SEARCH

SEARCH BY CITATION

Keywords:

  • balsam fir sawfly;
  • disturbance;
  • forest management;
  • Neodiprion abietis;
  • outbreak;
  • population fluctuation;
  • silviculture;
  • thinning

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • 1
    The stage-specific responses of eruptive populations to ecosystem alterations have received little attention, despite the relevance of the issue for conservation biology and pest management. This study examined the hypothesis that forest thinning treatments affect the outbreak cycles of Neodiprion abietis, a sawfly defoliator.
  • 2
    The densities of eggs, early instar larvae, late-instar larvae, cocoons and adults of N. abietis were monitored during an outbreak in a thinned and an adjacent untreated stand at each of three sites in western Newfoundland, Canada.
  • 3
    The amplitude of population fluctuations was greater in thinned than in adjacent untreated stands at certain points in the increasing or peaking stages of N. abietis outbreaks for all life stages sampled. However, the timing and magnitude of differences in fluctuations attributed to thinning varied among the life stages of N. abietis.
  • 4
    The largest difference between stand types occurred in the generation preceding peak egg density, when adult densities were three times higher in thinned than untreated stands. Conversely, only moderate differences in egg density between stand types were observed in the following generation. Combined with the observation that the modes of population fluctuations occurred earlier in thinned stands, this suggests that adult dispersal resulted in a redistribution of eggs between thinned and untreated stands.
  • 5
    Synthesis and applications. To our knowledge, this is the first study to show that silvicultural practices involving a reduction in forest stand density may alter the outbreak cycles of an eruptive population, potentially increasing the severity of outbreaks in treated, as well as in adjacent untreated, stands. Hence, this study emphasizes the need to consider proactively the potential impact of silvicultural practices on species such as N. abietis, which are most abundant in low-density stands. This field study also demonstrates that the response to ecosystem alteration of an outbreak population can vary with the different life stages of the studied organism, indicating that the detection of the impact of ecosystem alterations on a population can depend on the life stage sampled.

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The potential impact on populations of ecosystem alterations, including those that are human-mediated, is a central issue in conservation biology and pest management. Such alterations have been shown to affect the abundance, distribution and diversity of plants and animals in affected ecosystems, and have been held responsible for the decline of numerous species (Robinson et al. 1992; Wilcove et al. 1998), as well as for the emergence of indigenous species as new pests (Quiring 1990; Abate et al. 2000). Unfortunately, little is known of the impact of ecosystem alterations for eruptive populations, which may be the most likely to reach epidemic levels or be driven toward extinction following an alteration of their environment.

Forest management, the primary goal of which is sustained fibre yield, results inevitably in severe alterations to ecosystems, which in turn may affect resident organisms. The most notable example is that of forest fragmentation, which has been shown to affect animal (Roland 1993; Andrén 1994; Hargis et al. 1999; Wettstein & Schmid 1999) and plant (Turner et al. 1996; Benítez-Malvido & Martínez-Ramos 2003) popu-lations. More subtle alterations of forests, such as the simplification of stand conditions caused by selection cutting, plantation and thinning, may also alter the population dynamics of organisms dwelling in forest ecosystems. Thinning, a cutting treatment applied to forest stands to concentrate growth on a smaller number of retained stems, is one of the most commonly used silvicultural practices throughout the world. Thinning treatments modify stand architecture, illumination, soil temperature, decomposition, mineralization, foliar chemistry of residual trees and understorey composition (François et al. 1985; Wickman & Torgersen 1987; Bauce 1996; Thibodeau et al. 2000; Korb et al. 2003). These changes have been shown to affect both vertebrates (Thompson et al. 2003; Ransome et al. 2004; Sullivan et al. 2005) and invertebrates (Mason et al. 1992; Avtzis & Bombosch 1993; Bauce 1996; McMillin & Wagner 1998), but their effects on the dynamics of eruptive populations throughout an outbreak have not, to the best of our knowledge, been examined previously.

