Long-term (1956-1996) effects of clearcut logging and scarification on forest structure and biota in spruce, mixedwood, and pine communities of west-central Alberta.
John G. Stelfox1, J. Brad Stelfox2, Wayne C. Bessie3,, and Calvin R. Clark4
1C116, RR7, Desert
Cove Estates, Vernon, British Columbia, VIT 7Z3
2Forem Technologies, Box 805, Bragg Creek, Alberta, T0L 0K0
3Foothills Model Forest, Box 6330, Hinton, Alberta, T7V 1X6
currently, Golder and Associates Ltd., 940 - 6 Ave. SW., Calgary. AB. T2P 3T1
4Clark Ecodynamics, General Delivery, Bragg Creek, Alberta, TOL OKO
Full Report in PDF Format (1.9 MB)
John Stelfox was born in Rocky Mountain House, completed his undergraduate degree at the University of Alberta, his M.Sc. at Utah State University, and his Ph.D at the University of Montana. He worked as a wildlife biologist for the Alberta Fish and Wildlife Division from 1955 to 1966 and as a research scientist for the Canadian Wildlife Service from 1966 To 1986. He now lives in Vernon, British Columbia where he works as a forest consultant.
Special recognition is given to E.S. Huestis (former Deputy Minister, Alberta Dept. Lands and Forests), J. Clark (former Woodlands Manager of Champion Forest Products), and the late D. Crossley (former Chief Forester of St. Regis Ltd.) for their assistance in establishing and fostering this research project.
The support, guidance, and cooperative efforts of Weldwood of Canada Ltd., Alberta Natural Resources Service, Hinton Environmental Training Centre, and Canadian Wildlife Service have been critical to the completion of this project. Throughout the four decades of this study, the support and encouragement offered by K. Smith, G. Lynch, R. Udell, R. Bonar, D. Quintilio, and G. Wilde are gratefully acknowledged.
Field and lab support was ably provided by R. Brem, C. Briggs, R. Cormack, K. de Clercq, G. Kemp, J. Kemper, S. Lapointe, J. McGillis, D. Neave, H. Reynolds, D. Smith, E. Telfer, G. Wilde, and A. Woo.
We kindly acknowledge the financial support received by this study from Alberta Natural Resources Service, Canadian Wildlife Service, Weldwood of Canada Ltd., Hinton Division, Foothills Model Forest, Wildlife Habitat Canada, Alberta Recreation, Parks and Wildlife Foundation, and Alberta Fish and Game Association.
This report was improved significantly by the editorial reviews of K. Timoney and W. Helprin. The authors appreciate the financial assistance provided by Alberta-Pacific Forest Industries toward the printing of this document.
Multiple forest types (SPRUCE, MIXEDWOOD, and PINE) treated with clear-cutting logging in the mid to late 1950's in west-central Alberta experienced significant changes in plant community structure during the following four decades. Successional patterns in post-logging trajectories occurred for cover and species richness of graminoid, forb, shrub, and nonvascular taxa, and these patterns were influenced greatly by regeneration (both density and height) of hardwood and softwood tree species. In general, ground cover of herbaceous vegetation increased rapidly during Yrs 1-6, declined during Yrs 6-26, then increased during Yrs 26-39. Snags and down wood densities also underwent significant changes following logging. The post-logging successional patterns in aggregate forest structure (live trees, snags, down wood, understory structure, and forage availability) were related to changes in abundance of several wildlife groups.
A major aspect of this research focused on possible differences in forest structure, tree growth, and wildlife response to clearcuts that were either scarified or left unscarified following logging. In general, unscarified cutblocks maintained higher levels of forest structure than did scarified cutblocks. Major elements of increased structural complexity found in unscarified cutblocks included multiple age cohorts of regenerating trees, higher densities of residual trees, higher rates of conifer recruitment into taller height classes, and higher density and more varied rot classes of snags.
Composition and abundance of wildlife taxa were correlated with numerous aspects of forest structure, including tree age class diversity, ratio of hardwood/softwood species, understory visibility, and abundance and rot classes of snags and down wood. Combined use by cervids was higher on unscarified than scarified cutblocks for SPRUCE and PINE forests, and higher on scarified MIXEDWOOD cutblocks than on unscarified cutblocks. Combined use by cavity-dependent birds was higher on unscarified than scarified cutblocks for SPRUCE and MIXEDWOOD forests, and higher on scarified PINE cutblocks than on unscarified cutblocks.
