My primary interests are in field ecology, particularly in forests and other terrestrial ecosystems.  My research to date has focused on the importance of forest structure at multiple spatial and temporal scales, ranging from individual trees to large forested landscapes.  I am interested in how fires and other disturbances (such as wind, insects, or human land use) affect the structure of forests via successional processes and how these processes in turn may influence ecosystem processes such as carbon cycling and storage and forest productivity.  Much of my work is aimed towards understanding how the combination of site factors, biotic interactions, natural disturbances, and humans affect landscape patterns in forests.  My research always contains a field component and often incorporates multivariate statistics, landscape pattern analyses, remote sensing and aerial photo interpretation, GIS, and simulation modeling.  My study sites include forests in the Rocky Mountains, Yellowstone National park, northern Lower Michigan, and southeastern Michigan.

At Wayne State, I am hoping to develop a research program that will examine the effects of exotic insects and plants on the structure and function of deciduous forests in southeastern Michigan and native insects in coniferous forests of the Rocky Mountains.  I am also interested in the effects of past and current land use on the occurrence of invasive species in Michigan forests.

Landscape-level carbon cycling in Yellowstone National Park

Collaborators:  Mike Ryan, (US Forest Service), Bill Romme (Colorado State University), Dan Tinker (University of Wyoming), Monica Turner (University of Wisconsin), Erica Smithwick (University of Wisconsin)

Many studies have described ecosystem processes in small study areas, but the basis for understanding and predicting their variability across larger spatial scales is not well developed.  Previous research in Yellowstone has documented tremendous landscape variability in tree density that was created by the large fires of 1988, as well as by smaller fires that occurred before and since the 1988 fires.  Such variability is tightly linked to variability in ecosystem processes such as productivity, carbon allocation, and carbon storage.  Our current research is determining how initial post-fire structural heterogeneity in Yellowstone controls carbon dynamics as stands mature, and how climate-mediated changes in fire regimes may alter the behavior of the Yellowstone landscape as a net sink or source of carbon in the global carbon cycle.  We will determine the current carbon balance for the Yellowstone landscape and the carbon lost in the 1988 fires by mapping the current distribution of forest age and stand density; using replicated chronosequences to measure how annual net carbon storage (NEP) varies with age and density; estimating carbon lost in 1988 via direct combustion; and extrapolating stocks and fluxes to the landscape.

Current conditions and long-term trends in aspen forests of the northern Colorado Front Range

Collaborators:  Bill Romme (Colorado State University), Claudia Regan (US Forest Service

Trembling aspen is one of the only deciduous forest types in the US Rocky Mountains, and is therefore a valued species for its contribution to biodiversity, wildlife habitat, and aesthetics.  Several investigators have documented a general decline and senecence of aspen stands across the West due to factors such as climate fluctuations, fire suppression and increased elk browse.  We examined aspen stands in Larimer County, Colorado to assess the current condition and to predict long-term trends in aspen forests across this heavily managed landscape.  We identified six types of trends in aspen, distinguished mainly by the relative dominance and age distribution of aspen and co-existing conifers where applicable.  Only 35% of stands exhibited age distributions that suggested a self-replacing stand; 63% were declining, often as a result of successional interaction with conifers.  Aspen stands found amongst lodgepole pine forests suggested a dynamic relationship with pines whose relative dominance depended on the time since fire.  Because most of these aspen stands occurred within lodgepole pine forests less susceptible to fire suppression, this type of aspen decline is likely a normal successional process not significantly altered by humans.  Because only 24% of aspen stands showed evidence of a true "decline", and many aspen stands in the County may have regenerated due to human activities, such a result suggests that a decrease from the current amount of aspen area may still fall within the historic range of variability of aspen coverage.  Stand dynamics and successional characteristics are therefore critical in interpreting trends in aspen decline at large spatial and temporal scales.

Landscape variability and convergence in forest structure following large fires in Yellowstone National Park

Collaborators:  Monica Turner (University of Wisconsin), Craig Lorimer (University of Wisconsin), Bill Romme (Colorado State University)

The 1988 fires in Yellowstone National Park created a very complex, fine-grained mosaic of lodgepole pine seedling densities across the burned landscape.  Several investigators have speculated that this "footprint" of the 1988 fires will persist until a similar disturbance again occurs, few studies have actually examined the length of its persistence.  I examined the portion of the Yellowstone landscape left unburned by the 1988 fires to estimate how postfire heterogeneity in stand structure might change with successional time, and to determine the rates and mechanisms of these changes.  We first mapped the post-1988 seedling densities using digital orthophotography.  Analyses of age distributions and stand reconstructions using dendrochronology suggested that initial patterns of stand structure may persist for up to two centuries.  However, stand structural variability is over time because initially dissimilar stand structures develop though multiple mechanisms that act to compress the initial variability towards a common stand structure.  Stem growth and leaf area are also affected by this trend of "convergence", since both are related to stand density.  These changes in structural variability have direct influences on changes in landscape pattern, causing the landscape to become more coarse-grained as initially dissimilar stands coalesce into larger patches.  Thus large natural disturbances such as the 1988 Yellowstone fires may leave a structural imprint on the landscape, but the initial heterogeneity will decrease as succession occurs.

Landscape ecosystems and the occurrence and management of the Kirtland's warbler

Collaborators: Burt Barnes (University of Michigan), Wayne Walker (Woods Hole Research Center)

Kirtland's warbler is a federally endangered songbird that nests under young jack pine only in northern Lower Michigan.  The warbler nests in a given stand only for as long as there are lower live branches available to shelter the nest, such that it nests only in dense stands interpersed with small openings during a narrow window of duration of stand age.  Because jack pine regenerates over large areas only after fire, fire suppression has limited the amount of young jack pine within the warbler breeding range, and plantations are now the primary method for providing suitable habitat for the warbler.  Our work created a hierarchical classification of ecosystems inhabited by the Kirtland's warbler as a mean to understand and predict the timing and duration of use by the warbler of a given jack pine stand, whether natural or planted.  We noted significant differences in climate, physiography, soil, and vegetation between 10 landscape ecosystems at the ecological level of landforms.  Jack pine height growth rates differed significantly among the 10 ecosystems, and the landforms exhibited marked differences in the timing of initial colonization and duration of occupancy by the warbler.  Ecosystems favoring jack pine growth - those with a warmer microclimate or higher-quality soil - were typically colonized first but had the shortest duration of occupancy, while colder, drier, and less fertile ecosystems were colonized later but had longer durations of occupancy.  Such a description and classification of landscape ecosystems thus provides a useful ecological framework for conservation and management of Kirtland's warbler.