Developing outbreaks of Neodiprion abietis (Harris), the balsam fir sawfly, in thinned and untreated stands of western Newfoundland offered an opportunity to address this gap in our understanding. Recent studies with N. abietis have indicated that thinning may alleviate host-plant inducible responses to defoliation, increase or reduce N. abietis mortality caused by some natural enemies and affect the factors that determine the recruitment of new cohorts (i.e. shifts in sex ratios, variations in fecundity and adult dispersal) (Moreau 2004). Here, we examined the hypothesis that thinning treatments, through the sum of these effects or other mechanisms, alter the outbreak cycles of N. abietis. As the effect of variations in any environmental factor on an organism can vary with the developmental stage of the organism, we also tested the hypothesis that the effect of thinning on N. abietis depended on the developmental stage of the insect.

Materials and methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

study species

The biology and distribution of N. abietis have been described elsewhere (Struble 1957; Carroll 1962; Knerer & Atwood 1972, 1973; Martineau 1985; Wallace & Cunningham 1995; Piene et al. 2001; Anstey et al. 2002; Moreau et al. 2003; Moreau 2004). In western Newfoundland, natural populations of N. abietis are univoltine. Neodiprion abietis overwinters as eggs laid in the current-year foliage of balsam fir [Abies balsamea (L.) Mill.]. Eggs hatch 2–4 weeks after budburst the following spring and the gregarious larvae initiate feeding on old needles (≥ 1 year old). Male larvae have five instars and complete their development within 30 days, whereas females can have five or six instars and complete their development within 35 days (Carroll 1962). Last-instar larvae generally spin a cocoon directly on the foliage (Carroll 1962) and adults emerge 2–4 weeks later in late summer/early autumn. Although oviposition is observed only on balsam fir, larvae may also feed on white spruce Picea glauca (Moench) Voss, black spruce Picea mariana (Mill.) BSP and tamarack Larix laricina (Du Roi) K. Koch (Carroll 1962).

study area

Between 1997 and 2004, we monitored N. abietis populations for six to eight successive generations in a precommercially thinned and an adjacent untreated stand at each of three sites in western Newfoundland. Thinned stands had been treated more than 10 years previously. At each location, the pair of stands occurred in a continuous forest from which a section had been thinned and another not thinned. The pairs of stands were located at Lakeside (untreated stand: 48°35′59″ N, 58°6′23·1″ W; thinned stand: 48°35′59·1″ N, 58°6′15·5″ W), Stag Hill (untreated stand: 48°49′46·9″ N, 58°3′44·5″ W; thinned stand: 48°49′46·5″ N, 58°3′48·2″ W) and Stag Lake (untreated stand: 48°50′53·6″ N, 58°0′11·7″ W; thinned stand: 48°50′54·5″ N, 58°0′2·3″ W). Paired stands at each location were > 1 ha in size and were less than 5 m (Stag Hill and Stag Lake) or 60 m (Lakeside) apart from each other separated by either a trail or a gravel road. Uniformity in stand conditions was determined using local indices (Meades & Moores 1994) to eliminate the potential effects of environmental gradients (e.g. altitude, drainage, soil) on sawfly density that might overrule the effects of thinning. In the process of stand selection, care was taken to ensure that the area surrounding paired stands was comparatively homogeneous in all directions.

Each stand was a typical balsam fir stand found within the Corner Brook subregion of the western Newfoundland ecoregion (Meades & Moores 1994). Stands were composed of naturally regenerated balsam fir (over 90% of the basal area) growing at densities of 14 000–24 000 (untreated stands) or 2500–5000 (thinned stands) trees per hectare and mean tree ages (± SEM) for untreated and thinned stands were similar (respectively, 30·8 ± 1·5 and 30·4 ± 0·8 years old in Stag Hill, 26·4 ± 0·2 and 25·4 ± 0·7 years old in Stag Lake and 26·6 ± 0·7 and 27·8 ± 0·9 years old in Lakeside; n = five trees per stand; t-test for pairs of stands: t8  1·39; P  0·20). The understorey was composed mainly of Dryopteris spinulosa[(O.F. Muell.) Watt], Cornus canadensis (L.) and Clintonia borealis[(Ait.) Raf]. The moss layer was dominated by Dicranum majus (SM), Hylocomium splendens[(Hedw.) Schimp. in BSG] and Pleurozium schreberi[(Willd. ex Brid.) Mitt].