The use of regenerating cutblocks by several wildlife taxal groups (particularly cervids) appeared related to the proximity of unharvested forest strips between the experimental cutblocks. The removal of these residual strips at Yrs 12-15 was associated with significant reductions in cervid densities on the original cutblocks. This outcome suggests the critical importance of appropriate spatial distribution of young and old forests to certain wildlife species.
The major rationale for site scarification and subsequent planting is to improve establishment and growth performance of commercial tree species. The findings of this study over four decades do not ascribe any fiber growth advantage to scarification and subsequent plantings (or aerially seeding) in comparison to strategies where non-merchantable trees are retained, and natural regeneration occurs. Rather, for white spruce trees in both the canopy and the subcanopy strata of SPRUCE and MIXEDWOOD cutblocks, densities were significantly higher in unscarified than scarified cutblocks at the conclusion of this study.
Given the major changes in forest structure caused by scarification (generally leading to a more simple community structure), the reduced abundance of several wildlife taxa associated with this site preparation technique, and the absence of improved commercial tree growth on scarified sites, this report questions the economic and ecological wisdom of this commonly employed forest regeneration technique. This study suggests the need for additional studies that contrast the ecological, social, and economic differences of various harvest and regeneration strategies. It is imperative that these studies include experimental harvest and regeneration treatments that approximate those natural disturbance regimes that perturb forest communities.
Keywords: clearcut logging, scarification, silviculture, biodiversity, forest structure, succession, forest habitat, browse, ungulates
Acknowledgements
Abstract
Contents
Introduction
Study Area:
Logging and Silvicultural Treatments
Experimental Design
Methods:
Results:
Discussion
Summary
Forest Harvest Recommendations relevant
to wildlife habitat management
References
Appendix 1. Location of Photopoints
Appendix 2. Data Sheet Headers
Appendix 3. List of Abbreviations
Appendix 4. Species List
Appendix 5. Data Summary Tables
Appendix 6. Instructions for Future Data Collection
Societal expectations of public forests extend beyond fiber production, requiring forest managers to devise harvest and regeneration strategies that concurrently sustain fiber production, ecosystem processes, and biota (Hunter 1990, Hansen et al. 1991). Achieving these goals has proven challenging and has lead to a critical examination of many industrial forestry practices in Canada (Dancik et al. 1990, Alberta Forest Conservation Strategy 1997) and in the United States (Franklin 1990, Maser 1990). Forest harvest and regeneration systems based on removal of all merchantable trees (e.g., clearcutting) and the physical preparation of the cutblock soil surface (site scarification) for purposes of conifer establishment have dominated Alberta's and Canada's commercial logging history (Natural Resources Canada 1994). This prevalent harvest and regeneration strategy has been encouraged or required by government regulations defining acceptable wood utilization and regeneration standards (for example, Alberta Provincial Operating and Ground Rules 1999). Although the practice of clearcutting and scarification have been widely adopted by silviculturalists as an appropriate strategy for managing even-aged stands of white spruce, lodgepole pine, and jack pine forests, clearcutting practices in general have been criticized for altering above and below-ground forest structure (Berg et al. 1994, Ryden et al. 1997), for impairing natural regeneration of trees (De Grace 1950 Day 1963, Maser 1990, Hansen et al. 1991, Timoney and Robinson 1996), and for altering wildlife habitat and microclimate (Raphael and White 1984, Harmon et al. 1986). In reviewing the scientific foundation for various forestry practices, Timoney (1999) concluded that the ecological and scientific basis for clearcutting and physical site preparation in boreal forests is decidedly weak, and that it's pre-eminence as a harvest and regeneration strategy is based largely on economic considerations.
Based on extensive research of forest ecosystem dynamics in the Pacific Northwest and elsewhere, new harvest and regeneration methods (i.e., "new forestry") have been proposed that allow for the removal of timber volumes while protecting seed trees, providing forest structure, and maintaining shelter for seedling protection (Robertson and Salwasser 199, B.C Ministry of Environment 1995, Weetman 1995, Sauder 1992). In the boreal and Rocky Mountain forests of western Canada, numerous studies (De Grace 1950, Lees 1963, Waldron 1964, Jarvis et al. 1966., Coates and Burton 1997) have proposed various forms of partial cutting as ecologically preferable to clearcutting. Several forestry companies have tested these methods; however, they are sparingly implemented on the commercial landscape, while clearcut harvesting has remained the dominant cutting method in most jurisdictions. There are several reasons why most companies prefer clearcut harvesting. It is the most economical and safe method to harvest trees on a commercial scale. Clearcuts are easy to design on harvesting maps and it is relatively easy to determine pre-harvest tree volumes. However, there is a large body of information compiled that demonstrates that clearcut logging may lead to degradation of naturally functioning ecosystems. These include disruptions to soil flora, altered rates of nutrient cycling, damage to soil structure, and long-term loss of soil organic content. Temporal changes may include altered rates of forest succession and altered successional trajectories. Species diversity of trees can be simplified by clearcut harvesting through planting of one or few tree species. Understory diversity changes can occur by the spread of seeds on harvesting equipment and in the manure of animals that use clearcuts, and by wind dispersal from nearby areas. Off-site effects may include changes to wildlife use of the landscape, alterations to soil hydrology, and changes in regional nutrient budgets.
