No abstract

This analysis further refines Ropelewski & Halpert's (1987; Mon Wea Rev 115:1606-1626 analysis which investigates the relationship between El Niño-La Niña/Southern Oscillation events and southern United States precipitation. Comparisons are made between eigenvector-derived mid-tropospheric (500 mb) flow patterns over North America during extreme El Niño and La Niña months and a base climatology. In addition, the patterns are correlated to regional precipitation anomalies for the southern United States to determine mean surface responses. Cool season (November to March) months are divided into all winter months (AWM), positive anomaly months (PAM), and negative anomaly months (NAM). The extreme anomaly months were determined as any month with a Southern Oscillation Index (SOI) ± 1 standard deviation from the standardized mean. Therefore, the PAM and NAM anomalies represent the La Niña and El Niño extreme phases of the SOI, respectively. Results suggest that the positive (La Niña) SOI phase elicits a greater surface precipitation response than the El Niño phase. This is caused by substantial changes in the primary longwave flow during opposite SOI phases. During AWMs and NAMs, similar flow patterns, dominated by the Pacific/North American (PNA) teleconnection, prevail which induce similar regional precipitation responses. During PAMs, the mid-tropospheric flow shifts to a hybrid flow pattern which is between the PNA and the Tropical Northern Hemisphere teleconnections. Such displacement in the longwave flow variation centers ultimately affects jet stream flow and precipitation forcing, resulting in negative precipitation anomalies across the southern United States.

This analysis attempts to discern primary causes of interannual and interdecadal climate variations for precipitation and temperature regions of the conterminous United States. Varimax rotated principal components analysis of annual climate division data is used in the derivation of nine precipitation and five temperature regions. Each region's time series is examined for underlying linear trends, representing long-term climate change, and tests for variance changes, to determine regional climate variability shifts. The first six precipitation components, representing the entire eastern half of the country and the Northwest, displayed significant temporal increases. Of these, four displayed significant increases in interannual variability through time. For temperature, only the Southwestern region showed a significant change (increase) through time. However, significant reductions in temperature variability were confirmed for three regions. To determine the causes of the derived climate shifts, correlation analysis was performed with various atmospheric teleconnection indices. Precipitation trends are most strongly associated with variations in the Southern Oscillation Index (SOI) at the interannual time scale while interdecadal variations are associated more with variations in the Pacific/North American (PNA) teleconnection. Both interannual and interdecadal variations of regional temperature are most strongly related to the PNA, except for the Southwest, which showed a significant correlation to the SOI. This suggests that El Niño/Southern Oscillation (ENSO) events are the source for much of the precipitation change evident in the eastern and Northwestern United States and temperature change in the Southwest. [

No abstract