The North Atlantic Oscillation : El Nino's Young Cousin Prof. Martin Beniston Director Department of Geography University of Fribourg Switzerland The North Atlantic Oscillation (NAO) is one of the large-scale modes of variability in the climate system. As its name indicates, the NAO is centred on the North Atlantic Ocean basin. Here the atmospheric circulation normally displays a strong meridional (north-south) pressure contrast, with low pressure in the northern edge of the basin, centred close to Iceland, and high pressure in the subtropics, centred near the Azores. This pressure contrast drives the mean surface winds and the wintertime mid-latitude storms from west to east across the North Atlantic, bringing warm moist air to the European continent. It has long been observed that the monthly and seasonal (particularly wintertime) averaged sea level pressure in stations in Iceland and the Azores, display an out-of-phase relationship with one another. More precisely, there is a tendency for sea level pressure to be lower than normal in the Icelandic low pressure centre when it is higher than normal near the Azores and vice versa. This fluctuation is referred to as the NAO. It is related to noticeable changes in monthly and seasonal averaged wind speed and direction over the ocean, and concomitant changes in the paths of wintertime storms and their effect over the ocean and Europe. The NAO is the dominant mode of atmospheric variability in the North Atlantic sector throughout the year, but it is most pronounced during the winter season. The NAO exerts a dominant influence on wintertime temperature and precipitation across the North Atlantic basin and thus has major impacts on marine and terrestrial ecosystems. Changes in these climatic factors are of serious consequence to a wide range of human activities, as has been evident from recent scientific and media reports in Western European countries. Linear regression analysis shows that a considerable portion of the climatic fluctuations in surface temperatures and sea surface temperatures is directly related to the NAO index. Changes of more than 1( C associated with a one standard deviation change in the NAO index occur over the northwest Atlantic and extend from northern Europe across much of Eurasia. Changes in temperatures over northern Africa and the southeast U.S. are also notable. The changes in the mean circulation patterns over the North Atlantic are accompanied by pronounced shifts in the storm tracks and associated synoptic eddy activity which affect the transport and convergence of atmospheric moisture and can, therefore, be directly tied to changes in regional precipitation. Hurrell (1995) has shown that drier conditions during high NAO index winters occur over much of central and southern Europe and the Mediterranean, while wetter-than-normal conditions occur from Iceland through Scandinavia. This has been the case for much of the past two decades. In Portugal and Spain, for instance, severe drought conditions throughout Spain has affected olive harvests. In contrast, increases in wintertime precipitation over Scandinavia may be related to recent positive mass balances in the maritime glaciers of southwest Norway, one of the few regions of the globe where glaciers are not retreating. For Swiss precipitation no correlation to the NAO can be found (Widmann et al., 1998). Beniston(1997) showed that snow depth and duration in Switzerland is correlated with the NAO. Beniston and Rebetez (1996) found that snow depth and duration over the past several winters have been among the lowest recorded this century, causing economic hardships on those industries dependent on winter snowfall. However, as 1996 was a low NAO-index winter, Europe experienced a severe winter with record low temperatures and heavy snowfalls in many parts of southern Europe. Other studies which decrypt the influence of the NAO on climate parameters have been carried out by Hurrell (1995), who examined correlations between the NAO-index and climate parameters such as temperature and precipitation, and concluded that the behaviour of the NAO explains up to 34% of interannual variability in the extra tropical northern hemisphere. Beniston et al. (1994) conducted an exhaustive analysis of the relationship between climate characteristics in Switzerland and the NAO index, while Beniston (1997) investigated variations of snow depth and duration in the Swiss Alps. The NAO seems not to be a stationary stochastic (or deterministic) process in the time scales that are common in climate research. Appenzeller et al. (1998) showed by means of wavelet analysis, that in a 1400-year simulation of the ECHAM3 General Circulation Model (GCM) developed at the Max-Planck-Institute in Hamburg, as well as in ice-core data, the dominant frequencies of the NAO-index changes in time. One frequency in the NAO-index of ECHAM3 could be attributed to a coupled ocean-atmosphere mode which projects into the NAO-index (Timmermann, 1998). Another indication that the NAO can change its regime is the strong positive trend of the index since the late 1960s. During this latter part of the record, an 8-year oscillation may be observed This trend in the index may come from a very significant mechanism in a changing climate which GCMs must reproduce if climate projections in the North Atlantic region are to have any meaning. (see also section 2.2.1.3) Beniston et al. (1994) showed that there were significant links between NAO behavior and charateristics of Swiss climate. These links are particularly strong at times when the NAO index is either strongly positive or strongly negative. Differences between high and low index periods (essentially the 1980s and the 1950s, respectively) emphasize the high positive pressure, temperature, and