Long-term studies of ecosystem change are required to understand the processes of forest dynamics over the length of a typical forest cycle (~100 years). Much of our current understanding of long-term forest changes is based on predictions from forest theory (such as forest succession), studies on forests from other geographic locations (e.g. Eastern United States or Europe), or from forest chronosequence studies which link together observations from several different forest stands varying in age (Lee et al. 1995, Stelfox et al. 1995; Lee et al. 1999, Dix and Swan 1971, Holm and Torbjorn 1999). Theories of forest dynamics are critical to understanding general processes but have current limitations because of a lack of empirical evidence and because theories are only useful for the average forest in an area and cannot be used to predict changes on a specific forest site. Similarly, although we can gain substantial insight from studies from other geographic locations, several differences in forest dynamics can be expected to occur at each forest location due to regional or local differences in climate, soils, terrain factors and tree species. A major concern with chronosequence studies is that it is difficult to separate the changes in forest composition and structure that relate to aging or developmental changes from those which are due to site factors or stochastic processes.
To address concerns about the paucity of long-term data on forest and wildlife response to forest practices, a study was initiated in 1956 in the foothills of west-central Alberta to assess forest dynamics following clearcut logging and post-harvest site scarification. Three forest types (spruce, spruce/aspen/pine mixedwood, and pine) were selected. In addition to examining the effects of regeneration among forest types, the regeneration treatment (site scarification) was also compared in each forest type to a non-scarified regeneration strategy, and each regenerating forest was compared to conditions that existed in an uncut control forest. Long-term changes were examined on the following aspects of forest dynamics: tree regeneration (tree density and species composition), forest community structure, forest cover and browse availability, and abundance of selected wildlife populations.
The primary questions addressed during this study were: (6) How does forest community structure change following clearcut logging in each of spruce, mixedwood and pine forests? (7) How similar is each regenerating forest to the conditions in the unharvested controls? (8) How does site preparation (scarification) following clearcut logging affect tree regeneration and forest community structure in each forest type? (9) How do selected wildlife species respond to forest harvest, to site preparation, and to post-harvest forests as the forests develop towards a mature state? This report summarizes patterns in forest community structure and selected flora and fauna observed during the first 40 years following harvest treatment in spruce, mixedwood and pine forests. Previous reports of this study include Stelfox 1962, 1963, 1981, 1983, 1988; Stelfox, Telfer and Lynch 1973; Stelfox, Lynch and McGillis 1974, 1976; and Stelfox and Cormack 1962.
The issue of post-logging establishment and survival of commercial trees is most critical to this study, for its presumed performance is the rationale behind clearcut logging and scarification as the dominant harvest and regeneration strategy adopted in the region. From a silvicultural view, the ultimate gauge of performance must wait for the forest stand to attain merchantable volumes, for only then can one properly calculate meaningful harvest and regeneration costs ($/m3/yr). Since softwood forest stands in the Hinton region generally achieve harvestable volumes and piece sizes by 90-100 years, the forest stands examined in this study are approximately half-way to merchantable age. As such, any conclusions of this study regarding the economic wisdom of scarification and planting following clearcutting are preliminary. The reader is also reminded that the spruce and mixedwood sites examined in this study are located in the Athabasca River Valley bottomlands, a region that receives frequent winter chinooks, strong winds, windblow loess, and frost pocket events, and therefore patterns observed there may not be representative of upland sites. Given these caveats, the findings of this study over four decades do not ascribe any fiber growth advantage to scarification and subsequent plantings (or aerially seeding) in comparison to strategies where non-merchantable trees are retained, and natural regeneration occurs. This conclusion is best illustrated with white spruce regeneration on the spruce and mixedwood cutblocks 40 years after logging. For white spruce trees in both the canopy and the subcanopy strata, densities were significantly higher in unscarified than scarified cutblocks. One must remember that the scarified spruce and mixedwood cutblocks also received single or multiple plantings or aerial seedings of white spruce following scarification. The primary explanations for this discrepancy in white spruce response between scarification/planting and unscarified/natural regeneration appear to lie in two related plant community dynamics phenomena. The first focuses on the non-merchantable trees left on the site at the time of logging. These trees were of moderate density, were established, and displayed reasonable growth in response to the removal of a competing overstory. From an optimal capital investment perspective, their retention at the time of logging represented maintenance of existing natural equity, and the subsequent stand performance benefited from established trees accruing new growth on existing phytomass. In addition, these residual green trees appeared to provide a favorable emerging overstorey for development of seedling and sapling conifers, trees that contributed to the uneven age structure of the regenerating cutblock and would provide future residual trees for subsequent logging events.
