Comparing composition and structure in old-growth and harvested (selection and diameter-limit cuts) northern hardwood stands in Quebec
Introduction
One of the approaches proposed for implementing ecosystem management is to learn from the study of natural disturbance regimes and to use that knowledge for managing forests in order to minimize the ecological “distance” between managed and natural systems. It is thought that the probability of conserving the majority of the attributes, processes and species characteristic of natural ecosystems would be enhanced by the development and implementation of a natural-disturbance-based approach to forest management (Attiwill, 1994).
If compared to natural disturbance regimes, disturbances associated with forest management differ in their rate of occurrence, extent, severity (Bergeron et al., 2002) and synchronicity. In natural systems, disturbances are known to play a major role in determining patterns of recruitment, growth and mortality, as well as associated structural patterns observed in forests (e.g. Bormann and Likens, 1979, Runkle, 1984, Runkle, 1985a, Runkle, 1985b).
In the hardwood forests of northeastern North America, the natural disturbance regime is mainly characterized by the regular death and fall of branches, and single canopy trees or small groups of trees that create gradual or sudden openings in the canopy, (Bormann and Likens, 1979, Runkle, 1985a). This disturbance regime is known as the micro-gap regime. Runkle, 1982, Runkle, 1985a reported that the average rate of gap creation in old-growth mesic forests of the eastern United States was about 1% of total land area per year, ranging from 0.5 to 2%/year.
In the hardwood forests of Quebec, single-tree selection cutting has been the most widely used harvesting system since the early 1990s (Bédard and Majcen, 2003). It is believed that this silvicultural system is similar in some ways to the natural micro-gap regime (Majcen, 1994, Jones and Thomas, 2004) since it creates single- and multiple-tree canopy openings affecting approximately 30% of the initial basal area, with a return interval of approximately 25 years, i.e. a mean rate of harvest equivalent to the expected periodic net increment.
Many recent studies have assessed the impact of partial harvesting in hardwood stands, by focusing on particular features such as: snag and coarse woody debris (CWD) abundance (Gore and Patterson, 1986, Goodburn and Lorimer, 1998, Doyon et al., 1999, Hale et al., 1999, McGee et al., 1999, Fraver et al., 2002); tree size and structure (Hale et al., 1999, McGee et al., 1999, Solomon and Grove, 1999); gap characteristics (Kimball et al., 1995, Beaudet and Messier, 2002), light profiles (Beaudet et al., 2004); tree species composition (Majcen, 1995, Hale et al., 1999, Solomon and Grove, 1999); tree growth (Majcen, 1995, Bédard and Majcen, 2003, Jones and Thomas, 2004); the biodiversity of various groups of organisms (Bourque and Villard, 2001, DeBellis et al., 2002, Moore et al., 2004). In this study, a more global approach was used to characterize and compare many structural and compositional features of old-growth hardwood stands (OG) modeled under the micro-gap regime, and two types of partially cut stands. Since selection cutting has only relatively recently been widely implemented in Quebec, diameter-limit cuts (DLC), a form of partial cutting that has been widely used in the past (Bédard and Majcen, 2003), were also included in our comparison. This allowed us to assess the forest characteristics most susceptible to recover in the mid-term, and the compositional and structural legacies from that former type of harvesting.
Compared to the natural micro-gap regime, partial harvesting systems imply, among others: (1) a temporal synchronization of canopy opening and (2) a removal of large quantities of wood. The regeneration process driving the future stand composition and dynamics is influenced by canopy openings (Runkle, 1984). The rate of canopy gap formation and gap characteristics (size and spatial distributions) will thus influence the structure and successional path of the future stand (Runkle, 1985b). The micro-gap regime is a type of small-scale disturbance that induces local changes in resource availability, including light conditions, thereby promoting the recruitment and growth of regeneration in a particular location, at a particular moment. At the stand scale, this process of gap formation and regeneration response increases the structural heterogeneity through the formation of a mosaic of patches of vegetation at various ages and growth phase (Runkle, 1984). The harvesting-intensity-to-return-interval ratio of the current selection cutting regime in Quebec is similar to the natural gap formation rate (about 1% area/year). However, harvesting simultaneously affects large expanses of forests and therefore results in the synchronized formation of numerous canopy openings. Such a phenomenon does not correspond to the spatio-temporal pattern of gap formation under the natural micro-gap regime. It should be noted, however, that some natural disturbances, such as heavy ice storms, moderate blowdowns or low-severity fires might possibly impact the forest canopy and influence the regeneration dynamics similar to partial cutting. These more intense disturbance events occur in northeastern hardwood forests and play an important role in their natural dynamics (Lorimer and Frelich, 1994), but are far less frequent and/or affect relatively smaller fractions of the landscape compared to the micro-gap regime. The focus of this study were forests shaped by the micro-gap regime in their recent history.
