James Hurrell, Harry van Loon, and Shea completed a chapter for the American Meteorological Society (AMS) Monograph No. 49, "Meteorology of the Southern Hemisphere." They described the mean state of the troposphere using 15 years of global analyses and station data. Emphasis was placed on the climate of the Southern Hemisphere, although global maps were presented and some relevant comparisons with the Northern Hemisphere (NH) were made.

Van Loon, Karin Labitzke (Affiliate Scientist), and Silke Leder (both of Freie Universität Berlin, FUB) are working on a comparison between NCEP and FUB stratospheric analyses of the Northern Hemisphere. The interannual variability in the NCEP data is smaller, especially in the equatorial latitudes where the NCEP resolution of the quasi-biennial oscillation is poor.

Volume I of the CLIVAR implementation plan has now been published with major contributions from Trenberth, Co-Chair of the Scientific Steering Group, and also from Hurrell with Bob Dickson (MAFF Fisheries Laboratory) and Mike McCartney (Woods Hole Oceanographic Institution) on the D1 project on "North Atlantic Oscillation." Volume I describes the framework of global analysis and modeling activities plus regionally focussed programs through which CLIVAR will be implemented. It includes the overall vision and strategy for the program, as well as integrating themes, and a brief description of the Principal Research Areas (PRAs). The balance of the material, including details of the PRAs and global integration, will appear in a single-volume initial CLIVAR Implementation Plan, which would be the background for the CLIVAR conference to be held in December, 1998. A draft of this has also been produced. Other editorial contributions to the CLIVAR Newsletter have also been made by Trenberth to help advertise the CLIVAR Program.

David Blankinship is analyzing, with Trenberth, the revised World Ocean Atlas (WOA) data base to determine the heat storage associated with the annual cycle in the ocean, and as a direct follow on, determine the rate of change of that storage that relates directly to heat fluxes. We expect that most of the annual cycle is accounted for by the first two harmonics. Current approaches are making use of complex empirical orthogonal functions. Results will be tested for sensitivity to any small variations in parameters, criteria used, and the effects quantified. In this way we can help to put error bars on the numbers.

Roland Madden, Shea, Rick Katz (Environmental and Societal Impacts Group, ESIG), and John Kidson (National Institute of Water and Atmospheric Research, New Zealand) finished a study of the potential long-range (beyond determinist predictability limits of about two weeks) predictability of precipitation over New Zealand. Results indicate that only 30% or less of the total variance at stations is potentially predictable. Country-wide totals do not improve the situation. Persistence of the ENSO signal may help to realize a part of the potential predictability or about 5% of the total variance.

The review led by Trenberth of the tropical-extratropical interactions and especially the issues involved in determining the response of the extratropical atmosphere to tropical forcing associated with sea surface temperature (SST) anomalies for a special issue of the Journal of Geophysical Research on the Tropical Ocean and Global Atmosphere (TOGA) Program is now in press. A second review by Trenberth on short-term climate variations that describes the role of the TOGA Program in forging progress and the progression to the GOALS program and outlines the main areas believed to be fruitful for future research is now also published. Another review of ENSO has been prepared for a book chapter by Trenberth.

The TBO has been diagnosed from observations, as well as global coupled model results, by Jerry Meehl. The role of the northern spring season is highlighted in the sense of convective heating anomalies in the southeast Asian region contributing to subsequent south Asian monsoon strength in the TBO in agreement with other studies. Tropical Pacific SST anomalies undergo major transitions in the northern spring season, while the sense of the tropical forcing of the midlatitude circulation over Asia is similar to the winter season. These processes also act to set up the land-sea temperature contrast for the subsequent south Asian monsoon season. It is shown how the TBO itself arises both as a consequence of air-sea interactions in the tropical Indian and Pacific Oceans and tropical-midlatitude interactions. These result in a land-sea temperature contrast that exerts a strong control over monsoon strength. As noted in observations, in a global coupled general circulation model (GCM) and in an atmospheric GCM run with specified convective heating anomalies, south Asian snow cover can contribute to subsequent monsoon strength. However, it is most likely symptomatic of large-scale midlatitude circulation changes that affect land temperatures and thus land-sea temperature contrast.

