Linked disturbance in the temperate forest: Earthworms, deer, and canopy gaps.

Despite the large body of theory concerning multiple disturbances, relatively few attempts have been made to test the theoretical assumptions of how and if disturbances interact. Of particular importance is whether disturbance events are linked, as this can influence the probability and intensity of ecological change. Disturbances are linked when one disturbance event increases or decreases the likelihood or extent of another. To this end, we used two long-term, multi-disturbance experiments in northern Wisconsin to determine whether earthworm invasion is linked to canopy gap creation and white-tailed deer browsing. These three disturbances are common and influential within North American temperate forests, making any interactions among them particularly important to understand. We expected both deer and canopy gaps to favor invasive earthworms, particularly species that live close to or on the soil surface. However, we found only partial support of our hypotheses, as both deer exclosures and canopy gaps decreased earthworms in each experiment. Further, earthworm density increased the most over time in areas far from the gap center and in areas with deer present. Deer exclosures primarily decreased Aporrectodea and Lumbricus species, while gaps decreased Dendrobaena and Lumbricus species. Our findings show that earthworm invasion is linked to deer presence and gap-creating disturbances, which provides new insight in multiple disturbance theory, aboveground-belowground dynamics, and temperate forest management.

Multimodel simulations of forest harvesting effects on long­term productivity and CN cycling in aspen forests.

The effects of forest management on soil carbon (C) and nitrogen (N) dynamics vary by harvest type and species. We simulated long-term effects of bole-only harvesting of aspen (Populus tremuloides) on stand productivity and interaction of CN cycles with a multiple model approach. Five models, Biome-BGC, CENTURY, FORECAST, LANDIS-II with Century-based soil dynamics, and PnET-CN, were run for 350 yr with seven harvesting events on nutrient-poor, sandy soils representing northwestern Wisconsin, United States. Twenty CN state and flux variables were summarized from the models' outputs and statistically analyzed using ordination and variance analysis methods. The multiple models' averages suggest that bole-only harvest would not significantly affect long-term site productivity of aspen, though declines in soil organic matter and soil N were significant. Along with direct N removal by harvesting, extensive leaching after harvesting before canopy closure was another major cause of N depletion. These five models were notably different in output values of the 20 variables examined, although there were some similarities for certain variables. PnET-CN produced unique results for every variable, and CENTURY showed fewer outliers and similar temporal patterns to the mean of all models. In general, we demonstrated that when there are no site-specific data for fine-scale calibration and evaluation of a single model, the multiple model approach may be a more robust approach for long-term simulations. In addition, multimodeling may also improve the calibration and evaluation of an individual model.