Large and defective living trees as well as dead wood, whether standing (snags) or downed (stumps, coarse woody debris (CWD)), have been recognized as key structural features for numerous species in all forest ecosystems (Harmon et al., 1986, Hunter, 1990). Their roles are almost as diverse as their users; in the North American Northeast, more than 25% of forest wildlife species are associated with or dependent on some of these structural elements (DeGraaf et al., 1992). In addition to being involved in carbon budgets and nutrient cycling, dead wood provides nesting, denning, perching, sheltering, breeding and foraging sites for many vertebrates and invertebrates, as well as germination and growth substrate for many fungi, bryophytes and vascular plants (Harmon et al., 1986).
The main objective of this study was to compare OG and harvested northern hardwood stands to determine to what extent managed stands differ from unmanaged stands in terms of tree species composition and structural characteristics. More specifically, it was hypothesized that, compared to OG, managed stands would have a greater fraction of their area under gaps, a higher proportion of shade-intolerant and mid-tolerant species, and a more homogeneous horizontal distribution of foliage cover associated with a higher density of saplings. Also, because logging involves the removal of live trees that would potentially become large and/or defective trees, and eventually snags and CWD, it was also hypothesized that these structural elements would be less abundant in harvested stands.
Section snippets
Study area and site selection
Sampling was carried out from June to August 2002 in the Outaouais region, in southwestern Quebec, Canada (45°43′–46°06′N; 75°00′−75°51′W) (Fig. 1). The region is mantled by thin glacial till deposits and bedrock outcrops are common. Relief ranges from 350 to 400 m above sea level (Robitaille and Saucier, 1998). Mean annual temperature ranges from 2.5 to 5 °C, and mean annual precipitation is about 1000 mm, of which 25% falls as snow (Robitaille and Saucier, 1998). Stands were dominated by sugar
Gaps
DLC had the lowest percentage of land area under gaps (4.4 ± 1.3%), followed by OG (9.4 ± 1.7%) and SC (32.0 ± 3.0%), the latter with an average percentage of area under gaps more than three times higher than in OG. All stand types were significantly different from each other (Table 3).
Gap size was only evaluated in OG and DLC, as explained previously. Gaps in OG were 36.5 ± 6.0 m2 in size on average and their size generally followed an inverse J-shaped distribution characterized by a high proportion of
Gaps
Comparing the fraction of land area under gaps and gap characteristics observed in this study with values reported in the literature is difficult since many different gap definitions and methodologies are used (Runkle, 1992). The results in this study in OG are, however, in general agreement with other studies (Table 2). The inverse J-shaped distribution of gap size was also observed by other authors (Runkle, 1982, Runkle, 1985a, Payette et al., 1990, Krasny and Whitmore, 1992).
Because of the
Acknowledgments
We are indebted to S. Côté and P. Côté for assistance with fieldwork and data entry. We also thank N. Villeneuve for helpful comments on the proposal and for providing access and information about old-growth stands, and P. Cyr and G. Forest from the MNR who provided helpful information about history and location of harvested sites. C. Lorimer, F. Doyon, J.-M. Roberge as well as three anonymous reviewers provided helpful comments on the manuscript. Funding for this project was provided by the
References (70)
The disturbance of forest ecosystems: the ecological basis for conservative management
For. Ecol. Manage.
(1994)- et al.
Variation in canopy openness and light transmission following selection cutting in northern hardwood stands: an assessment based on hemispherical photographs
Agric. For. Meteorol.
(2002) - et al.
Ecology of coarse woody debris in temperate ecosystems
Adv. Ecol. Res.
(1986) - et al.
Patterns of early cohort development following shelterwood cutting in three Adirondack northern hardwood stands
For. Ecol. Manage.
(1999) Stand development and overstory-understory interactions in an aspen-northern hardwoods stand
For. Ecol. Manage.
(1992)Disturbance regimes in temperate forests
- et al.
Effects of uneven-age management intensity on structural diversity in two major forest types in New England
For. Ecol. Manage.
(1999) - et al.
Woody understory response to changes in overstory density: thinning in the Allegheny hardwoods
For. Ecol. Manage.
(1998) - et al.
Understory light profiles in northern hardwood stands: recovery process following selection cutting
J. Ecol.
(2004) - Bédard, S., Brassard, F., 2002. Les effets réels des coupes de jardinage dans les forêts publiques du Québec en 1995 et...