Proponent Information

Name: Brian H. Aukema or Matthew D. Klingenberg Organization: Canadian Forest Service, Natural Resources Canada Mailing Address: ℅ University of Northern British Columbia Enhanced Forestry Laboratory 11-123 3333 University Way Prince George, BC V2N 4Z9 Email Address: [email protected] or [email protected] Telephone: (250) 960-5924 or 960-6337 Fax: (250) 960-6498

Project Information

Project Number: M075046 Project Title: Stand and forest dynamics following MPB: how spatial patterns of salvage harvesting affect Warren root collar weevil pressure in regenerating stands

Project Purpose / Management Implications:

Project Description: The outbreak of mountain pine beetle in BC currently covers 9.2 million ha, although the majority of the outbreak in the northern interior of the province has run its course (Aukema et al. 2006). Salvage operations will continue at unprecedented rates throughout the next decade, transforming the landscape into a mosaic of numerous patches of regenerating stands adjacent to killed timber. Despite widespread salvage operations, little is known about how the spatial patterns of salvage logging will affect recruitment and success of both planted and naturally regenerating stands. Large-scale disturbance events such as insect outbreaks frequently affect plant and animal species assemblages in successional communities (Matsuoka et al. 2001). In the current context of mountain pine beetle disturbance, a significant threat to regeneration is the Warren root collar weevil (Hylobius warreni) both concentrating in and migrating from residual stands. The Warren root collar weevil is prevalent in lodgepole pine and mixed lodgepole pine/spruce stands throughout BC. Larvae develop and feed subcortically on the root collars of trees. Large diameter trees typically tolerate feeding, although the feeding wounds may predispose surviving trees to root pathogens. Larvae frequently girdle and kill small diameter trees, and cumulative mortality in past studies has ranged as high as 16% (Schroff et al. 2006). Adults may live up to five years, and females lay 25-30 eggs per year. Of these, approximately 25% typically survive to the damaging larval stages (Cerezke 1994). Several site conditions favour Warren weevil development, and many of these may be created by timber salvage operations following MPB. First, an increased duff layer, such as from logging slash, may create conditions favourable for weevil proliferation. For example, a study by Czereke (1994) noted that the number of weevils per tree increased up to 5× after a pre-commercial thinning treatment, and the proportion of trees sustaining injury doubled from 25% to 50%. Second, open site conditions, such as within a clearcut, allow increased soil temperatures amenable to larval development. Third, planted lodgepole pine often has significant root deformation (Robert and Lindgren 2006), and such trees may suffer increased mortality due to weevil attack (Robert 2004). Finally, weevils are most abundant on moist to wet sites, which may become more prevalent as water tables rise due to extensive lodgepole pine mortality (Cerezke 1994). Salvage logging of MPB-killed stands might act to decrease endemic resident weevil populations, but in fact, the opposite may be true. MPB-caused mortality would essentially remove dead trees from the host pool, and weevils would typically concentrate on the remaining live stems. After salvage harvesting, larvae will continue to develop in residual stumps (Cerezke 1973), while adults may migrate to nearby regenerating stands, such as within clearcuts, although the extent of the latter activity is unclear. Aerial surveys in the Nadina Forest District by Ken White and Carolyn Stevens (Project Partners with the BC Ministry of Forests and Range) indicate mortality of up to 40% in regeneration openings, frequently within 100m of MPB-killed stands. Quantifying this migration and subsequent mortality to subsequent regeneration in relation to spatial arrangement of MPB-killed stands is the focus of this research. This project will lead to an understanding of the spatial pattern of attack/mortality of Warren root collar weevil in regenerating stands adjacent to MPB-killed stands, and allow us to infer processes that may be driving such patterns. By determining spatial variability within and between regenerating stands, especially focusing on weevil distributions and patterns of injury and mortality within regeneration, effective weevil management strategies may be found early in stand-replacement phases for the northern British Columbia interior. Understanding process may inform future management strategies, such as non-host species selection for harvesting on high-risk sites, deployment of potential chemical or visual attractants and traps, and/or creation of buffer-habitat unsuitable for migration.

