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

Ecology, Oct, 1998 by Janice L. Fuller

In summary, at Graham Lake several forest taxa appear to have increased in abundance following the hemlock decline, probably resulting in forest succession. By contrast, at High Lake, pollen percentages of T. canadensis did not reach levels as low as at Graham Lake, and there was a less dramatic response by other forest taxa. Pollen influx values of most taxa examined did not appear to increase exponentially at High Lake, unlike the situation at Graham Lake, indicating that there was less mortality of hemlock.

Recovery of hemlock populations

At both Graham Lake and High Lake, T. canadensis remained at lower abundances for [approximately]1500 yr before increasing to reach former abundances. A similar pattern is seen at other sites in southern Ontario (Mott and Farley-Gill 1978, Liu 1981, McAndrews 1981, Bennett 1987, Ritchie 1987). Why did T. canadensis take such a long time to recover? It seems unlikely that the populations of the pathogen remained at high levels after T. canadensis stands had been killed throughout its mid-Holocene range, although pathogens may have maintained T. canadensis populations at low levels for some time. Principal Components Analysis indicates that forest composition around Graham Lake, after T. canadensis populations recovered and re-expanded on the landscape, was different from that prior to the hemlock decline. F. grandifolia and A. saccharum were more abundant, whereas P. strobus, P. banksiana/resinosa, and Quercus were less common. This difference may be due in part to a shift in the climatic regime, associated with long-term changes suggested to have occurred during the Holocene. Several studies indicate that climate in northeastern North America has become cooler and moister in the latter part of the Holocene (Davis et al. 1980, Kutzbach 1987, Jackson 1989, Spear et al. 1994). However, after it had declined, T. canadensis reinvaded a forest type different from that which it invaded in the early postglacial period. T. canadensis populations re-expanded within a closed-canopy forest with high abundances of shade-tolerant species, such as F. grandifolia and A. saccharum. In addition, life history constraints would have affected rates of population expansion. T. canadensis is a slow-growing tree that does not reach reproductive maturity until at least 40 yr of age and has specific requirements for germination (Godman and Lancaster 1990).

By contrast, at High Lake, T. canadensis does not appear to have declined dramatically, and forest composition in the late Holocene was not markedly different from that prior to the hemlock decline. This suggests that where T. canadensis declined significantly, there was a long-term impact of the decline on forest composition. Whereas at High Lake, although climate and environmental conditions may have changed during the mid- to late Holocene, forest composition did not appear to change as markedly as it did at Graham Lake. Comparable data from other sites are required to explore the possible long-term effects of the hemlock decline on forest composition.