sampling procedure

Sampling procedures were similar to those described in Moreau et al. (2003) and Parsons et al. (2005). From 1997 to 2002, sampling was conducted before egg hatch, between first and second instars, between third and fourth instars and at the end of adult emergence to determine the density of eggs, early instar larvae, late-instar larvae, cocoons and adults, respectively. In 2003 and 2004, sampling was conducted only before egg hatch in all stands because sampling low densities of larvae and cocoons would have required a sampling intensity that could have caused significant damage to stands. Frequent monitoring of field populations enabled us to determine when previously specified stages of development had been reached. To account for habitat heterogeneity, long, thin sampling areas were used (Krebs 1999). In each stand, along a 150-m transect, a 15-m diameter plot was established every 30 m for a total of five plots per stand. In each plot, at every sampling date, one midcrown branch was sampled from five dominant or codominant balsam fir trees selected randomly without repetition, for a total of 25 branches per stand per sampling date. Sawfly densities are more stable throughout juvenile development in the midcrown than in other crown levels (Anstey et al. 2002). The numbers of eggs, larvae and eclosed and uneclosed cocoons present on the branches were recorded. To express N. abietis density in terms of surface area of foliage, the branch surface area (foliated and defoliated) was estimated using the branch maximum length and mean width. Density estimates were averaged by stands.

data analysis

To examine stage/age-specific (hereafter, stage-specific) fluctuations of populations, we plotted sawfly densities of eggs, early instar larvae (L1–L2), late-instar larvae (L3–L4), cocoons and adults (hereafter life stages) in thinned and untreated stands against the number of generation(s) from the time when egg densities peaked at individual stands (Fig. 1). To avoid drawing inferences from limited evidence, data were examined only for the period from 2 years before to 3 years after peak population density (i.e. the only period for which data were collected in all six stands). For each life stage and thinning treatment, a polynomial Poisson regression was fitted to determine the trend of fluctuations in density and the associated error.

image

Figure 1. Relationships between the densities of (a) eggs; (b) early instar larvae (L1–L2); (c) late-instar larvae (L3–L4); (d) cocoons; and (e) adults and the number of generation(s) from the time when egg densities peaked in thinned (closed shapes) and untreated (open shapes) stands at Lakeside (diamonds), Stag Hill (circles) and Stag Lake (triangles). Curves in thinned (thick solid lines) and untreated (thick dashed lines) stands and their 95% confidence intervals (fine lines) were fitted using the Poisson regressions described in Table 1. Shaded areas indicate significant difference in density between thinned and untreated stands (χ2 tests of least-square means; P < 0·01).

Download figure to PowerPoint

The adjusted r-square values were calculated using the methods described in Mittlböck & Waldhör (2000) and Mittlböck (2002).

To evaluate whether sawfly densities differed in thinned and untreated stands, we subtracted the mean density of sawflies in the treated stand from that of the adjacent untreated stand. Because longitudinal data thus produced had non-Gaussian distributions, they could not be examined using a standard repeated-measures anova. Instead, to determine in which generation mean contrasts differed from zero while accounting for repeated measurements within sites, χ2 tests of least-square means were produced for each life stage using generalized estimating equations (Liang & Zeger 1986) that included generation as the independent variable. Sequential Bonferroni adjustments were performed on χ2 matrices to maintain levels of significance at α = 0·05. To examine the trend of fluctuations in thinned and adjacent untreated stands, slopes and intercepts of regressions of curve modes carried out by thinning treatment were compared. The value of the exponent in regressions was calculated by determining the best fit for the overall model including both treatments.

All statistical analyses were performed using SAS (SAS Institute 1999). Residuals were examined to ensure that postulates of parametric analyses were respected. All data in tables and figures are presented in their original, untransformed scale.

