Genecology of jack pine in north central Ontario
Van Niejenhuis, Annette
Master of Science
DisciplineForestry and the Forest Environment
MetadataShow full item record
To understand the pattern of adaptive variation of jack pine in north central Ontario better, short-term provenance tests were established. Seed was collected from 64 sites to the east and west of Lake Nipigon and grown in three common garden tests, including a greenhouse trial at Lakehead University, a farm field trial at Lakehead University, and a field trial near Raith. Eight growth variables were measured (two annual heights from the greenhouse trial and three annual heights from each of the field trials), fourteen phenological variables were determined (elongation initiation and cessation dates, elongation duration and needle flush date at each trial; and foliage purpling at the greenhouse and Lakehead University field trials), and survival at the Raith trial was examined. Variation expressed among seed sources was significant for all growth variables and many phenological variables. Multiple regressions were run for 18 of the 23 variables against climatic variables interpolated using geographic information systems techniques from weather data of 56 weather stations, as well as spatial, soil, and vegetative variables which described the environment at seed origin resulting in coefficients of determination as high as 0.57. Principal components analysis (PCA) was used to summarize the variables examined, with 33 and 21 per cent of the variation accounted for by the first and second component respectively. Multiple regressions were run on the factor scores produced from PCA against the variables describing environment at seed origin. These regression models had coefficients of determination of 0.323 and 0.429 for the first and second factor scores respectively. The pattern of variation in this portion of the range as displayed in the mapping of the predicted factor scores was clinal with numerous irregularities. July and average annual temperatures, heating degree days, frost dates, and soil and vegetation variables were included in the predictive models. The contrast displayed in height performance between seedlings from the southwestern portion of the range and those from the north shore of Lake Superior reflects trends seen in a previous study of cone and needle characteristics.