Meehl continued to collaborate with Roger Lukas (University of Hawaii), George Kiladis (NOAA), Matthew Wheeler (University of Colorado), and Ian Matthews (NOAA) to examine time and space scale interactions that involve interannual global-scale processes (e.g., ENSO, TBO, etc.) that alter the base state of tropical SSTs and midlatitude circulation. This base state then affects the evolution and development of subseasonal phenomena (Madden-Julian Oscillation (MJO), submonthly), and these subsequently affect synoptic timescale (1- to 6-day) events where strong air-sea coupling takes place. These processes could lead to situation-dependent forecast skill on the submonthly timescale.

Madden, Hoar (GSP), and Ralph Milliff (Oceanography Section, OS) are working with satellite scatterometer surface winds to better understand that aspect of the 40-50 day tropical oscillation. In a related project, they are studying the frictional stresses and torques that drive the Antarctic Circumpolar Current using the scatterometer winds.

The World Climate Research Programme (WCRP) CLIVAR Program was asked to define El Niño. Following an exploratory article in the CLIVAR newsletter by Trenberth on this, a full article is now in press. A review is given of the meaning of the term "El Niño" and how it has changed in time, so there is no universal single definition, and precision can only be achieved if the particular definition is identified in each use. For quantitative purposes, possible definitions are explored that match the El Niños identified historically after 1950 and suggest that an El Niño can be said to occur if 5-month running means of SST anomalies in the Niño 3.4 region (5ºN to 5ºS, 170ºW to 120ºW) exceed 0.4ºC for 6 months or more. With this definition, El Niños occur 31% of the time and La Niñas (with an equivalent definition) occur 23% of the time. An advantage of such a definition is that it allows the beginning, end, duration, and magnitude of each event to be quantified. Most El Niños begin in the northern spring or perhaps summer and peak from November to January in SSTs.

The moisture budget in the atmosphere is documented in a book chapter by Trenberth. Extensive work on the NCEP reanalyses that resulted in a 300-page NCAR Tech Note by Trenberth and Chris Guillemot (Lamont-Doherty Earth Observatory) has been followed up with further work. Monthly, seasonal, and annual moisture-related fields, and various statistics and time series of precipitable water, evaporation E, precipitation P, E - P, the vertically integrated flux of moisture (as a vector), the divergence of the latter, the tendency in moisture, E – P, from the moisture budget, and the differences between the two estimates of E - P are detailed. The base period is 1979 to 1995. Similar statistics have been computed of precipitable water from NASA Water Vapor Project (NVAP) and P from Xie-Arkin (1996 dataset), along with differences, to document the mean bias, the standard deviations of the differences (indicating the typical errors), and the correlation between the anomalies. The moisture budget is shown not to balance and that the imbalances reveal problems in how moisture is assimilated in the analyses, as bull's eye features show up over many island stations.

Meehl, Hurrell, and van Loon have shown that the reduction of the semiannual oscillation at mid and high latitudes of the Southern Hemisphere that occurred around 1980 was related to a change in the seasonal cycle of surface temperatures near 50ºS and 65ºS. At both those latitudes, crucial for the mechanisms of the semi-annual oscillation (SAO) as first described by van Loon (1967), there was annual mean warming of surface temperatures after 1980, but the warming was not evenly distributed throughout the year. The changes in the seasonal cycles of surface temperature at those two latitudes resulted in an anomalous intensification of the surface temperature gradient in May and November. This acted to modulate the normal intensification of the gradient in March and September, thus reducing the amplitude of the SAO after 1980.

Hurrell and van Loon continued their work on North Atlantic climate variability. They demonstrated that the NAO has exhibited considerable variability at quasi-biennial and quasi-decadal periods over the past 130 years, and the latter have become especially pronounced the second half of this century. Since 1980, the NAO has tended to remain in one extreme phase and has accounted for a substantial part of the observed wintertime surface warming over Europe, dry conditions over southern Europe and the Mediterranean, and wetter-than-normal conditions over northern Europe and Scandinavia. Changes in the monthly mean flow over the Atlantic are accompanied by a northward shift in the stormtracks and associated synoptic eddy activity, and these changes help to reinforce and maintain the anomalous mean circulation in the upper troposphere.

Hurrell is pursuing the study of the mechanisms of decadal variability over the Atlantic and its relationships with ENSO and the global oceans, making use of several long CCM runs in which the model is forced with both observed and climatological SSTs. Emphasis is also being placed on an extensive documentation of the intraseasonal variability over the North Atlantic, and its interannual and decadal variability, making use of historical data and the new reanalysis datasets. Dickson (MAFF Fisheries Laboratory) and Hurrell are completing a study documenting the impact of the NAO on the Arctic Ocean.