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Populations increased and declined during the study period at all sites (Fig. 1). For each life stage and thinning treatment, these population fluctuations were remarkably similar at the different sites, with regressions capturing between 74–91% (thinned stands) and 42–89% (untreated stands) of the variation in density (Fig. 1; Table 1). The predictive power of regressions (i.e. the r-square value) for annual variations in sawfly densities in thinned stands was equal to or greater than that obtained in untreated stands, and declined progressively in the later stages of insect development (Table 1). Accordingly, confidence intervals were wider in the later stages of development (Fig. 1).

Table 1.  Coefficient estimates of variables (± SE) and r-squares of Poisson regressions illustrated in Fig. 1. The regression with the highest significant polynomial term was kept. The regression model is log(µ) = ax + bx2 + cx3 + dx4 + ex5 + f + ɛ
Life stageTreatmentabcdefr2
  • *

    These coefficient estimates of lower-order terms did not differ from zero at α = 0·05 but were kept to maintain the integrity of the model.

EggsThinned 0·42 ± 0·05−1·31 ± 0·05−0·18 ± 0·020·15 ± 0·015·91 ± 0·030·90
Untreated 0·98 ± 0·06−1·28 ± 0·06−0·30 ± 0·020·16 ± 0·015·67 ± 0·030·89
Early instar larvaeThinned−0·23 ± 0·08−1·32 ± 0·07 0·33 ± 0·060·18 ± 0·02−0·06 ± 0·015·59 ± 0·040·91
Untreated 0·35 ± 0·09−0·85 ± 0·09 0·06 ± 0·07*0·09 ± 0·02−0·02 ± 0·014·93 ± 0·050·79
Late-instar larvaeThinned−0·50 ± 0·11−0·98 ± 0·10 0·32 ± 0·090·14 ± 0·03−0·05 ± 0·014·84 ± 0·050·87
Untreated 0·74 ± 0·09−1·14 ± 0·09−0·21 ± 0·040·13 ± 0·024·81 ± 0·050·79
CocoonsThinned−1·40 ± 0·19−1·07 ± 0·16 0·87 ± 0·130·17 ± 0·04−0·11 ± 0·023·97 ± 0·080·74
Untreated−0·54 ± 0·21−1·04 ± 0·19 0·41 ± 0·160·14 ± 0·05−0·06 ± 0·033·58 ± 0·100·62
AdultsThinned−1·97 ± 0·35−0·45 ± 0·32* 0·85 ± 0·240·03 ± 0·08*−0·17 ± 0·042·73 ± 0·180·75
Untreated−0·60 ± 0·18−0·33 ± 0·08 0·09 ± 0·052·32 ± 0·150·42

χ2 tests of least-square means and Poisson regressions indicated that the amplitude of population fluctuations was greater in thinned than in adjacent untreated stands at certain points in the increasing or peaking stages of N. abietis outbreaks for all life stages sampled (Fig. 1). However, the timing and magnitude of differences in fluctuations attributable to thinning varied among the life stages of N. abietis; these differences occurred at the generation preceding peak egg density in eggs, late-instar larvae and adults (Fig. 1a,c,e), and in both the generation preceding peak egg density and the generation of peak egg density in early instar larvae and cocoons (Fig. 1b,d). The largest difference between stand types occurred in the generation preceding peak egg density when adult densities were three times higher in thinned than untreated stands (Fig. 1e).

Regressions were fitted to the modes of population fluctuations, estimated by Poisson regressions in Fig. 1

(untreated stands: y0·00001 = 1·00002 + 0·00003x; r2 = 0·93; F1,3 = 53·26; P < 0·01; thinned stands: y0·00001 = 1·00004 + 0·00003x; r2 = 0·97; F1,3 = 142·48; P < 0·01) (Fig. 2). Slope estimates of regressions were similar in thinned and untreated stands (t6 = −0·26; P = 0·80) and intercept estimates were higher in thinned than in untreated stands (t6 = 3·81; P < 0·01). This indicates that regression lines were parallel, although mode values occurred earlier and at greater densities in thinned stands.

