Аннотация:Time series representing two of the climate systems leading patterns of variability, namely El Niño–Southern Oscillation (ENSO) and the Arctic Oscillation (AO), are used together with 50 yr of daily mean surface air temperature data over the conterminous United States to diagnose relationships between winter temperature extremes and interannual climate variability. The aim is to focus attention on some of the physical phenomena that climate models must be able to simulate in order to be deemed credible for use in weather and climate forecasts and assessments. Since the 1950s there has been considerable decadal variability in winter surface air temperature extremes. At most locations in the United States the number of daily extremes is reduced during El Niño, and increased during La Niña and ENSO-neutral years. These changes are qualitatively consistent with a decrease in the daily mean surface air temperature variance during El Niño relative to La Niña and ENSO neutral. Changes in the number of warm extremes during a particular AO phase are largely compensated for by changes in the number of cold extremes, so that the net change in the numbers of surface air temperature extremes is close to zero. However, the AO is associated with larger changes in mean temperature than ENSO.
