The Wetherill Mesa Archeological Project devoted a portion of its funds and personnel to an attempt to reconstruct the past climate at Mesa Verde National Park. Data for this study were obtained from the thorough tree-ring dating of the archaeological excavations, from environmental measurement stations, and from measurements of tree growth. The results of the dendrochronological study on Wetherill Mesa include very long tree-ring chronologies; large clusters of dates from each site excavated on Wetherill Mesa; and a tentative climate reconstruction for Mesa Verde.

A long Douglas-fir tree-ring chronology was needed for use in a climatic interpretation of Mesa Verde. The archaeological portion of the chronology was derived from 21 archaeological specimens obtained from excavations on Wetherill Mesa and extended from A.D. 435 to 1276. The modern portion, extending to 1963, was derived from directional and non-directional cores from six very old living Douglas-fir trees. In addition, four directional cores from each of the six trees were used to compute tree-ring indices for the years 1860 to 1963, which have been used for an analysis of paleoclimate.

Contains: References, Indices for Four Directional Cores from SOT-1 to -6, Douglas-fir Indices for Climatic Analysis, Composite Species Indices for all Archaeological Specimens from Wetherill Mesa, Juniper Indices for Wetherill Mesa Archaeological Sites (A.D. 301-1274), Pinyon Pine Indices for Wetherill Mesa Archaeological Sites (A.D. 161-355; 486-1278), Douglas-fir Indices for Wetherill Mesa Archaeological Sites (A.D. 435-1276), Ponderosa Pine Indices for Wetherill Mesa Archaelogical Sites (A.D. 1060-1269) / The 1,916 wood and charcoal specimens obtained from five years of survey and excavations on Wetherill Mesa produced 501 tree-ring dates. The dated specimens filled a gap between A.D. 674 and 733, greatly lengthening the combined species' tree-ring series for Mesa Verde so that it now extends from A.D. 161 to 1280. In addition a very long tree-ring series was established for juniper, a tree rarely dated from Mesa Verde in the past.

Dendrochronological research in the Mexican state of Oaxaca in 1970 proved negative due to complacent ring series. It is suggested that this is caused by a flexible growing season triggered only by the onset of the rains. Pine and fir were sampled from eleven sites. No old age trees were located and crossdating could not be accomplished.

A method is presented for making probability statements about past climatic conditions for the state of Arizona given the corresponding relative width of tree rings. The probability statements about periods of extreme climate from 1650 to 1899 are based upon the joint occurrence of the state-wide average seasonal climate and ring widths during 1899-1957. The ring-width values used are index chronologies selected from four different areas within Arizona. Spatial homogeneity among the four chronologies is evaluated by using digital filtering and correlation techniques. The chronologies are then normalized, averaged to form a state-wide series, and the values of state-wide growth for each year placed into one of nine equally probable classes. Similarly, seasonal temperatures and seasonal precipitation are placed into three equally probable classes and the joint occurrences between temperature and precipitation become nine climatic classes. Contingency tables are used to establish the joint occurrence of the nine climatic classes and the nine ring-width classes. A number of 10 and 20 year intervals since 1650 are identified as periods of unusually high or low probability of occurrence of above or below normal precipitation for any season of the year. The period 1880-1889 is of special interest as it was a period when downcutting was initiated in many Arizona streams and is also one of the periods in which the probability for below normal seasonal precipitation was unusually high (p = 0.48).

In the last ten years tree-ring analysis in Europe and its application has been extended considerably. By varied methods, a number of chronologies have been established for conifers and deciduous trees from different climatic regions that partly reach back to periods before the birth of Christ. The focus of dendrochronological research is usually dating and climatological studies are carried on only sporadically.

A crossdating program for tree-ring research has been written to compare ring patterns of individual trees and composites. The program written in FORTRAN calculates the t value for correlation at every point of overlap of the two chronologies. The program is small enough to be used on a routine basis with a large number of trees. As the chronologies must be free from errors, the program is more suited to the study of oaks than coniferous trees.

Bristlecone pines were sampled at four sites ranging from the arid lower forest border to the upper treeline in the Snake Range of eastern Nevada. Maximum ring-width response to environmental variation is found at the upper and lower forest limits. Ring-width index series from individual trees, as well as the mean site chronologies, were compared by cross-correlation analysis and principal component analysis, combined with digital filtering to emphasize variations in different frequency ranges. Positive correlation exists between the high-frequency variations at all sites, but the longer term trends and fluctuations at the upper treeline are negatively correlated with fluctuations at the lower altitude sites. Cross-spectral analysis substantiates the results of cross-correlation analysis and indicates that the associated variations in the upper treeline and lower forest border chronologies are concentrated in certain frequency ranges that may have climatic significance. From examination of the climatic response functions, the negatively correlated low frequency variations are tentatively judged to be related to warm- season temperature fluctuations, whereas the positively correlated high frequency variations may be related to precipitation. Frequency-dependent relationships between tree-ring chronologies, or between tree-rings and climate should be considered in the analysis of large arrays of tree-ring

Ring-width variation in trees at upper treeline in the high mountains of temperate latitudes is a potentially important indicator of past climatic variations, especially temperature variations. Bristlecone pines (Pinus longaeva D.K. Bailey and P. aristata Engelm.) were sampled at nine sites in western United States. Plotted annual ring-width indices are given for chronologies that range in length from 532 years in New Mexico, 1409 years in Colorado, and 1239 years in Nevada to 1501 years in eastern California. Possibilities for increasing the length of these chronologies by incorporating tree-ring data from logs and remnants are good in several of the areas, and a 5405-year upper treeline chronology has been developed in California. Tree-ring statistics show that crossdating is poorer, the climatic response is smaller, and the autocorrelation (a measure of year-to-year persistence) is greater in trees at upper treeline sites than at sites near the arid lower forest border. Climatic response functions differ in many details, but generally indicate a positive response of ring growth to warm temperatures in the previous late summer and autumn and current spring and summer. There is a negative response to warm temperatures during some winter and early spring months at several of the sites. The effect of precipitation varies greatly, but a positive response to precipitation during the previous summer or autumn, and during the current spring or summer is indicated. Variations in needle length are related to summer temperature, and may be important in explaining the high autocorrelation of upper treeline ring-width series. Ring-width departures from the long-term mean during the past 500 years were calculated from upper treeline data for 30-year subperiods. The departures are in the same direction over the whole region during many of these subperiods, indicating that climate, rather than local ecological factors, is responsible for the ring-width variations. Comparison of tree-growth fluctuations with meteorological observations at selected stations shows that a general warming trend between the periods 1901-1930 and 1931-1960 is reflected by an upward trend in tree growth. However, low rates of tree growth during an earlier warm period (1850-1869) may be due to a lag in the response of ring -width growth to climatic changes at upper treeline.

A widely spaced grid of tree-ring chronologies most suitable for dendroclimatic analysis of western North America is selected objectively on the basis of 1) numbers in the sample, length in years, and site locations, 2) statistical characteristics of the chronologies, and 3) correlation of chronologies with those on neighboring sites. The chronology statistics are then analyzed to characterize the quality of the selected set. The procedures used in this study are recommended for future climatic analysis to assure objectivity in the selection of quality tree-ring data and to allow comparisons of the statistics for new chronologies to the established data sets.