image

Figure 2. Overlay of the Poisson regressions from Fig. 1 and Table 1. The modes of the regression curves (black and white squares) were joined using a regression shown as a solid (thinned stands) and a dashed (untreated stands) thick line.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Results from this study support the hypothesis that thinning treatments alter the outbreak cycles of N. abietis. For all life stages, at certain points in the increasing or peaking stages of N. abietis outbreaks, the amplitude of population fluctuations was greater in thinned than in adjacent untreated stands. Contemporary studies carried out in the same area indicated that thinning resulted in larger tree diameter, longer terminal shoots, higher numbers of young shoots on each branch and higher foliar content of monoterpenoids (G. Moreau, unpublished data). We may speculate that differences in N. abietis outbreak dynamics are attributable to some of these changes in host plants, to the effects of thinning on host-plant inducible responses to defoliation (Moreau 2004), to thinning effects on the mortality caused by natural enemies and on recruitment of new cohorts (Moreau 2004), as well as to other factors. Similar mechanisms may also explain why N. abietis (Atwood 1960; Drooz 1985; Martineau 1985) and several other Neodiprion species (reviewed in McMillin & Wagner 1993; McMillin et al. 1996) generally reach high population levels in naturally low-density stands, on forest edges and on isolated host plants. To the best of our knowledge, this study is the first to show that changes in tree density, independent of its effect on crown closure (Ostaff & Quiring 2000), may alter the outbreak cycles of a defoliator population.

Increases in sawfly density attributable to thinning appeared in the generation preceding peak egg density and disappeared in the generation following peak egg density. The largest difference between stand types occurred in the generation preceding peak egg density, when adult densities were three times higher in thinned than untreated stands. Even greater differences in egg density might have been expected in the subsequent generation as potential fecundity, which is influenced by both population sex ratio and female realized fecundity, is generally greater in thinned than in untreated stands (Moreau 2004). However, only moderate differences in egg density between stand types were observed at peak egg densities, suggesting that adult predation is higher in thinned stands or that adult dispersal resulted in egg redistribution between stand types. Such adult movement from thinned to untreated stands would result in oscillations occurring later in untreated stands, but not necessarily a complete generation later, as population growth and decline in untreated stands are not dependent solely upon immigration from thinned stands. This condition was supported by the present data. As oviposition, as indicated by egg density, varies little between thinned and untreated stands, our study suggests that the large differences in density observed sometimes during the larval, cocoon and adult stages resulted from greater juvenile survival in thinned stands.

Our study suggests, therefore, not only that thinning treatments may increase the density of N. abietis in thinned stands by increasing juvenile survival, but also that they may, because of adult dispersal, increase sawfly densities in adjacent untreated stands. Greater densities of early instar and late-instar larvae in the early phase of outbreaks should result in greater levels of defoliation (Parsons et al. 2003, 2005), which may cause significant losses of wood volume (Piene et al. 2001; Parsons et al. 2003). This interpretation is supported by current and historical trends of N. abietis defoliation in Newfoundland, which indicate that current outbreaks in both untreated and thinned stands are unprecedented in severity (Anonymous 1943–96), compared with previous outbreaks that occurred only in natural stands before intensification of silvicultural practices.

management implications

Thinning has been employed widely in western Newfoundland to increase the growth of retained stems. In fact, the small number of untreated stands in the outbreak area restricted the extent of replication in this study. Silvicultural treatments must be adapted to environmental characteristics (Schütz 1999; van der Kelen 2003) and it may well be possible that large-scale thinning treatments are not suitable for the ecological conditions of all regions in Newfoundland. The apparent preference of N. abietis for open-grown trees (Atwood 1960; Drooz 1985; Martineau 1985) may have predisposed this species for increased pest status in thinned stands. Thus, this study emphasizes the need to consider proactively the potential impact of silvicultural practices on species, such as N. abietis, that are most commonly observed in low-density stands. This would permit forest managers to estimate potential losses and include them in management plans, to develop pest management strategies in advance or to prescribe alternative strategies for the management of forest areas. Nevertheless, in the absence of alternative silvicultural methods to improve the yield of conifer stands at low cost, thinning remains an essential tool for the management of boreal forests. In this context, aerial applications of the baculovirus NeabNPV may provide an efficacious yet environmentally friendly means to reduce the undesired impact of thinning on population outbreaks of N. abietis (Moreau et al. 2005).