Project Duration: 1 Apr 2006 to 31 March 2008

Methodology Overview: We select between 6-10 regenerating cutblocks aged 4-7 years post stocking, adjacent to MPB-killed stands. We attempt to select relatively flat, uniform sites. Slope, elevation, aspect, block:edge ratio, age, and stocking density are measured for each site. Our hypothesis is that MPB causes loss of habitat, and thus leads to net movement of Warren root collar weevils into new regenerating stands. Weevil attack as quantified by injury to the root collar of trees and tree mortality are used as proxies for weevil numbers. In each stand, we establish 100m transects from the margin of the stand edge adjoining the MPB-killed stand, proceeding toward the centre of the regenerating block. Along the 8-m wide transect, we measure all trees: location, height, origin (planted/natural/unsure), condition (e.g., forest health problems such as stem rusts), and vitality (i.e., live or dead; dead trees will include those with yellow or red needles and lack of bud growth). Each 50m, we conduct a more intensive sampling of an 8m portion of the transect for weevil injury (i.e., feeding scars and circumference girdled) on every tree. These intensive subplot samplings involve careful examination of the root collar on each tree. To obtain estimates of prior year mortality, all dead trees in each transect are harvested. Examination of the growth rings are used to determine year of tree death for planted trees, because we have prior knowledge of tree age at planting, and years since planting for these trees. Comparisons between spatially referenced current and previous mortality provide estimates of weevil movement through time. Quantification of the spatial patterns of both trees and mortality within the transects is accomplished using appropriate point pattern analyses, such as Ripley’s univariate K(t) function or its linearized analog, L(t). Interactions between the two patterns are assessed using Ripley’s bivariate K(t)12 function. This provides information on spatial pattern of planting and attack (regular, random, or clustered distributions). Presence of an inhomogeneous point process (such as a gradient of mortality) is examined using linear regression of mortality as a function of distance. Potential changes in this relationship by accounting for ages of tree death (as determined by counting growth rings of each dead tree) sheds light into relative rates of migration along transect vs. random localized increases. Linear models of the intensively-sampled subplots against distance from MPB-stand margin (number of trees attacked/number trees present, number of trees killed/number of trees present, etc.) will be used to provide information on potential weevil ingress into plots. Between- cutblock comparisons will be conducted using the strip plots as subsamples in appropriately nested analytical designs.

Project Scope and Regional Applicability: Prince George, Nadina, Quesnel, and Vanderhoof Forest Districts.

Interim Results/Conclusions Graphical analysis has shown a gradient of mortality due to feeding by Warren root collar weevil exists in approximately 60% of stands examined in the first year of this project. We are currently working to determine mechanisms that may be responsible for the patterns found. Conclusions will follow year two of this project.

Contact Information: Dr. Brian Aukema, [email protected], 250-960-5924 Mr. Matthew Klingenberg, [email protected], 250-960-6337

References

Aukema, B.H., Carroll, A.L., Zhu, J. , Raffa, K.F., Sickley, T.A., and Taylor, S.W. (2006) Landscape level analysis of mountain pine beetle in British Columbia, Canada: spatiotemporal development and spatial synchrony within the present outbreak. Ecography. 29: 427-441.

Cerezke, H.F. (1973) Survival of the weevil, Hylobius warreni Wood, in lodgepole pine stumps. Canadian Journal of Forest Research. 3: 367-372.

Cerezke, H.F. (1994) Warren rootcollar weevil, Hylobius warreni Wood (Coleoptera, Cuculionidae), in Canada: ecology, behavior, damage relationships, and management. Canadian Entomologist. 126: 1383-1442.

Matsuoka, S.M., Handel, C.M., and Ruthrauff, D.R. (2001) Densities of breeding birds and changes in vegetation in an Alaskan boreal forest following a massive disturbance by spruce beetles. Canadian Journal of Zoology-Revue Canadienne De Zoologie. 79: 1678-1690.

Robert, J. A. (2004) Effects of root deformation on growth and susceptibility of lodgepole pine to red attack by the Warren root collar weevil. M.Sc Thesis, University of Northern British Columbia, Prince George, BC.

Robert, J.A. and Lindgren, B.S. (2006) Relationships between root form and growth, stability, and mortality in planted versus naturally regenerated lodgepole pine in north-central British Columbia. Canadian Journal of Forest Research-Revue Canadienne De Recherche Forestiere. 36: 2642-2653.

Schroff, A.Z., Lindgren, B.S., and Gillingham, M.P. (2006) Random acts of weevil: A spatial analysis of Hylobius warreni attack on Pinus contorta var. latifolia in the sub-boreal spruce zone of Northern British Columbia. Forest Ecology and Management. 227: 42-49.