Fillipo Giorgi (GDS), Hurrell, Maria Marinucci (GDS), and Martin Beniston (University of Fribourg) have shown that temperature changes over the Alps associated with the NAO show a strong elevational dependence, such that the higher the elevation the greater the warm anomaly during a high NAO index winter. Consistent with this finding are results from present day and doubled CO₂ experiments over the Alpine region with a nested regional climate model that show that there is a substantial elevational dependence in the simulated temperature change signal, mostly during winter and spring, with more pronounced warming at high elevations. Changes in other climate variables, such as precipitation, also show an elevation dependency, which may have important implications for impact assessments in mountainous regions and the early detection of anthropogenic signals.

Wigley and Rob Wilby have been developing statistical methods to "downscale" coarse spatial and temporal resolution GCM-derived precipitation data to the single-site and daily timescales. In addition, they have coordinated an intercomparison of different statistical downscaling methods using: standard stochastic simulation techniques (Dan Wilks, Cornell University); neural networks (Bruce Hewitson, University of Cape Town); and circulation-based methods (Phil Jones, University of East Anglia). The various methods have been applied to validate GCM-derived daily precipitation output for wet-day precipitation amounts, unconditional wet-day probabilities, wet-wet and dry-dry conditional probabilities, etc., and to assess how these statistics might change in the future.

Van Loon, Labitzke (FUB), and Leder (FUB) are continuing work defining the decadal period (ten to twelve year) oscillation in the stratosphere. They have completed a study of the annual cycle in total ozone and in the temperature of the ozone layer. Single value decomposition in 30 millibar height is being used to bring out the 10 to 12 year oscillation (65% of the variance in summer) and decadal trend (19% of the variance in summer) to identify the associated unique patterns in the troposphere and SST. NCEP data from 1968 to 1996 reveal that the pattern of correlations between stratospheric heights and temperatures and the solar flux are the same in both hemispheres, and that the highest correlations with the sun move from one summer hemisphere to the other.

Van Loon is studying several aspects of the decadal variations in the troposphere including, with Shea, a study of the changes in eddy fluxes associated with the global change in the late 1970s, and with Hurrell, further study of changes in the annual cycle in the Southern Hemisphere, going back as far as station records allow.

A commentary was given on the use and abuse of models by Trenberth, mainly in the context of climate change experiments. This elaborated on the capabilities of global climate models and noted that they must be carefully evaluated before they can be usefully employed as tools for climate change experiments. Trenberth and Wigleyn also participated in producing a UNEP/WMO booklet on common questions about climate change, and in particular, global warming.

Decadal variability of ENSO and relationships to climate change have been pursued by Trenberth and Hoar following on the earlier comprehensive statistical analysis of how an index of the Southern Oscillation changed from 1882 to 1995, with a focus on the unusual nature of the 1990-1995 ENSO warm event in the context of an observed trend for more El Niño and fewer La Niña events after the late 1970s. The conclusions have been challenged by two studies that deal with only the part of the results pertaining to the length of runs of anomalies of one sign in the Southern Oscillation Index (SOI). They, therefore, neglect the essence of Trenberth and Hoar, which focussed on the magnitude of anomalies for certain periods and showed that anomalies during both the post-1976 and 1990-mid-1995 periods were highly unlikely given the previous record. With updated data through mid-1997, additional tests were performed using a regression model with autoregressive-moving average (ARMA) errors that simultaneously estimates the appropriate ARMA model to fit the data and assesses the statistical significance of how unusual the two periods of interest are. The mean SOI for the post-1976 period is statistically different from the overall mean at < 0.05% and so is the 1990-mid-1995 period. The recent evolution of ENSO, with a major new El Niño event underway in 1997, reinforces the evidence that the tendency for more El Niño and fewer La Niña events since the late 1970s is highly unusual and very unlikely to be accounted for solely by natural variability.

A physically-based conceptual framework is put forward by Trenberth that explains why an increase in heavy precipitation events should be a primary manifestation of the climate change that accompanies increases in greenhouse gases in the atmosphere. Increased concentrations of greenhouse gases in the atmosphere increase downwelling infrared radiation, and this global heating at the surface not only acts to increase temperatures but also increases evaporation that enhances the atmospheric moisture content. Consequently, all weather systems, ranging from individual clouds and thunderstorms to extratropical cyclones, which feed on the available moisture through storm-scale moisture convergence, are likely to produce correspondingly enhanced precipitation rates. Increases in heavy rainfall at the expense of more moderate rainfall is the consequence, along with increased runoff and flooding. However, because of constraints in the surface energy budget, there are also implications for the frequency and efficiency of precipitation. It follows that increased attention should be given to trends in atmospheric moisture content, and datasets on precipitation rates and frequency need to be developed and analyzed, as well as total accumulation.

