Reasons for variability in spruce budworm caused tree mortality and defoliation
Classification and regression trees were used to analyze the spatial relationships between the presence and frequency of SBW defoliation at the landscape scale, revealing that the boundaries of SBW defoliation were related to host species tree abundance and that defoliation frequency increased with less precipitation in summer and low temperature in spring (Candau and Fleming 2005). Nonetheless, according to Gray and Mackinnon (2007), the spatial patterns of SBW defoliation cannot be explained by host species abundance, i.e., host species volume. Also, by examining annual digitalized SBW defoliation maps using cluster analyses and cross-correlations, it was concluded that SBW outbreaks were synchronized in large landscape because of high dispersal rate as well as a “Moran effect” (i.e., regional stochasticity; Williams and Liebhold 2000) instead of extended epicenters (forest stands with low susceptibility; Hardy et al 1983) or synchronized intrinsic factors (Royama 1984). But Peltonen et al. (2002) indicated that spatial synchrony was not directly dependent on insect dispersal, yet primarily influenced by the “Moran effect”. Other research also indicated that climate conditions played an essential role in affecting outbreak location, duration, and severity (Gray 2007, 2013). In terms of the temporal patterns, Candau et al. (1998) concluded that a complicated temporal fluctuation of SBW outbreaks at a large scale consisted of a basic oscillation (around 36 years) and a secondary oscillation which can be explained by fluctuations at smaller scales.
Craighead (1925) determined that trees with more vigor corresponded to lower mortality, but the mortality did not have a significant relationship with density, basal area, or species composition. It was declared by Turner (1952), however, that balsam fir mortality was related to fir proportion among all tree species and the actual basal area in the stand. In addition, Batzer (1969) found that the proportion of spruce basal area, combined with the proportion of non-host species and total fir basal area explained 56% of the mortality variation by statistical analysis. Patterns of fir mortality after an uncontrolled SBW outbreak were assessed in 20 stands on Cape Breton Island, suggesting that fir death was negatively correlated with tree vigor, crown position, and DBH at the beginning state of the outbreak (MacLean and Ostaff 1989). As Bouchard and Pothier (2010) stated, the degree of SBW caused fir mortality is a complicated phenomenon affected by climatic or other random factors.
Previous studies methods and results
Defoliation
At the landscape scale, analysis of aerial defoliation survey data has suggested that severe defoliation patterns during the past 1970s-1990s SBW outbreak in New Brunswick tended to be in central-north of the province, and light defoliation tend to be around the province boundary (Zhao et al. 2014). By using historical aerial survey data of SBW defoliation in Quebec from 1965 to 1992, Grey et al. (2000) determined 25 representative spatiotemporal patterns of SBW defoliation and managed to predict the cause of the following SBW outbreak accordingly.
Mortality
Spruce budworm caused tree mortality and stand recovery over 20 years were examined for 10 plots in a stand located in north-western New Brunswick (Baskerville and MacLean 1979). The research determined a high variability of tree mortality between the plots in the uniform stand; the proportion of tree mortality ranged from 34 to 84% for balsam fir and from 0 to 24% for spruce trees. The variability was concluded that not related to stand characteristics including density or species composition but was a function of variable budworm pressure (e.g. uneven within-stand adult dispersal), and the range of mortality was strongly affected by plot size. Additionally, apparent contagious distribution of mortality patterns was observed within the plots, that obvious “holes” developed in the affected stand as the tree mortality progressed. Also, high spatial plot-to-plot variability in SBW-caused tree mortality occurred in spaced and unspaced stands on Cape Breton Island, Nova Scotia; mortality ranged from 31-49% and 11-32% in spaced and unspaced young fir plots, with the degree of variability related to plot sizes (MacLean and Piene 1995).
Mortality patterns were also studied previously. For example, Baskerville and MacLean (1979) noted that sequential tree mortality tended to a distribution that leads to “holes” in the stand as tree death progressed. High spatial plot-to-plot variability in tree mortality was also concluded by MacLean and Piene (1995). However, it should be noted that mortality is the result of severe defoliation, and their variability does not necessarily to be similar, particularly for those lightly defoliated stands.
Possible reasons for clustered defoliation patterns
Oviposition site selection by female moths
The number of eggs of each mass varies considerably, but a typical mass contains about 20 eggs (Morris 1955). A large proportion of the egg masses are generally deposited on the peripheral shoots of the crown (Morris 1955). Population densities of larvae are usually greater in the upper portions of tree crowns because of the ovipositional behavior of the female moths (Beckwith and Burnell 1982).
According to Morris and Mott (1963), the frequency distribution of budworm numbers per tree in a stand are positively skewed, namely, a small proportion of the trees supporting populations much greater than the average. It should be caused by unequal attraction for female moths to trees in the stand. This either due to trees of the same species differ in the strength of their stimulus to the olfactory responses of the insect just as do trees of different species (Morris and Mott 1963), or due to the position or exposure to light of individual trees in a stand makes them unequally susceptible to attack (Morris 1955). It was suggested that female ovipositional behavior is greatest around tallest and dominant trees which are well exposed to light, and those trees tend to receive the highest egg amounts year after year (Morris 1955).
