Integrated landscape genetics of the mountain pine beetle system
The mountain pine beetle (MPB; Dendroctonus ponderasae) outbreak in western Canada has had significant ecological and economic consequences. Outbreak dynamics are the product of interactions among the beetle itself, associated pathogenic fungi (Ophiostomaspp.), host trees (Pinus spp.), physical landscape features, and climate. As part of the TRIA project, a large interdisciplinary NSERC Strategic Network, this research uses genetic data to investigate the spatial interactions among species in the beetle-fungi-pine complex and relate them to landscape features to better understand the causes and consequences of outbreaks and to inform risk management models.
We are also interested in the question of whether rapid evolution has played a role in the expansion of the MPB into novel habitats. The first step to address such a question is to identify loci that are potentially under selection. However, demographic processes such as rapid range expansion can create spatial patterns in allele frequencies that mimic those created by adaptive processes. Landscape genomics endeavours to identify loci potentially under selection through correlations between allele frequencies and environmental heterogeneity but do not necessarily consider the influence of this potentially confounding demographic context. Using a simulation approach, we are examining how range expansion, population dynamics, and spatial context affect our capacity to meaningfully identify loci under selection.
Scanning electron microscopy image of an adult mountain pine beetle, Dendroctonus ponderosae. Photo credit: Janice Cooke
Parasitoid meta-community dynamics
Theory and empirical data indicate that forest structure (i.e., composition and configuration) can affect insect population dynamics and thus has the potential to modulate the severity and extent of forest insect outbreaks. Forest structure affects population irruptions directly through the availability and connectivity of suitable host trees. However, the forest can also indirectly affect outbreaks through its effects on predator (parasitoid) community diversity, abundances, and mobility. This project combines tools and techniques from community ecology and landscape genetics to evaluate the relative contributions of direct and indirect factors on insect trophic dynamics and outbreaks. Specifically, we are evaluating the relative importance of the “silvicultural” and “natural enemies” hypothesis of spruce budworm (SBW) outbreaks.
To address this idea, SBW larvae were collected in the summer of 2014 from 30 sites across Quebec in regions currently affected by the SBW. Larvae were reared and emerged parasitoids were identified. In total, or 50 000 larvae were individually reared which provided over 7000 parasitoids of over 15 species of hymenoptera and diptera.
Current projects using these data include:
1) Assessing how forest structure affects spatial variation in parasitism rates at different spatial scales using remotely sensed data.
2) Determining the relative importance of spatial context, environmental variation, and outbreak characteristics of parasitoid community beta-diversity at broad spatial scales.
3) Investigating how climate driven phenological mismatches between SBW and parasitoids may affect future outbreak dynamics in northern forests.
4) Characterizing the spatial genetic structure and estimating dispersal capacity of two parasitoid species: Apanteles fumiferanae & Glypta fumiferanae.
5) Assessing the context-dependent drivers of community assembly in the SBW parasitoid complex.
We hope that our results will contribute to our understanding of the diversity-stability relationships in forest insects systems and to provide recommendations for spatial management strategies that will allow us to augment the control that parasitoid populations exert on outbreaking populations and thus to reduce forest losses.
Students: Simon Legault
Marrec R, Pontbriand-Paré O, Legault S, and James PMA. 2018. Spatiotemporal variation in drivers of parasitoid metacommunity structure in continuous forest landscapes. Ecosphere 9(1):e02075. doi: 10.1002/ecs2.2075
Photo credit: Simon Legault
Photo credit: Simon Legault
Using spatial genetics to characterize spruce budworm dispersal
The spruce budworm (SBW; Choristoneura fumiferana) is a lepidopteran forest pest that devastates huge areas of spruce and fir forest during its periodic outbreaks. SBW outbreak dynamics are shaped by the complex interactions among climate, forest structure, communities of natural enemies, and dispersal. Despite the significance of movement to the spatial dynamics of SBW outbreaks, little is known about SBW dispersal, how it varies with spatial context and over course of an outbreak, and how it affects spatial synchrony in outbreak dynamics. This research applies tools and methods from spatial population genetics to characterize genetic connectivity among outbreak patches in the current outbreak in eastern north America. Using this information on genetic connectivity, and how it varies within and among years, we will infer patterns of gene flow and dispersal and how it varies as a function of intervening land-cover (isolation by resistance) and local environmental conditions (isolation by environment). Concurrently, we are developing dispersal models using predicted phenology (BioSIM) and demographic data collected from pheromone traps (adult males). Together, this work will increase our knowledge of how SBW movement varies in different forest and landscape contexts and will be used to improve simulation models that predict insect population dynamics and forecast future outbreak risk. This work will also address the important question of the potential efficacy of currently proposed early intervention strategies.
