Extended Abstract (196K)P3.10
The relationship between low-frequency North Atlantic sea surface temperatures and eastern North American climate
Anthony Arguez, COAPS/Florida State University, Tallahassee, FL; and S. R. Smith and J. J. O'Brien
A time series of North Atlantic sea surface temperature (SST) anomalies is constructed by grid-averaging monthly anomalies from the Labrador Sea to Iceland from 1901-1999. The relationship between this North Atlantic SST Index (NASSTI) and temperature and precipitation anomalies over eastern North America is investigated. The NASSTI is low-pass filtered to isolate low-frequency fluctuations and separated into three parts: a warm phase (top 20%), a neutral phase, and a cold phase (bottom 20%). Average cold-neutral and warm-neutral deviations, as well as 3-month seasonal average deviations, are calculated for both temperature and precipitation. Precipitation anomalies during extreme NASSTI phases do not reveal any large-scale climate signals in eastern North America. While the deviations can exceed 50% for particular stations during certain seasons, there is no consistent temporal or spatial coordination between the anomalies. The NASSTI is, however, associated with coherent temperature anomaly patterns over eastern North America. The cold-neutral deviations reveal a dipole-like pattern in winter temperatures, with considerably colder conditions in Quebec and warmer temperatures in much of the United States. Warm phase temperature deviations reveal a general warming from late fall to early spring over much of eastern North America. The signal arrives in fall in the Ohio River Valley and the Great Lakes and propagates northeastward in time, amplifying as it reaches northern Quebec before subsiding in the spring. The NASSTI is compared to low-pass filtered time series of the Arctic Oscillation (AO) and the North Atlantic Oscillation (NAO). The AO and NAO indices are categorized in a fashion similar to the NASSTI, yielding a high phase, a neutral phase, and a low phase. The NAO and AO time series are negatively correlated with the NASSTI. This may be due to the dominance of the oceanic response to atmospheric forcing, as stronger (weaker) westerlies are associated with increased (decreased) evaporation, leading to cold (warm) SST anomalies. The high-neutral temperature signals of the AO and NAO are nearly indistinguishable from the NASSTI’s cold phase temperature deviations. The low-neutral AO/NAO pattern is characterized by warmer conditions in eastern Canada and cooler temperatures in the eastern United States. Thus, the low-neutral signal does not exactly correspond to the NASSTI’s warm phase temperature deviations. However, the two signals do correspond in eastern Canada, where the anomalies in both cases are large and positive during winter. It is hypothesized that the NASSTI, AO, and NAO temperature anomaly patterns over eastern Canada are interrelated by the variability in the strength of the Icelandic Low. An excited (relaxed) Icelandic Low is associated with the high (low) phases of the AO and NAO, negative (positive) SST anomalies in the NASSTI’s domain, and increased (decreased) cold air advection into eastern Canada.
Poster Session 3, Poster Session III
Thursday, 13 February 2003, 9:45 AM-11:00 AM
Previous paper Next paper