This study may be the first to show that the response of an eruptive population to ecosystem alteration varies during an outbreak according to the different life stages of the studied organism. Studying all the stages enabled us to detect indirectly adult dispersal and to suggest an explanation for the increased defoliation observed during current N. abietis outbreaks in Newfoundland. For practical reasons, the appraisal of the impact of perturbations on populations is usually carried out with a census of the life stages that can be sampled easily. However, in many instances a direct census is not carried out and the human impact on animal populations is assessed only through changes in animal behaviour (e.g. Gosling & Sutherland 2000; Gill et al. 2001). This study has shown that a better understanding of the impact of ecosystem alterations can be gained through a census of the life stages responsible for the recruitment of the next cohort and, in the case of pest species, of life stages that cause damage. Such knowledge would significantly contribute to our understanding of disturbances and of ways to attenuate undesirable impacts of anthropogenic disturbances on populations.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

We thank R. Graves, S. Heard, P. Price, J. Sweeney and anonymous reviewers for comments on an earlier version of this manuscript; J. Buffett, G. Butt, B. Davis, A. Morrison, J. McIsaac, J. Park, J. Parsons, K. Parsons, A. Sharpe, M. Sharpe, M. Slaney, S. Stryde, R. Sutton and J. Warren for technical assistance; H. Piene and R. West for help with site selection; and W. Brown, H. Crummey, J. Evans, G. Fleming, D. Ostaff and G. VanDusen for their valuable collaboration. This work was supported by the Canadian Forest Service, Forest Protection Limited, the Biocontrol Network, a NSERC Discovery Grant, a NSERC postdoctoral fellowship and by a CFS/NSERC Research Partnership Grant with Abitibi Consolidated, Corner Brook Pulp and Paper and the Newfoundland and Labrador Department of Natural Resources as industrial contributors.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Abate, T., van Huis, A. & Ampofo, J.K.O. (2000) Pest management strategies in traditional agriculture: an African perspective. Annual Review of Entomology, 45, 631659.
  • Andrén, H. (1994) Effects of habitat fragmentation on birds and mammals in landscapes with different proportions of suitable habitats: a review. Oikos, 71, 355366.
  • Anonymous (194396) Annual Report. Newfoundland Forest Protection Association, St John's, NL.
  • Anstey, L.J., Quiring, D.T. & Ostaff, D.P. (2002) Seasonal changes in intra-tree distribution of immature balsam fir sawfly (Hymenoptera: Diprionidae). Canadian Entomologist, 134, 529538.
  • Atwood, C.E. (1960) Present status of the sawfly family Diprionidae (Hymenoptera) in Ontario. Proceedings of the Entomological Society of Ontario, 91, 205215.
  • Avtzis, N. & Bombosch, S. (1993) Auswirkungen vershiedener Pflegemabnahmen auf die Nahrungsqualitat von Fichten fur Phytophage Insekten. Allgemeine Forst und Jagdzeitung, 164, 2122.
  • Bauce, É. (1996) One and two years impact of commercial thinning on spruce budworm feeding ecology and host tree foliage production and chemistry. Forestry Chronicle, 72, 393398.
  • Benítez-Malvido, J. & Martínez-Ramos, M. (2003) Impact of forest fragmentation on understorey plant species richness in Amazonia. Conservation Biology, 17, 389400.
  • Carroll, W.J. (1962) Some aspects of the Neodiprion abietis (Harr.) complex in Newfoundland. PhD Dissertation, Syracuse University, Syracuse, NY.
  • Drooz, A.T. (1985) Insects of Eastern Forests. USDA Forest Service Miscellaneous Publication 1426. USDA Forest Service, Washington, DC.
