Forest Investment Account (FIA) - Forest Science Program
FIA Project Y091028

    Modeling phenology and outbreaks of the western spruce budworm
 
Project lead: Nealis, V.G. (Canadian Forest Service)
Author: Nealis, Vince G.
Subject: Forest Investment Account (FIA), British Columbia
Series: Forest Investment Account (FIA) - Forest Science Program
Description:
The western spruce budworm, Choristoneura occidentalis, is the most destructive insect on Douglas-fir in BC. Damaging outbreaks have become chronic in the IDF zone, especially in the southern interior, since the 1970s (Maclauchlan et al. 2006). More than 500 000 ha have been defoliated every year since 2004. The striking characteristic of these recent outbreaks is their extension over the entire ecological range of Douglas-fir in the central and southern interior of the province. This represents a dramatic northward extension of damage into the Cariboo region as well as upward in elevation to mixed Douglas-fir/spruce/lodgepole pine forests and is unprecedented in historical times. One indicator of this concern to forestry was the largest Btk spray program ever conducted for this insect in BC in 2006. (https://www.for.gov.bc.ca/rsi/ForestHelath/Index.html).

A strong influence of weather and climate on western spruce budworm outbreaks has long been suspected. Trees in dry, stressed areas seem to be at greatest hazard to western spruce budworm (Maclaughlan et al. 2006) and the observation of defoliation in specific and shifting elevation bands across the landscape is common. Early, survey-based models invoked weather-driven variation in the synchrony of insect emergence and bud-burst as a driving factor for these outbreak patterns (Thomson et al. 1984). A recent application of this early model suggests that the absence of outbreaks in coastal Douglas-fir since the 1930s is related to a warming trend in the coastal climate which has desynchronized the critical relationship between budworm feeding and bud-flushing dates of coastal Douglas-fir (Thomson and Benton 2008). Similarly, an improvement in synchrony associated with current climate conditions in the BC interior may be contributing to the exceptional severity, duration, and extent of the outbreaks there. If so, not only has the hazard profile for western spruce budworm changed over an extensive area, but many of the forests at risk have little or no historical signal of previous losses in their growth profiles. Today’s depletions are new, and perhaps unaccounted, reductions in expected volume projections. As future timber supply in the wake of the mountain pine beetle will rely more heavily on non-pine species, increased protection from spruce budworm may be required for interior Douglas-fir not only in mature but in juvenile stands. Non-timber values are also at risk from increased budworm disturbance as Douglas-fir plays an important role in hydrology and wildlife ecology in dry, interior forest and range ecosystems. Assessment of these potential changes now will provide more management options for mitigation and adaptation.

This project will develop a seasonal, process-oriented phenology model for western spruce budworm that can be applied to questions of current and future risk of BC’s Douglas-fir forests to damaging outbreaks, and for aiding operational decisions around control. Process models describe data using key mechanisms or processes that determine the intrinsic structure and behavior of a system. By comparison, empirical models, often used in forestry, for example, to estimate hazard, emphasize statistical relationships with more emphasis on descriptive than causative structure. A process model for phenology will use parameters which describe eco-physiological relationships between weather variables and insect development and survival to analyze and predict target events such as seasonal occurrence of life stages and rates of change in population densities for any given location. Because the parameters of the process model are general, they remain relevant for new conditions, whether these be the need to predict target events in different locations or times, or in different weather regimes likely under climate change. Process models are thus flexible tools of great use at multiple spatial and temporal scales which characterize the information challenges of modern resource management. They can be used to examine seasonal occurrence of life stages in specific locations in any one year or project changes in hazard over landscapes for several years Examples of BC pest management programs currently using process-based phenology models include gypsy moth eradication (Nealis et al. 2001) and history and spread of mountain pine beetle (e.g. Taylor et al. 2006).

A comprehensive and credible process-based phenology model requires insight into key mechanisms influencing the seasonal ecology of insects, sufficient data to adequately estimate parameters, and independent field data to calibrate the model. Fortunately each of these basic elements is relatively well-developed for the western spruce budworm system. A research program at the Pacific Forestry Centre has been tracking the population ecology of western spruce budworm in the southern interior since 1997. This program has compiled profiles of budworm seasonal development and changes in population density in several sites for many years. Available laboratory measurements of development rates at variable temperatures can provide preliminary estimates of rate equations (see Fig. 1 in downloads). Analogous experiments in the lab and field are underway to examine winter mortality. While an initial model may reveal the need for additional measurements to achieve sufficient refinement for the broadest possible applications, the current data are certainly adequate for defining the structure of the model, initializing parameter values and producing realistic output which can be compared to patterns in the field. Historic survey information and published degree-day models for Douglas-fir will also be used to incorporate tree development and thus enable coupled seasonal development of the insect and its tree host. Recent history of the expansion of outbreak levels of western spruce budworm in BC is available from the BC Ministry of Forests and Range.

Another necessity for process-based phenology models is a modeling environment that enables rapid input of spatio-temporal information, specification of species-specific models, integration of these models with weather data, and production outputs in terms of desirable target events in different formats of use to decision-makers. Such a modeling environment, BioSIM, has been developed by J. Régnière of the Canadian Forest Service (Régnière 1996). It has been successfully applied to a variety of pest management issues worldwide and is used extensively to examine risk and guide management decisions regarding the gypsy moth and mountain pine beetle in western North America. Since the early development of BioSIM was driven largely by interest in management decisions regarding eastern spruce budworm (Cooke and Régnière 1996), BioSIM is well-suited for adaptation to the western spruce budworm system. The phenology model for western spruce budworm proposed here will apply the conceptual approach already used for the eastern spruce budworm model but will estimate the necessary, independent parameters from original data collected on the western spruce budworm in BC and include available information on tree phenology as well (Nealis and Nault 2005). BioSIM will integrate the statistical models describing temperature-driven seasonality of the insect and the tree with spatially-explicit climatic information from BC. The model can then be used to examine the associations between recent geographic expansion of the outbreaks and climate change and enable investigative scenarios of future states. Direct application of model output to assist in risk assessment, mapping current and future hazard and guide protection programs will also be possible.

    Deliverables:

Final report (0.3Mb)
Western spruce budworm-BioSIM model (page 8 in Information Forestry newsletter, August 2008) (0.8Mb)

To view PDF documents you need Adobe Acrobat Reader, available free from the Adobe Web Site.

Updated August 16, 2010 

Search for other  FIA reports or other Ministry of Forests and Range publications.

Please direct questions or comments regarding publications to For.Prodres@gov.bc.ca