Disturbance Interactions During Ecological Change chaired by Doug Shinneman and Brian Palik

Forest Insect Defoliation and Carbon Dynamics: Simulating Multiple Defoliator Species in Shared Landscapes With Landis-II

J. Foster et al. — Tue, 23 Jun 2009 08:00:00 PDT

Defoliation outbreaks are dynamic forest disturbances with unique spatial and temporal characteristics that produce distinct changes in forest composition and carbon balance. We simulated defoliation outbreaks using a new module for the forest disturbance and succession model, Landis-II, to better understand the long-term consequences of defoliation on forest carbon. This new module recreates the spatial dynamics of defoliation outbreaks by stochastically drawing parameters that describe spatial pattern from empirical distributions derived from Landsat defoliation maps. The module also captures species specific growth and mortality responses to accumulated defoliation stress. We demonstrate how these simulated defoliation events mimic spatial and temporal patterns of gypsy moth (GM, Lymantria dispar L.) defoliation outbreaks and their effects in the central Appalachian mountains of western Maryland, U.S.A. We simulated aboveground carbon dynamics over 400 years with and without GM defoliation in this mixed deciduous landscape. These simulations facilitated estimation of (1) the impacts of a generalist defoliator on long-term changes in forest composition and carbon storage, and (2) comparison of aboveground carbon dynamics expected in the absence of GM with those following introduction. Simulations were also run with forest tent caterpillar (Malacosoma disstria Hbn.) defoliation, individually and with GM, to examine how multiple defoliators with shared hosts alter long-term aboveground carbon dynamics. Preliminary results show that the introduction of GM disturbance changes the trajectory of forest species composition, facilitating increases in non-host species that would not otherwise occur. Forest carbon storage is temporarily reduced following individual outbreaks, as are long-term means, once GM enters the landscape. Changes in forest carbon storage are even more pronounced when a native defoliator has periodic outbreaks in the same landscape. The results directly illustrate how temperate forest carbon cycles, particularly aboveground carbon pools, are affected by interacting insect disturbances that are a fundamental, but changing part of forest ecosystems.

Forest Ecosystem Dynamics in a Non-Linear World

Sybille Haeussler — Tue, 23 Jun 2009 08:20:00 PDT

Forest ecosystems across North America are under increasing stress from the accelerating pace of global change which involves simultaneous changes in resource availability (temperature, moisture, nutrients), disturbance regimes (fire, insects, diseases, extreme weather, logging, urbanization) and (3) species distributions (invasive organisms, threatened species). Interactions among the agents of global change can generate emergent or unexpected ecosystem behaviour. Complex systems science provides a strong theoretical foundation for understanding these factor interactions and provides many new mathematical and simulation modeling tools that can generate complex, non-linear behaviour and provide improved understanding of ecosystem response to global change. I present an updated version of Jenny’s (1941) classic state factor model of soils and ecosystems that allows three variables (resources, disturbance and plant-soil functional groups) to interact to generate higher orders of complexity through self-organizing plant-soil feedback switches. An interactive agent-based model allows the user to vary the degree of change in resource availability, disturbance frequency or severity and the strength of negative and positive plant-soil feedbacks and to measure the effect of these changes on ecosystem resilience, diversity and landscape complexity. The model is currently being used to simulate the dynamics of lodgepole pine-Cladina-lichen dominated landscapes in central British Columbia that has been massively affected by interactions among changing fire regimes, mountain pine beetle and a warming climate.

Scale Specific Management Legacies in Patterns of Spruce Budworm Host Species in the Border Lakes Region

