Methods for Analyzing Climatic Variations in the North Pacific Sector and Western North America for the Last Few Centuries
|Methods for Analyzing Climatic Variations in the North Pacific Sector and Western North America for the Last Few Centuries
|Year of Publication
|University of Wisconsin
The investigation of summer and winter climatic variations in the North Pacific sector and western North America during the last few centuries is the subject of this study. Tree ring widths from western North America are used as indicators of the past climate. It is shown that large scale spatial patterns of temperature and precipitation anomaly which are in turn related to large scale spatial patterns of pressure anomaly, i.e., to the general circulation. A spatial correlation method is chosen to identify and describe the major types of general circulation, as reflected in anomaly patterns of sea-level pressure, during the 20th Century. Five such anomaly type-patterns are identified for summer and four for winter. These are each associated with an assemblage of generalized weather patterns and a corresponding pattern of temperature and precipitation anomaly in the United States, as well as with a spatial anomaly pattern of tree ring widths from 49 sites over western North America. The occurrence of one of these ring width patterns for some year in the past is suggestive of the corresponding occurrence of the associated climatic anomaly type. Orthogonal eigenvector techniques are then selected for use in the development of a statistical model to estimate departure patterns of sea-level pressure using the ring width departures as predictor data. The model is first calibrated using available pressure data since 1899. The model is then applied to estimate winter pressure departure patterns since 1700 A.D. As a means of summarizing these climatic reconstructions, the estimated pressure departure pattern for each winter is compared with each of the type-patterns using correlation coefficients as a measure of comparison. The time series of correlation coefficients between a type-pattern and each winter’s estimate departure pattern provides an indicator of the occurrence, or non occurrence, of the corresponding anomaly type through time. Graphs of the time series of correlation coefficients corresponding to each of the four type-patterns are presented as indicators of reconstructed winter climatic variations for approximately the last two and one-half centuries. If an estimated pressure departure pattern is highly correlated with one of the type-patterns, the simultaneous occurrences of the temperature and precipitation anomalies associated with 20th Century occurrences of that pressure type-pattern are implicitly specific. These implicit estimates of temperature and precipitation anomaly are then independently verified using available data for the United States from the last half of the 19th Century. The climatic reconstructions are in good agreement with the recorded data and are found to complement and augment the findings of other investigators.