The primary goal of this research is to gain a better understanding of the spatial and temporal relationships between atmospheric circulation features and winter climate variability in the southwestern United States, and to investigate the variations in these relationships over the past three centuries. A set of six circulation indices is compiled that describes circulation features important to winter climate variability in this region. This set includes pre-existing indices such as the SOI and a modified PNA index, as well as regionally-tailored indices. A network of 88 tree-ring chronologies is then used to reconstruct the indices and the regional winter climate variables: numbers of rainy days (a variable not previously reconstructed with tree rings) and mean maximum temperature. Analyses suggest that three types of circulation features have influenced winter climate in the Southwest over the past three centuries. Although ENSO-related circulation patterns have been an important factor, especially in the 20th century, circulation patterns featuring a southwestern low appear to be as important if not more important to climate in some time periods. Results suggest that low frequency variations in atmospheric circulation patterns have occurred over the past three centuries and have had spatially and temporally varying impacts on winter climate in the Southwest.

Thesis (Ph. D.-- Geosciences) University of Arizona, 1996. / Includes bibliographical references (leaves 175-187).

Annual tree-ring width of Southwestern conifers growing on dry sites exhibits sensitivity to variation in climatically created moisture stress. Douglas-fir, ponderosa pine, and pinyon in the eastern San Juan Basin in northwestern New Mexico and southwestern Colorado were sampled at four sites to investigate covariation of climate with intra-annual anatomy. The sites possessed characteristics that created different amounts of physiological stress in trees. Increment borer samples were glued into wooden mounts and machined to approximately 1.0 mm thickness by a special router-planer. All samples were crossdated by comparing climatically controlled synchronous patterns of ring widths. Moving slit X-ray densitometry (at Forintek Canada Corporation Western Forest Products Laboratory, Vancouver, British Columbia) objectively defined the earlywood zone (large, low density cells) and latewood zone (smaller, denser cells formed late in the growing season) in each ring. The densitometer measured eight parameters for each ring: ring, earlywood, and latewood width, minimum and maximum density, and mean ring, earlywood, and latewood density. Individual radial series were standardized (i.e, transformed to indices with 1.0 mean and homogeneous variance) by fitting curves and dividing annual values by the corresponding curve values. Density series proved more difficult to standardize than widths and usually correlated more poorly among individual radii of the same data data type. Statistical characteristics of site summary density chronologies differed from width chronologies. Response functions using monthly mean temperature and total precipitation showed climate influenced all data types. Low moisture stress increased ring, earlywood, and latewood width and ring, maximum, and latewood density. High moisture stress increased minimum and early— wood density. No width or density type consistently covaried more than any other with climate. Linkage of climatic variation with density parameters differed considerably from that reported in the literature for conifers growing in wetter, cooler climates. Southwestern conifers posed unique densitometric technical difficulties. Selection of sites that caused moderate physiological stress and samples with few missing rings proved critical. Acquisition of density data required much more time and effort than optical measurement of ring width, but yielded valuable intra—annual data. Intra—annual densitometric data hold great promise for reconstruction of seasonal paleoclimate.