P. James et al. — Tue, 23 Jun 2009 08:40:00 PDT

The spruce budworm (SBW) is a native defoliator of fir and spruce forests in Canada. Increases in the extent and severity of SBW outbreaks over the last century may be due to changes in forest composition and configuration as a result of forest management. Better understanding of long term interactions between human and natural disturbance processes is essential to sustainable forest management. Using the Border Lakes Landscape (BLL) of northern Minnesota and north-western Ontario as a case study, we investigate whether differences in spatial patterns in SBW host species can indeed be attributed to regional differences in forest management. The BLL contains three zones of historical management: (1) fine scale management in Minnesota, (2) coarse scale management in Ontario, and (3) no management in a wilderness region comprised of the Boundary Waters Canoe Area and Quetico Provincial Park. Multi-temporal LANDSAT data representing species composition and basal area were analyzed using spatial pattern metrics and wavelets to better understand the role of management legacies on host species spatial structure. Results indicate that forest composition and host species basal area are finely structured over the entire BLL and strongly constrained by patterns of lakes and wetlands. Patch metrics indicate larger patches and a larger proportion of mixed wood forest in the unmanaged region between the actively managed regions of Ontario and Minnesota and smaller patches of deciduous forest in the wilderness and Ontario regions. Wavelet analysis revealed fine scale spatial structure in host species across the study region. Some support was also found for the existence of locally significant regions of host species basal area that correspond to political boundaries in the BLL demonstrating the potential influence of forest management on insect disturbance dynamics.

Effects of Forest Management Legacies on Spruce Budworm Outbreaks

L-E. Robert et al. — Tue, 23 Jun 2009 09:00:00 PDT

It is postulated that landscape structure changes connectivity of insect populations by affecting movement and dispersal. Spruce budworm outbreaks that have become more severe than in the past may be due in part to the effects of forest management on landscape structure, although this hypothesis remains to be tested. The main objective of this study was to evaluate the effects of landscape structure resulting from different types of management on spruce budworm outbreaks under similar biophysical conditions. The Border Lakes Landscape, with its contrasting management legacies on each side of the Canada-U.S border, provides an opportunity to study the effect of landscape structure on outbreak dynamics. We compared spruce budworm outbreak characteristics between forests managed at a fine scale (e.g., 10 hectare harvests) in Minnesota, USA, forests managed at a coarse scale (e.g., 100 hectare harvests) in Ontario, Canada, and a conservation zone overlapping the US-Canadian Border (i.e., Boundary Waters Canoe Area Wilderness and Quetico Provincial Park). We sampled sites to minimize the effect of other confounding factors such as climate and host abundance that also have an effect on spruce budworm cycle. Dendrochronological reconstructions of outbreaks in each zone allowed the detection of differences in the periodicity and synchrony of outbreaks. In comparison to the conservation zone where periodicity varied between 15 to 40 years, outbreaks were more frequent within the fine-scale managed zone (6-12 years) in the United States. Managed Canadian forests had an intermediate periodicity of 13 to 30 years. Results are consistent with the hypothesis that fragmentation and reduction of host abundance reduced the synchronization of outbreaks in the fine-scale zone thus explaining the difference in periodicity.

Simulated Effects of Wind and Harvesting on Detritus, Soil Organic Carbon, and Species Composition in Northern Wisconsin

R. Scheller et al. — Tue, 23 Jun 2009 09:20:00 PDT

Total forest carbon (C) storage is determined by forest succession, multiple interacting disturbances, climate and the edaphic properties of a site or region, including soil texture and depth. How these complex processes interact will determine forest carbon dynamics at landscape and regional scales. We have developed a new succession extension for the LANDIS-II forest landscape simulation model that incorporates the belowground soil C dynamics of the Century soil model. This extension simulates three primary soil organic matter (SOM) pools (fast, slow, passive), litter dynamics, and nitrogen (N) feedbacks to overstory production. The extension was validated against data from the Willow Creek experimental forest in Wisconsin, USA. We subsequently initialized the full model to simulate forest dynamics of 9800 ha of the surrounding forest landscape with harvesting and wind, two common disturbances in mesic forests of the Lake States. We simulated three management scenarios: no harvesting, standard harvesting with slash left in place, and whole tree harvesting where all slash is removed. A historic wind throw regime (50 year wind rotation period, including light, moderate, and extreme events) was included in all scenarios. We estimated forest compositional change and C storage. Our results indicate a strong feedback from harvesting to litter C and the slow SOM pools. The fast and passive SOM pool was not significantly altered. Overall, the simulations with no harvesting had substantially greater total C and continued to sequester C. Standard harvest simulations had more C than the whole tree harvest simulations, although the differences were slight. Under both harvest regimes, C accrual was not evident after 150 years. In conclusion, we have developed a robust model of C and N cycling that leverages an existing forest modeling framework to simulate landscape and regional scale interactions among forest disturbances, climate change, and soil processes.