  • François, F., André, P. & Devillez, F. (1985) Effet de l’intensité de l’éclaircie sur l’extinction du rayonnement solaire en jeune futaies de Picea abies (L.) Karst. Annales des Sciences Forestières, 41, 439448.
  • Gill, J.A., Norris, K. & Sutherland, W.J. (2001) Why behavioural responses may not reflect the population consequences of human disturbance. Biological Conservation, 97, 265268.
  • Gosling, L.M. & Sutherland, W.J. (2000) Behaviour and Conservation. Cambridge University Press, Cambridge.
  • Hargis, C.D., Bissonette, J.A. & Turner, D.L. (1999) The influence of forest fragmentation and landscape pattern on American martens. Journal of Applied Ecology, 36, 157172.
  • van Der Kelen, G. (2003) Sylviculture(s): féminin pluriel. L’aubelle, 143, 1924.
  • Knerer, G. & Atwood, C.E. (1972) Evolutionary trends in the subsocial sawflies belonging to the Neodiprion abietis complex (Hymenoptera: Tenthredinoidea). American Zoologist, 12, 407418.
  • Knerer, G. & Atwood, C.E. (1973) Diprionid sawflies: polymorphism and speciation. Science, 179, 10901099.
  • Korb, J.E., Johnson, N.C. & Covington, W.W. (2003) Arbuscular mycorrhizal propagule densities respond rapidly to ponderosa pine restoration treatments. Journal of Applied Ecology, 40, 101110.
  • Krebs, C.J. (1999) Ecological Methodology, 2nd edn. Addison-Welsey Educational Publishers Inc., Menlo Park, CA.
  • Liang, K.Y. & Zeger, S.L. (1986) Longitudinal data analysis using generalized linear models. Biometrika, 73, 1322.
  • Martineau, R. (1985) Insectes nuisibles des forêts de l’est du Canada. Ressources Naturelles Canada, Service Canadien des Forêts Cat. Fo64–32/1984F. Ressources Naturelles Canada, Broquet, Ottawa, ON.
  • Mason, R.R., Wickman, B.E., Beckwith, R.C. & Paul, G.H. (1992) Thinning and nitrogen fertilization in a grand fir stand infested with western spruce budworm. Part I: Insect response. Forest Science, 38, 235251.
  • McMillin, J.D., Hengxiao, G., Wagner, M.R. & Long, X. (1996) Spatial distribution patterns of pine sawflies (Hymenoptera: Diprionidae) in Arizona, US and Sichuan, P.R. of China. Forest Ecology and Management, 86, 151161.
  • McMillin, J.D. & Wagner, M.R. (1993) Influence of stand characteristics and site quality on sawfly population dynamics. Sawfly Life History Adaptation to Woody Plants (eds M.Wagner & K.F.Raffa), pp. 333361. Academic Press, San Diego, CA.
  • McMillin, J.D. & Wagner, M.R. (1998) Influence of host plant vs. natural enemies on the spatial distribution of a pine sawfly, Neodiprion autumnalis. Ecological Entomology, 23, 397408.
  • Meades, W.J. & Moores, L. (1994) Forest Site Classification Manual A Field Guide to the Forest Types of Newfoundland. Canada − Newfoundland Forest Resource Development Agreement. FRDA Report 003. Natural Resources Canada/Canadian Forest Service, St John's, NL.
  • Mittlböck, M. (2002) Calculating adjusted R2 measures for Poisson regression models. Computer Methods and Programs in Biomedicine, 68, 205214.
  • Mittlböck, M. & Waldhör, T. (2000) Adjustments for R2-measures for Poisson regression models. Computational Statistics and Data Analysis, 34, 461472.
  • Moreau, G. (2004) The influence of forest management on defoliator populations: a case study with Neodiprion abietis in precommercially thinned and natural forest stands. PhD Dissertation, University of New Brunswick, Fredericton, NB.
  • Moreau, G., Lucarotti, C.J., Kettela, E.G., Thurston, G.S., Holmes, S., Weaver, C., Levin, D.B. & Morin, B. (2005) Aerial application of nucleopolyhedrovirus induces decline in increasing and peaking populations of Neodiprion abietis. Biological Control, 33, 6573.
  • Moreau, G., Quiring, D.T., Eveleigh, E.S. & Bauce, É. (2003) Advantages of a mixed diet: feeding on several foliar age classes increases the performance of a specialist insect herbivore. Oecologia, 135, 391399.