Although seedling and sapling densities of white spruce in the scarified cutblocks were moderate (mixedwood cutblocks) to high (spruce cutblocks) by Year 40 (reflecting the manual planting episodes), they did not demonstrate high levels of survival, nor favorable recruitment into the subcanopy or canopy strata. The causal mechanisms for this poor recruitment in scarified cutblocks is unclear, but may reflect unfavorable microclimates associated with an absent overstory or disruptions to the ground surface (and associated microbes) caused by the scarification event. If current stand development trends continue, it is logical to expect that the spruce and mixedwood cutblocks created by natural regeneration on unscarified cutblocks will be ready to receive their second harvest event 1 to 2 decades sooner than those receiving scarification and planting events. The wisdom of scarification and planting strategies becomes even more dubious when one considers the significant costs of site preparation (scarification), and subsequent hand-planting or aerial seeding of conifers. A survey of several forest companies in Alberta revealed 1999 cutblock scarification and planting costs for white spruce at (~$100200/ha) and (~$1000/ha), respectively.
As stated earlier, the spruce and mixedwood cutblocks are found along the Athabasca River valley bottomlands and therefore may not reflect patterns in upland sites. Accordingly, examining tree growth response patterns in the upland pine cutblocks might reveal a different pattern. The fiber growth trajectory of lodgepole pine on scarified and unscarified Pine cutblocks is somewhat different from the spruce and mixedwood cutblocks. In this forest trajectory, young pine established equally well on both treatment types, recruitment into higher strata was favorable, yielding canopy strata densities of ~8001000 stems/ha by the end of the 4th decade. The major differences between scarified and unscarified Pine cutblocks was not in the canopy strata, but the structure and composition beneath. For each of lodgepole pine, white spruce, black spruce and balsam fir, understorey densities were higher on the unscarified cutblocks. This difference in understory tree density was most pronounced for white spruce. Although the site preparation strategy had generated similar volumes of lodgepole pine in the canopy by Yr 40, it came at a cost of reduced volume and tree species composition offered by white spruce, trembling aspen, black spruce, and balsam fir. Although there was no direct seeding or planting cost associated with the scarified Pine cutblocks, there were significant costs associated with scarification following the logging event. It would therefore appear that a management objective of an even-age pine monoculture can be accommodated with a clearcut/scarification strategy. In contrast, the unscarified pine cutblocks had similar lodgepole pine volumes, but had greater structural diversity and volumes of secondary tree species (white spruce, trembling aspen, black spruce, balsam poplar).
Albertans appreciate the environmental, societal, and economic services provided by forest ecosystems to society (Alberta Forest Conservation Strategy 1997), As such, societal demands of forest companies are increasing, requiring forestry practices that ensure reasonable tree regeneration, ecological function, and biodiversity on regenerating forest cutblocks. For these reasons it is imperative that longterm studies such as the one described in this report examine the response of plant community structure and wildlife to different harvest and regeneration strategies.
A significant and predictable change in plant community dynamics occurred following clearcut logging, largely in response to physical changes in plant community structure caused by the logging event. The most pronounced meteorological changes involved temperature and light regimes associated with the removal of all merchantable trees in the forest canopy. From a plant community dynamics perspective, the physical removal of trees also created a large and immediate increase in ground space available to non-tree flora for establishment or expansion. Graminoid cover increased rapidly during the early post-logging years, but declined as cover of forbs and shrubs increased. Shrubs did not respond numerically to logging as quickly as did herbaceous species, but maintained consistent increases in cover throughout the study. As expected, moss cover remained very low or absent during the first 3 decades following logging in all forest types, but had increased considerably by decade 4.