Can Landscape-Scale Management Influence Insect Outbreak Dynamics? A Natural Experiment For Eastern Spruce Budworm

B. Sturtevant et al. — Tue, 23 Jun 2009 09:40:00 PDT

The balance of evidence suggests forest insect outbreaks today are more damaging than ever because of changes in forest composition and structure induced by fire suppression and post-harvest proliferation of tree species intolerant to herbivory. We hypothesize that landscape connectivity of acceptable host trees increases defoliator population connectivity, altering the dynamics and spatial structure of defoliator populations, and thus increasing forest susceptibility to insect pest damage. We evaluate this hypothesis for eastern spruce budworm (Choristoneura fumiferana; Clemens) within a 2 million ha “experimental” landscape at the international border between Minnesota and Ontario, containing wilderness plus two contrasting patterns of harvesting (coarse vs. fine). Wavelet analysis of forest composition maps produced by remote sensing revealed fine spatial structure and offered some support to the existence of separate zones of host species spatial structure corresponding with political boundaries of the study landscape. Yet current spatial structure in host trees was complicated by abundant lakes and as well as overlapping disturbances including harvest, fire, budworm, and wind. Initial application of AFLPs from spruce budworm larvae found geospatial genetic structure within the study landscape, where populations within the northern region appear more connected than within the southern region. Tree ring chronologies show a clear 32-year outbreak cycle in the undisturbed wilderness region, a poorly synchronized and lower intensity 15-year cycle in fragmented Minnesota, and a heterogeneous mix of the two signals in coarsely fragmented Ontario. Collectively, our results are consistent with the hypothesis that fragmentation and reduction of host abundance reduced the synchrony of outbreaks in the fine-scale zone thus explaining the difference in population connectivity as well as the periodicity and intensity of outbreaks. Our study is among the first to show that forest management can not only influence insect damage associated with budworm outbreaks, but also the nature of the outbreaks themselves.

Legacy of Insect Defoliators: Increased Wind-Related Mortality Two Decades After a Spruce Budworm Outbreak

S. L. Taylor et al. — Tue, 23 Jun 2009 10:30:00 PDT

The effects of spruce budworm (Choristoneura fumiferana (Clem.)) outbreaks on growth and survival of balsam fir (Abies balsamea (L.) Mill.) and spruce (Picea spp.) are well documented, but few studies extend beyond 10 years after defoliation ceased. We used inventory data from 106 permanent sample plots (PSP) in >50 year old balsam fir stands in northern New Brunswick, Canada, to determine legacy effects of the 1969-1993 budworm outbreak on stand development up to 29 years after defoliation ceased. Defoliation data were based on annual aerial surveys from 1945 to 1993 and PSP ground sampling from 1985 to 1993. PSPs were stratified into net stand volume development categories (decreasing, stable, and increasing stemwood volume from 1985-2005), and related to outbreak phases (outbreak, direct 1-10 years after defoliation ceased, and legacy >10 years), number of years of defoliation (1-4, 5-8, 9-12), and stand age (mature, overmature). Trend and rate of volume development over time was related to past outbreak severity (?2 = 6.681, df = 2, p<0.05). For example, 93% of 15 PSPs that had entered the stand break-up phase (i.e., <100 m3/ha live standing volume) experienced ?5 years of defoliation. Stand age was an important factor influencing outbreak severity (e.g., r2 = 0.383, p<0.01) and vulnerability to post-outbreak wind disturbance events. A pulse of post-outbreak wind-related mortality peaked at 11 m3/ha/yr 11-15 years after defoliation ceased. Hence, the compounding effect of post-outbreak wind disturbance drastically increased the impact and longevity of the spruce budworm disturbance event, especially in ageing post-outbreak stands. Our study demonstrates the need to consider the complexity of disturbance dynamics, as a single disturbance event rarely operates in isolation. Stand development projections and wood supply models need to take into account losses that result from the interaction of disturbances events to ensure the security of future wood supplies.