  • Ostaff, D.P. & Quiring, D.T. (2000) Population trends of a specialist herbivore, the spruce bud moth, in young white spruce stands. Canadian Entomologist, 132, 825842.
  • Parsons, K., Quiring, D., Piene, H. & Farrell, J. (2003) Temporal patterns of balsam fir sawfly defoliation and growth loss in young balsam fir. Forest Ecology and Management, 184, 3346.
  • Parsons, K., Quiring, D., Piene, H. & Moreau, G. (2005) Relationship between balsam fir sawfly density and defoliation in balsam fir. Forest Ecology and Management, 205, 325331.
  • Piene, H., Ostaff, D. & Eveleigh, E. (2001) Growth loss and recovery following defoliation by the balsam fir sawfly in young, spaced balsam fir stands. Canadian Entomologist, 133, 675686.
  • Quiring, D.T. (1990) Effet de l’établissement de plantations sur l’entomologie forestière aux Maritimes. Revue d’entomologie du Québec, 35, 1824.
  • Ransome, D.B., Lindgren, P.M.F., Sullivan, D.S. & Sullivan, T.P. (2004) Long-term responses of ecosystem components to stand thinning in young lodgepole pine forest. I. Population dynamics of northern flying squirrels and red squirrels. Forest Ecology and Management, 202, 355367.
  • Robinson, G.R., Holt, R.D., Gaines, M.S., Hamburg, S.P., Johnson, M.L., Fitch, H.S. & Marinko, E.A. (1992) Diverse and contrasting effects of habitat fragmentation. Science, 257, 524526.
  • Roland, J. (1993) Large-scale forest fragmentation increases the duration of tent caterpillar outbreaks. Oecologia, 93, 2530.
  • SAS Institute (1999) SAS/STAT User's Guide, Version 8. Statistical Analysis System Institute, Cary, NC.
  • Schütz, J.P. (1999) Close-to-nature silviculture: is this concept compatible with species diversity? Forestry, 72, 359366.
  • Struble, G.R. (1957) Biology and control of the white-fir sawfly. Forest Science, 1, 196209.
  • Sullivan, T.P., Sullivan, D.S., Lindgren, P.M.F. & Ransome, D.B. (2005) Long-term responses of ecosystem components to stand thinning in young lodgepole pine forest. II. Diversity and population dynamics of forest floor small mammals. Forest Ecology and Management, 205, 114.
  • Thibodeau, L., Raymond, P., Camiré, C. & Munson, A.D. (2000) Impact of precommercial thinning in balsam fir stands on nitrogen dynamics, microbial biomass, decomposition, and foliar nutrition. Canadian Journal of Forest Research, 30, 229238.
  • Thompson, I.D., Baker, J.A. & Ter-Mikaelian, M. (2003) A review of the long-term effects of post-harvest silviculture on vertebrate wildlife, and predictive models, with an emphasis on boreal forests in Ontario, Canada. Forest Ecology and Management, 177, 441469.
  • Turner, I.M., Chua, K.S., Ong, J.S.Y., Soong, B.C. & Tan, H.T.W. (1996) A century of plant species loss from an isolated fragment of lowland tropical rain forest. Conservation Biology, 10, 12291244.
  • Wallace, D.R. & Cunningham, J.C. (1995) Diprionid sawflies. Forest Insect Pests in Canada (eds J.A.Armstrong & W.G.H.Ives), pp. 193232. Natural Resources Canada, Canadian Forest Service Cat. Fo24–235/1995E, Ottowa, ON.
  • Wettstein, W. & Schmid, B. (1999) Conservation of arthropod diversity in montane wetlands: effect of altitude, habitat quality and habitat fragmentation on butterflies and grasshoppers. Journal of Applied Ecology, 36, 363373.
  • Wickman, B.E. & Torgersen, T.R. (1987) Phenology of Douglas-fir tussock moth, Orgyia pseudotsugata, egg eclosion and mortality in a thinned and unthinned stand (Lepidoptera, Lymantriidae). Pan-Pacific Entomologist, 63, 218223.
  • Wilcove, D.S., Rothstein, D., Dubow, J., Phillips, A. & Losos, E. (1998) Quantifying threats to imperilled species in the United States. Bioscience, 48, 607615.