Temporal patterns in understory plants were generally similar between scarified and unscarified cutblocks. Herbaceous cover was marginally higher in unscarified than scarified cutblocks during Yrs 110, but this difference generally disappeared during later stages of the study. There was some evidence that forbs were advantaged by scarification and their cover levels were higher on scarified than unscarified cutblocks in mixedwood and pine forests. Dwarf shrub cover was generally higher during Yrs 127 on unscarified than scarified cutblocks. Although no clear difference in moss cover was detected following treatment types in Spruce and pine forests, moss cover was significantly higher in unscarified mixedwood cutblocks.
Plant species richness generally increased with spruce cutblock age throughout post-harvest succession and also in mature unharvested forest as it progressed through post-rotational seral stages. Plant species richness achieved maximum levels at or near the end of the study in both scarified and unscarified cutblocks; both at values higher than recorded in the mature unharvested forest. Forb species were the highest contributor to species richness, generally followed by shrubs, graminoids (grasses, sedges, rushes), dwarf shrubs, trees, and fern/allies. Increased post-harvest species richness occurred primarily through the addition of forb species. Whereas no major differences in species richness of plant groupings was observed between scarified and unscarified Spruce cutblocks, species richness was generally higher on scarified than unscarified cutblocks in pine and mixedwood cutblocks.
For the Spruce cutblocks on which browse data was most consistently collected, browse biomass was low immediately following logging and then increased sharply during Yrs 19. Maximum browse biomass levels occurred at Yr 17 for unscarified (1,068 kg/ha) and Yr 26 for scarified (934 kg/ha) cutblocks. Browse biomass levels were higher on unscarified than scarified cutblocks during Yrs 117 and did not differ appreciably during Yrs 2639. In general, unscarified cutblocks had higher browse stem densities than did scarified cutblocks.
Snag densities were significantly higher in mature forests than in regenerating cutblocks. At Yrs 32 and 39 for all forest types, snags were more common on unscarified cutblocks than in ones subjected to scarification; this difference in snag density, however, had largely disappeared by Yr 39 for Mixedwood and Pine cutblocks. Generally, snag densities declined between Yrs 32 and 39, suggesting that snag recruitment rates had dropped or that fall-over rates of snags had increased. Snags containing cavities were equally common in mature forests and unscarified cutblocks for spruce and mixedwood forests, but were absent (spruce, mixedwood) or of low densities (pine) on scarified cutblocks.
Mature spruce and mixedwood forests had higher down wood cover than their respective regenerating 40 Yr old cutblocks (Table 15), while scarified and unscarified cutblocks did not differ in amount of down wood. Mature spruce and mixedwood forests had an average down wood cover of 45%, while regenerating scarified and unscarified forests of spruce, mixedwood, and pine cutblocks had down wood covers of 12%.
Mature spruce forests had higher average diameter of down wood than did the regenerating cutblocks (Table 16, Figure 76). No differences in average diameter of down wood occurred between cutblock treatment types for spruce, mixedwood or pine forests, suggesting that scarification had no affect on average diameter of down wood in regenerating cutblocks at Yr 40.
Horizontal visibility through the forest understory, at the conclusion of the study, increased with distance upward from the ground surface, reflecting visual obstruction caused by the dense herbaceous and lower shrub strata. Visibility did not differ between mature spruce forests and regenerating cutblocks (Yrs 32, 39) but was higher on scarified mixedwood cutblocks than either mature forests or unscarified cutblocks. No differences in visibility between scarified and unscarified pine cutblocks were observed.
Security cover and thermal shelter for cervid (deer, elk, moose) species was lower on scarified cutblocks than unscarified cutblocks because of the removal of overstory vegetation and the leveling of understory woody vegetation associated with clearcut logging and subsequent site preparation. Conifer densities for trees greater than 2 m, and hence thermal protection to cervids, was greater on unscarified cutblocks than scarified SPRUCE and MIXEDWOOD cutblocks at Yrs 32 and 39. The extent that scarified and unscarified cutblocks were used by cervids was significantly reduced when the intervening unharvested forests were harvested at Yr 13, indicating the importance of juxtaposition of young regenerating cutblocks and mature forests.
The loss of residual green trees and snags and the lowered structural diversity of scarified cutblocks were associated with lower cavity-dwelling bird richness and abundance in comparison to unscarified cutblocks.