sunshine duration anomalies mostly from late autumn to early spring, and corresponding negative precipitation anomalies. The precipitation anomaly does not appear if the NAO-index is linearly correlated with Swiss precipitation (Widmann, 1998). In Beniston (1997), it is shown, that large-scale forcing, and not local or regional factors, plays a dominant role in controlling the timing and amount of snow in the Alps, as evidenced by the abundance or dearth of snow over several consecutive years. Furthermore, since the mid-1980s, the length of the snow season and the snow amount have substantially decreased, as a result of pressure fields over the Alps which have been far higher and more persistent than at any other time this century. These pressure fields are well correlated with the particularly high NAO indices during this period. A detailed analysis of a number of additional Alpine stations for the last 15 years shows that the sensitivity of the snow-pack to climatic fluctuations diminishes above 1750 m. Widmann et al. (1998a) found a low pressure field in the north sea that is well correlated with high precipitation over Switzerland in this century. This pattern is able to reconstruct much of the observed Swiss precipitation trend since 1960. The reconstruction fails before 1960. They also assessed the relation between this coupled pattern and the NAO. As expected, they find a correlation between the NAO and the pressure pattern, whereas a correlation between the NAO and Swiss precipitation is almost absent. Nevertheless they assume a relation between the NAO-trend and Swiss precipitation trend that has to be examined. Widmann et al.(1998b) examined the same pattern in NCEP-reanalysis and found good agreement with the coupled pattern of Widmann et al. (1998a). However, a trend in Swiss precipitation is absent in the NCEP reanalysis. Bresch (1998) examined in a two-part study the response of storm track activity and NAO to caracteristic SST-anomalies in the Atlantic in reanalysis data. He found that a positive SST-anomaly produces an NAO-like pattern at the 500 hpa-field. Changes in SST could explain over 30% of the NAO month-to-month variability. He also found a narrowing of the Atlantic storm track width due to positive SST-anomalies. These results were compared with sensitivity studies that he carried out with a RCM. He found a good qualitative agreement between observations and the numerical experiments. The influence of these findings on Swiss climate have not been investigated to date. Coupled General Circulation Model (GCM) simulations usually represent the variability of the NAO quite well. (Osborn et al., 1998; Tschuck,1997; Ulbrich and Christoph, 1998); however, the NAO trend since 1960 has not been convincingly captured in any GCM-run. Osborn et al. (1998) did not find a trend in a 240-year scenario-run with the British HadCM2 GCM. In comparable runs with the ECHAM4-GCM model at T42 spectral (2.8° latitude/longitude) resolution coupled with OPYC3-ocean Christoph and Ulbrich found only a weak trend in the NAO and a eastward-shift of the pressure field. Paeth finds in the same run a realistic trend at first sight, but he used a different definition of the NAO-index which includes the location of the NAO-pattern. In an uncoupled model, Tschuck found no NAO-trend at all in his control-run for this century. He used the ECHAM4 model at T106 spectral (1.1° latitude/longitude) resolution. Sea surface data were provided by the GISST-dataset. Because of these uncertainties related to the capability of GCMs in reproducing NAO events under current climatic conditions, an accurate representation of the NAO in a future warming climate seems to be very difficult. References Appenzeller, C., Stocker, T. F. and Anklin, M. 1998: North Atlantic Oscillation Dynamics Recorded in Greenland Ice Cores, Science, conditional accepted. Beniston, M., 1997: Variations of snow depth and duration in the Swiss Alps over the last 50 years: links to changes in large-scale climatic forcings. Climatic Change, 36, 281-300 Beniston, M. and M. Rebetez, 1996: Regional Behavior of Minimum Temperatures in Switzerland for the Period 1979--1993. Theor. Appl. Climatol., 53, 231-243 Beniston, M., Fox, D. G., Adhikary, S., Andressen, R., Guisan, A., Holten, J., Innes, J., Maitima, J., Price, M., and Tessier, L., 1996: The Impacts of Climate Change on Mountain Regions. Second Assessment Report of the Intergovernmental Panel on Climate Change (IPCC), Chapter 5, Cambridge University Press, pp. 191 - 213 Bresch, D, 1998: Coupled Flow and SST Patterns of the North Atlantic, Diss. ETH 12878 Franke, 1998: http://www.dkrz.de/clivar/vol2/pd1_new.html#D532 Hurrell, J.W.,1995: Decadal Trends in the North Atlantic Oscillation: Regional Temperatures and Precipitation, Science, 269, 676-679, . Osborn, T., Briffa, K.R., Tett S.;Jones, P.D. and Tringo, R., 1998:Evaluation of the North Atlantic Oscillation as simulated by a climate model. Timmermann, A., M. Latif., R. Voss, and A. Grötzner, 1998: North Atlantic variability: a coupled air-sea mode. J. Climate, in press. Tschuck, P., 1997: Atmospheric Blocking in a General Circulation Model, Zürcher Geographische Schriften, Heft 70, Geogr. Institut ETH Zurich, 125 pp. Ulbrich, U.; Christoph M. 1998:A shift of the NAO and Increasing Storm Track Activity over Europe due to Anthropogenic Greenhhouse Gas Forcing, submitted to Climate Dynamics Widmann, M. and C. Schaer, 1998a: Relations between Swiss wintertime precipitation and large-scale Pressure and temperature changes on monthly to decadal time scales In preparation. Widmann, M. and C.S. Bretherton, 1998b: "Validation of mesoscale precipitation in the NCEP reanalysis using a new gridpoint dataset for the Northwestern US". submitted to J. Climate.