Dendrochronology
In seasonal climates, trees preserve a continuous record of annual events, in particular, climate. Dendrochronology, the study of the annual growth in trees, is the only method of paleoenvironmental research that produces proxy data of consistently annual resolution. Trees add a cone of wood each year. Initially the cells are thin walled to conduct the abundant spring soil moisture. As soil water declines through the summer, the cells become thicker-walled and more dense. Thus each annual ring consists of early (light) and late (dark) wood. Tree-ring series can be classified as either complacent (uniform ring widths where moisture and heat are sufficient throughout the growing season) or sensitive (pronounced year to year variation in ring width, where conditions are frequently near the limits of the trees tolerance, e.g. tree lines, dry sites).
The search for proxy climatic data was the original application of tree rings. A.E. Douglass, the 'father' of dendrochronology was interested in the affect of sunspots on the earth’s climate. In 1901, he noticed ring-width variations on a cut log and reasoned that these were controlled by the tree's environment (Fritts, 1976). Douglass (1920) illustrated the relationship between climate and ring width by plotting both against time, and introduced the technique of cross dating by correlating ring-width signatures (sequences of wide and narrow rings) among trees distributed over large areas.
In western Canada, dendrochronology has been largely confined to the montane and boreal forests (Case and MacDonald, 1995; Luckman and Innes, 1991). An investigation of fire and insect infestation frequency in the jack pine forests of Manitoba (Gill, 1930) was the first Canadian study to use ring-width data and cross-dating techniques to develop a tree-ring chronology. Shortly afterwards, Powell (1932) compared variation in wheat yields in Saskatchewan to ring-width variation in white spruce and some hardwood species. Much of the tree ring research in western Canada has at the Laboratory of Tree Ring research in Tucson, Arizona, including the first studies of Douglas Fir in Alberta (Schulman, 1947), the first regional dendrochronological network for western North America (Drew, 1975), regional climatic reconstructions (Fritts, 1971; Fritts, et al., 1979; Fritts and Lough, 1985), and a dendrohydrological study of the Peace-Athabasca delta (Stockton and Fritts, 1973). One major exception is the research group at the University of Western Ontario which has established long tree-ring chronologies for the Canadian Rockies, to reconstruct climatic and glacial history of the last millennium and evaluate tree growth - climate relationships at altitudinal treeline (e.g. Luckman, 1995, 1996; Luckman and Colenutt, 1992).
The general absence of trees has obviously discouraged the pursuit of dendrochronology in the southern Interior Plains. Although a belt of aspen parkland extends across the prairie provinces, much of the original aspen polar was removed for crop production and this tree species is much inferior to coniferous trees for tree ring research (Fritts, 1976). However, conifers in the treed uplands and sheltered coulees tend to be climatically sensitive (Sauchyn and Beaudoin, 1999).
Methods
The cores and disks are processed using standard laboratory techniques (Stokes and Smiley, 1968). Cores are glued to a grooved pieces of wood such that the tracheids were approximately 30o from their original vertical orientation, ensuring maximum visibility of the latewood to earlywood transition between successive years. The wood is sanded with progressive finer paper to expose the growth rings for counting and measurement of ring width. .
The fundamental technique in dendrochronology is cross-dating, whereby distinctive series of narrow and wider tree rings are identified and matched among trees of different ages. Calendar years can then be assigned to rings from dead wood. This extends tree-ring chronologies beyond the life spans of living trees, and enables dating of pre-historic events (e.g. droughts) that affected tree growth. Cross dating among tree ring series of about the same length and age enables the detection of missing and false rings. Annual growth rings can be missing in unusually cold or dry growing seasons. False rings represent renewed growth after a cold or dry weather causes the formation of late wood mid way through a growing season. That is, two poorly defined rings correspond to one calendar year and thus one is considered false.
Sensitive tree-ring series contain the best signatures but can also have missing and false rings, and partial rings (growth truncated during growing season). Cross-dating of multiple ring-width series and chronological control (absolute dates) enable long master chronologies (e.g. 8200 years for the White Mountains of California where Bristlecone Pine lives for more than 4500 years) and records of episodic geomorphic and hydroclimatic events.
Dendrogeomorphology
The timing of recent geomorphic and hydrologic events can be established, where they have interrupted or disturbed the growth of coniferous trees (Shroder, 1980). Landslides generally result in tree mortality, especially along the lateral and terminal margins, and on the lower parts of landslides where the substrate tends to be severely disturbed. The year of mortality can be determined by cross-dating of ring width signatures from the dead and living trees. Some trees survive, although generally under poor growing conditions and tilted. On the upper valley sides, slump blocks and the associated soil and vegetation commonly remain intact, despite the considerable displacement of bedrock. The failure is deep below the surface, over a curved plane, causing the block to rotate and the trees to tilt backward towards the scarp face. Titled trees produce reaction wood and grow asymmetrically, so that the trunk curves upward to resume a vertical growth. Dating of landslides is thus "based on changes in the amount and eccentricity of tree growth" (Alestalo, 1971: 72). Dendrogeomorphology is a reliable means of determining the age of recent slope failures, where a single event such as the rotational failure of a slump block dominates the signal. Delayed growth response to disturbance, and errors in tree ring counting caused by missing or false rings, prevent exact results.
The disturbance of trees by floods produces similar growth responses, plus the scarring of cambium from the impact of boulders and logs carried in the flow. In general there is better agreement among the samples from flood affected trees than with landsliding, since they occupy level ground, where overbank stream discharge is the only geophysical disturbance. There are meteorological data, photographs and written documentation for the three most recent floods. Whereas the use of these methods is limited by the relatively short life spans of most tree species, evidence of earlier floods and landslides should exist in dead wood. Reconstructing geophysical events from dead wood is more involved, however, since the logs are not in situ and cross-dating is required to determine the years of anomalous growth.