Ecological impact of the mid-Holocene hemlock decline in southern Ontario, Canada

Ecology, Oct, 1998 by Janice L. Fuller

Thuja/Juniperus appears to have increased in abundance after the initial increase by other forest taxa at Graham Lake, possibly replacing T. canadensis in the wetter locations. Thuja occidentalis is underrepresented in the pollen record, and as it is shade tolerant (Johnston 1990), may have become quite abundant following the hemlock decline.

Similar increases in abundance of several forest taxa have been recorded at other sites in the Northeast. Mirror Lake in New Hampshire records an increase in pollen influx of several taxa after the hemlock decline: Betula appears to have been one of the first pollen taxa to increase, followed by F. grandifolia, A. saccharum, and Ostrya/Carpinus (Davis 1978, 1981). Sediments from Nutt Lake, Ontario record a small increase in pollen influx of F. grandifolia and A. saccharum after the decline (Bennett 1987), whereas pollen percentages of Betula, Quercus, and Larix laricina increased at Found Lake, Ontario (McAndrews 1981). As Davis (1978) suggests, the patterns of forest dynamics after the decline vary regionally. This is probably a reflection of differences in environmental conditions, species composition, and/or variation in the nature of the decline. The taxa that most commonly appear to have increased after the hemlock decline include Betula, F. grandifolia, Acer, and Quercus (Davis 1978, 1981, McAndrews 1981, Mott and Farley-Gill 1981, Bennett 1987).

Pollen data thus indicate that the hemlock decline triggered an increase in abundances of several other forest taxa. The sharp increase in pollen influx of several tree taxa at Graham Lake, however, appears to have been short-lived. Before T. canadensis populations recovered, pollen influx values of most the taxa that had previously increased at Graham Lake, declined again [ILLUSTRATION FOR FIGURES 5 AND 9 OMITTED]. The decline was not synchronous for all taxa, and there is no evidence from the age-depth model for a marked increase in sediment accumulation rates at this time that might account for this trend. On the other hand, the pollen percentages of these taxa did not decrease markedly, indicating that the decrease in pollen influx may have been due to sedimentary changes. The death of numerous individuals of T. canadensis throughout the landscape must have provided a large number of potential sites for the establishment of other taxa. The creation of this newly available space may have been followed by a burst of regeneration and growth of other tree species. Although some species would have responded more quickly than others, a more or less even-aged stand may have developed. It seems plausible, therefore, that the decline in pollen influx values of several taxa (after their initial increase) observed at Graham Lake may have been due simply to the aging of these populations (Bormann and Likens 1979) and their subsequent thinning. Thinning is a natural phase of forest development, during which competition is intense, establishment of new seedlings is minimal, and stem density steadily decreases, as there is not sufficient space for all the offspring produced (Peet 1992). It is probable that the observed decline in pollen influx values at Graham Lake was the result of a combination of changes in sedimentation rates, competitive interactions among taxa, and thinning.