Check out our photo gallery illustrating our sampling network and developing results.
Students: Jeremy Larroque, Simon Legault
Collaborators: Rob Johns (CFS, AFC)
SBW larva hanging by a thread. Photo: Olivier Pontbriand-Paré
Spatial genetic structure is influenced by ecological context. Genetic structure in cyclic and irruptive populations will depend on the timing of data collection (peak or trough), context-dependent dispersal in the species, as well as the amplitude and frequency of population oscillations. See James et al (2014). for more details.
Outbreaks of forest insects are a significant agent of disturbance in Canada’s boreal and mixed-wood forests that affect forest landscape structure, including the accumulation of combustible fuels. As a result of repeated defoliation over consecutive years, defoliation by the spruce budworm (SBW; Choristoneura fumiferana) creates large patches of dead fir or spruce that have the potential to affect fire activity. Although it is generally believed that forest insects affect fire activity, how they have such an influence, and how this affect varies through time, remains equivocal. Expected northward expansion of multiple species of forest insect pest in combination with forecast climate-related increases in forest fire activity in the boreal forest means that there a a great deal of uncertainty regarding future disturbance interactions, their effects on forest composition and connectivity, and consequent ecosystem service provisioning.
In this research, we are seeking to better understand how historical defoliation by the spruce budworm affects fire risk including the probability of ignition, the probability of escape from initial control, and final fire size, in combination with multiple spatial and temporal covariates (e.g., weather and climate).
Recently, we completed a project modelling the relationship between historical fire ignitions and defoliation in Ontario using a a series of generalized additive logistic regression models. Using these models we contrasted fire-defoliation relationships between spring and summer fire seasons, as well as between ecoregions in eastern and western Ontario. We found that, in general, spruce budworm activity increases the risk of ignition 8-10 years after defoliation occurred, but decreases this risk immediately following defoliation (< 1 year).
The long term goal of this research is to produce predictive spatial models that can be used in combination with forecasted future climate and fire weather to predict changes in fire risk in repsonse to changing forest landscapes and insect-induced changes to fuel sturcture and connectivity.
James PMA, Robert, LE, Wotton BM, Martel D, Fleming RA. 2016. Lagged cumulative spruce budworm defoliation affects the risk of fire ignition in Ontario, Canada. Ecological Applications. 27(2): 532-544
Photo credit: Terry Chapin. http://www.lternet.edu/node/49513.
Forest ecosystem modelling
Researchers are often faced with questions at spatial and temporal scales that exceed our manipulative capacities. For example, understanding the long term consequences of climate change or different forest management regimes on forest dynamics requires data at spatial and temporal scales much larger than are usually feasible in traditional ecological studies. Spatially explicit simulation modelling can be used to address these questions, with the caveat that models are not reality, but instead represent a simplified version of reality containing the main processes in which we are interested. In this way, simulation models should be considered as a way for us to observe the consequences of our assumptions about how systems work even when direct manipulation and observation may not be possible.
This research theme focuses on the long term consequences of forest management on landscape composition and configuration and how these spatial patterns affect habitat availability, landscape connectivity, and biodiversity conservation. Using spatially explicit simulation models and the SELES simulation platform, we examine how interactions between the spatial legacies created by forest management policies affect other ecological processes such as climate, fire and insect disturbance, forest succession, as well as community and population dynamics.
Much of the other ongoing research in the lab will contribute new parameters, sub-models, and conceptual frameworks to current models of forest ecosystem dynamics and will improve their capacity to evaluate the sustainability of current land-management practices.
Irruptive population dynamics are affected by both bottom-up and top-down processes. However, trophic chains such as this do not exist in isolation. Indeed, the food webs in which irruptive taxa are nested include multiple connections with and between trophic levels. We are investigating how variation in the amount of available host affects irruptive population dynamics when assuming that a reduction in the amount of host (e.g., confiers) increases the availability non-host (e.g., deciduous species) available for other herbivores, which in turn serves as a reservoir for shared natural enemies. This work is motivated by the spruce budworm outbreak system in which forest diversity has been shown to reduce the intensity of defoliation, presumabley thorugh an influence on a diverse community of associated generalist parasitoids. Using mathematical models (ODEs) to formalize this conceptual model, we are specifically investigating under what conditions a manipulation of in host and non-host density can modulate the intensity of defoliator outbreaks.
Collaborators: V. Krivan (CAS), É. Filotas (TELUQ)