigate the responses. J. Haigh (Imperial College) gave an invited talk on the response to solar variability, and A. Robock (Rutgers Univ.) summarised the impact of the 1991 Mt Pinatubo eruption based on the Pinatubo Model Intercomparison Project. On the other hand, V. Ramaswamy (GFDL) pointed out the difficulty of getting statistically significant results in the high-latitude winter due to large internal variability as a result of the GFDL "SKIHI" experiments.
The last sub-session on trends, which was also the last part of the assembly, had an invited talk by D. Karoly (Monash Univ.) who gave a perspective of the Intergovernmental Panel on Climate Change for detection and attribution of a stratospheric role in climate change. Stratospheric forcing processes include not only natural forcing variations discussed in the third sub-session but also anthropogenic variations such as in greenhouse gases or stratospheric ozone. Model studies of stratospheric trends in ozone, temperature and water vapour were reported by several authors. For example, B. Boville (NCAR) showed a result of climate change simulations with the NCAR Climate System Model (a coupled ocean-atmosphere general circulation model) for the period of 1870-2100 with several scenarios of trace gas changes. Importance of the modulation due to large internal dynamical variations was repeatedly pointed out during the sub-session.
SESSION 4: UV OBSERVATIONS AND MODELLING
(Conveners: S. Diaz and R. McKenzie)
The UV session invited papers on measurement studies as well as modelling studies of UV radiation in the troposphere and at the Earth's surface. We were particularly interested in receiving contributions that relate to changes in UV to changes in ozone, cloud, and aerosols. Particular emphasis was placed on the co-ordination of ground-based measurements of UV to assess the accuracy of the estimations of UV derived from satellites since in years to come it is expected that these will play a major role in UV radiation assessment, in the same way that satellite derived ozone has already proved crucial to our global understanding of ozone depletion.
This session attracted more interest than expected, with over 50 papers being accepted. Only 10 of these were given oral presentations, and it could be argued that greater emphasis be given to this area in future assemblies of SPARC, since changes in UV are the major driving factor for this research. There was a good mix of topics and geographical coverage: it was especially pleasing to see an excellent representation of the wide range of research into UV and its impacts that is being undertaken in South America.
The posters presented were of a very high quality and the focused discussion sessions provided an excellent forum for informal discussion with the authors, despite the inclement weather (but low UV) which prevailed most of the time at the meeting. They attracted a lot of interest among the atmospheric scientists as well as with the biological scientists who convened the associated workshop on impacts of UV radiation on terrestrial and aquatic ecosystems on the Saturday following the SPARC meeting.
Several papers (orals and posters) focused on relating ground-based measurements of UV to satellite-derived estimates of UV. Most of these compared of ground-based measurements of UV with estimates using data products from the TOMS instruments. There was a general consensus that while satellite products provide a useful global coverage of UV patterns at the surface, significant discrepancies remain. The discrepancy appears to be related to uncertainties in allowing for tropospheric extinction (e.g. from ozone and aerosols). At unpolluted sites differences between satellite-derived estimates of UV and ground-based measurements are smaller. Improvements may be achievable with a better understanding of regional scale differences in tropospheric extinction and with higher resolution topography and cloud imagery.
Several speakers (including the invited speakers to this session Drs Bais and Seckmeyer) discussed the present status of UV research, and the extent to which recent internationally co-ordinated activities have improved the measurement accuracy and our understanding of processes that affect UV radiation. One of these was the IPMMI (International Photolysis Frequency Measurement and Model Intercomparison) campaign, which focused primarily on actinic radiation. It was argued (S. Madronich) that the effects of air pollution can be larger on actinic UV fluxes than irradiances and that more emphasis should be given to measuring actinic fluxes, particularly in polluted environments.