Other work on moisture and relationships of droughts, floods, and temperatures has been done by Aiguo Dai in collaboration with Trenberth and Tom Karl (NOAA). One study relates diurnal temperature range (DTR) to cloud cover, precipitable water, precipitation, and streamflow, and reveals physically consistent trends since the late 1940s, with a decrease in DTR associated with the enhanced hydrological cycle. Greenhouse warming of water vapor and clouds dominate in winter while cooling by precipitation and surface evaporation, and solar radiation reflection by clouds dominate the trend in summer for reduced DTR. A second study examines the changes in the Palmer drought severity index and relationships with streamflow and the Southern Oscillation. The latter is dominant globally and contributes to trends since 1950 towards more droughts in some areas (e.g., Australia, Africa) and wetter conditions in other areas (such as the central United States).

Wigley has contributed to a series of Technical Papers (TPs) produced by Working Group I of the IPCC: as a co-lead author of TP2 (on simple climate models) and TP3 (on the implications of the stabilization of greenhouse gas concentrations) and as convening lead author of TP4 (on the implications of greenhouse gas emissions limitation proposals). These papers were produced as key background documents for the Third Conference of Parties meeting held in Kyoto, Japan, December, 1997. He has also carried out additional work on the implications of CO₂ emissions limitations restricted to so-called "Annex I" (i.e., developed) countries, given a global goal to stabilize atmospheric CO₂ concentration.

Wigley’s work on detection has focussed on pattern correlation methods, comparing model-predicted and observed changes in horizontal (latitude-longitude) and vertical (latitude-height) patterns of temperature. Statistically significant results have been obtained demonstrating the existence of anthropogenic signals in the observational record due to the combined effects of carbon dioxide, aerosols, and stratospheric ozone depletion. This work has been influential in leading to the IPCC statement that the "balance of evidence suggests a discernible human influence on global climate." The work has been carried out in conjunction with Ben Santer (Lawrence Livermore National Laboratory, LLNL), Karl Taylor (LLNL), Joyce Penner (LLNL), Phil Jones (University of East Anglia), Peter Jaumann (Waterstone Consultants, Inc.), and others.

Further statistical analyses have been carried out to test the robustness of the pattern correlation detection method and to determine the expected behavior of the correlation test statistic (R(t)) under the influence of anthropogenic forcing, compared with its behavior under the null hypothesis of no external forcing. The method employs synthetic observational data made up of various prescribed anthropogenic "signals" combined with internally-generated natural variability "noise" from coupled ocean-atmosphere GCM control runs. The results show that expected R(t) magnitudes are similar to those observed. They also help to explain why, given the available observational data network, it has been so difficult to detect a greenhouse gas signal compared with a signal arising from both greenhouse gas and aerosol forcing. This work also shows that the R(t) statistic is not expected to show a simple linear increasing trend as the anthropogenic signal increases relative to the noise of natural variability. Instead, R(t) can show low-frequency excursions about an upward trend depending on the relative magnitudes of greenhouse gas and aerosol forcing. On theoretical grounds, R(t) should have decreased slightly in recent decades, as is observed.

Wigley has also carried out new time series studies of surface temperature at the global and hemispheric scales. He has demonstrated the consistency between the observed global-mean warming trend (which amounts to 0.6ºC over the past 100 years) and model estimates of this trend. The best match between model and observations occurs when solar forcing effects are included with anthropogenic (greenhouse gas and aerosol) effects. Solar forcing explains about one-third of the overall trend. The suggestion that the observed 20th century warming may only be a "recovery" from an early globally cool period (such as the so-called "Little Ice Age") has also been investigated. The work shows that, because of the finite memory of the climate system, any such recovery could, at most, explain only a very small fraction of the warming over the past century.

Steven Smith, working in collaboration with Wigley, has investigated the Global Warming Potential (GWP) concept. GWPs are the primary tool proposed for comparing the effects of reducing emissions of different greenhouse gases in terms of reductions in future climate change. The present analysis compares complete calculations of the effects of emissions reductions with results based on the use of GWPs as a comparison index. Under realistic scenarios, GWPs lead to major errors for analyses that extend beyond a few decades into the future, implying that further thought is required by policy makers before applying GWPs to implement a comprehensive multi-gas approach to reducing future climate change.