SBW pressure is uneven? Larval dispersal?
Dispersal of a forest insect can contribute to population redistribution within and between tree crowns and stands (Beckwith and Burnell 1982). Throughout the L1 and feeding stages of SBW, dispersal only happens within trees or a stand, but budworms may move several times not only during the L1 stage but also during their feeding stages before locating at a suitable feeding location.
In fact, mass movements are observed only in very severely infested stands where the food supply becomes exhausted (Morris and Mott 1963). Also, a large segment of the population must be involved in dispersal movements to cause such remarkable changes in tree-to-tree population levels.
SBW egg masses are usually on the needles in the up-crown of dominant host trees (Beckwith and Burnell 1982). Emerging from eggs, the photopositive L1 redistribute within the canopy to move toward the branch tip (Morris 1963). McKnight (1969) also reported that densities of hibernacula are reportedly higher in lower and mid-crown levels because of redistribution of L1 emerging from eggs and favorable overwintering sites. After emerging from hibernacula, L2 move from these interior or mid-crown to exterior reaches of up-crowns (Moody and Otvos 1980). L2 also tends to dispersal between crowns. L2 dispersal occurs between crowns more often than within trees (Regniere and Fletcher 1983). Generally speaking, redistribution within stands should be lateral or downward (Beckwith and Burnell 1982).
According to (Regniere and Fletcher 1983), differences between L1 and L2 dispersal behavior are profound: 1) L1 dispersal mostly happen within the crown, while L2 dispersal usually between crowns, 2) L1 landing on the ground do not crawl back up toward the host trees while L2 do, 3) dispersing number is larger in top-crown in L1 than in L2 compared to lower-crown. There were no obvious changes in density or vertical distribution of airborne larvae with distance from host trees. They noticed that capture trends on ground traps in different study locations were quite different, which was explained by different lower canopy structures, and they suggested that larval settling under host trees is in direct relation to dispersal in low-crown.
The distance that larvae can travel after emerging from eggs or hibernacula is not well-known but is probably limited to a few kilometers (Blais 1985). Generally, not much air movement is required to affect larval drift because of body weight and silk thread of SBW although wind direction varied drastically (Beckwith and Burnell 1982). Closely located tree crowns have a progressive filtering effect on larval dispersal, which is a function of stand density and air currents (Beckwith and Burnell 1982). In addition, how far small larvae can transport could vary considerably over years, depending on climate and food sources (Blais 1952 The relationship of the SBW to the flowering condition of BF). Also, some factors were believed to have an impact on the success of L2 dispersal in finding appropriate feeding spots, including the height of hibernation concerning the crown canopy, stand density, species composition, and air turbulence during dispersal (Morris and Mott 1963). The timing of spring dispersal of L2 varies as well because it corresponds closely with the timing larvae emerging from hibernacula (Regniere and Fletcher 1983).
Asynchrony
Beckwith and Kemp (1984) indicated that the synchrony of host trees and defoliator phenology has an impact on defoliation intensity. Kemp (1985) suggested that springtime larval survival was greater when adequate host shoots were sufficiently expanded than in years when the reverse was true. Beckwith and Burnell (1982) suggested that dispersal activity and development rates of 2nd-instar budworm would be the same, whereas bud-to-needle development would be much slower in the second case. The authors believed that under colder soil temperature conditions, early dispersing budworm would be more apt to find successful feeding sites on an alternate host, such as western larch, which has earlier bud development (Beckwith and Burnell 1982)
Larval survival
Larval dispersal by air within a stand may contribute to three following processes: 1) population redistribution within a stand, 2) population spreading to adjacent stands, and 3) population loss due to non-host trees.
However, according to Regniere and Fletcher (1983), L2 losses due to landing on the ground can be very low and depend on stand density and composition, and such losses would be partly compensated by larval crawling back up toward the host trees. Regniere and Nealis (2008) determined that survival of early instars had a density-dependent relationship condition of host trees, and was affected by resource limitation. Specifically, defoliation-caused tree damage would result in increased losses of L2, which were exacerbated by maternal fecundity, infection of a pathogen, or weather-related effects. They also indicated that survival of early-stage budworm larvae in persistent outbreaks declined and the likelihood of other density-related factors such as the rate of mortality from natural enemies increased.
Application of the study of spatial variability of mortality
The understanding of spatial variability of mortality is essential to understand the stand vulnerability to spruce budworm disturbance.
References
Batzer, H.O., 1969. Forest character and vulnerability of balsam fir to spruce budworm in Minnesota. Forest Science, 15(1), pp.17-25.
Baskerville, G.L. and MacLean, D.A. 1979. Budworm-caused mortality and 20-year recovery in immature balsam fir stands. Can. For. Serv., Fredericton, NB, Canada. Inf. Rep. M-X-102.
Beckwith, R.C. and Burnell, D.G., 1982. Spring larval dispersal of the western spruce budworm (Lepidoptera: Tortricidae) in north-central Washington. Environmental Entomology, 11(4), pp.828-832.
Beckwith, R.C. and Kemp, W.P., 1984. Shoot growth models for Douglas-fir and grand fir. Forest science, 30(3), pp.743-746.
Blais, J.R., 1952. The relationship of the spruce budworm (Choristoneura fumiferana, Clem.) to the flowering condition of balsam fir (Abies balsamea (L.) Mill.). Canadian Journal of Zoology, 30(1), pp.1-29.
Blais, J.R. 1985. The ecology of the Eastern Spruce Budworm: a review and discussion. In Recent Advances in Spruce Budworms Research: Proceedings of the CANUSA Spruce Budworms Research Symposium, Bangor, Maine, September 16-20, 1984. CANUSA: Canadian Forestry Service, Ottawa, Ont. pp. 90–91.
Bouchard, M. and Pothier, D., 2010. Spatiotemporal variability in tree and stand mortality caused by spruce budworm outbreaks in eastern Quebec. Canadian Journal of Forest Research, 40(1), pp.86-94.
Candau, R., Belli, A., Millet, G.Y., Georges, D., Barbier, B. and Rouillon, J.D., 1998. Energy cost and running mechanics during a treadmill run to voluntary exhaustion in humans. European journal of applied physiology and occupational physiology, 77(6), pp.479-485.
Candau, J.-N. and Fleming, R.A. 2005. Landscape-scale spatial distribution of spruce budworm defoliation in relation to bioclimatic conditions. Can. J. For. Res. 35(9): 2218-2232.
Craighead, F.C., 1925. Bark-beetle epidemics and rainfall deficiency.
Gray, D.R. and MacKinnon, W.E. 2007. Historical spruce budworm defoliation records adjusted for insecticide protection in New Brunswick, 1965–1992. J. Acad. Entomol. Soc. 3: 1-6.
Gray, D.R. 2013. The influence of forest composition and climate on outbreak characteristics of the spruce budworm in eastern Canada. Can. J. For. Res. 43(12): 1181-1195.
Hardy, Y.J., Lafond, A., and Hamel, L. 1983. The epidemiology of the current spruce budworm outbreak in Québec. For. Sci. 29(4): 715-725.
Hardy, Y., Mainville, M., and Schmitt, D.M. 1986. Spruce budworms handbook: an atlas of spruce budworm defoliation in eastern North America, 1938-1980. U.S. Department of Agriculture, Forest Service, Cooperative State Research Service, Washington, DC, USA.
MacLean, DA and Ostaff, DP, 1989. Patterns of balsam fir mortality caused by an uncontrolled spruce budworm outbreak. Canadian Journal of Forest Research, 19 (9), pp. 1087-1095.
Morris, R.F., 1955. The development of sampling techniques for forest insect defoliators, with particular reference to the spruce budworm. Canadian Journal of Zoology, 33(4), pp.225-294.
Morris, R.F. and Mott, D.G., 1963. Dispersal and the spruce budworm. The Memoirs of the Entomological Society of Canada, 95(S31), pp.180-189.
Moody, B.H. and Otvos, I.S., 1980. Distribution of hibernating spruce budworm larvae within crowns of balsam fir trees in Newfoundland (p. 21). Newfoundland Forest Research Centre.
Peltonen, M., Liebhold, A.M., Bjørnstad, O.N., and Williams, D.W. 2002. Spatial synchrony in forest insect outbreaks: roles of regional stochasticity and dispersal. Ecology 83(11): 3120-3129.
Régnière, J. and Fletcher, R.M., 1983. Direct measurement of spruce budworm (Lepidoptera: Tortricidae) larval dispersal in forest stands. Environmental entomology, 12(5), pp.1532-1538.
Régnière, J., Nealis, V. and Porter, K., 2008. Climate suitability and management of the gypsy moth invasion into Canada. In Ecological Impacts of Non-Native Invertebrates and Fungi on Terrestrial Ecosystems (pp. 135-148). Springer, Dordrecht.
Royama, T. 1984. Population dynamics of the spruce budworm Choristoneura fumiferana. Ecol. Monogr. 54(4): 429-462.
Turner, E.G., 1952. Roman Oxyrhynchus. The Journal of Egyptian Archaeology, 38(1), pp.78-93.
Williams, D.W. and Liebhold, A.M. 2000. Spatial synchrony of spruce budworm outbreaks in eastern North America. Ecology 81(10): 2753-2766.
Zhao, K., MacLean, D.A., and Hennigar, C.R. 2014. Spatial variability of spruce budworm defoliation at different scales. For. Ecol. Manage. 328: 10-19.