Research Philosophy:
Since its inception in 1982, the overarching philosophy of the Cedar Creek LTER has been, and remains, the synthesis and integration of the principles and processes of population, community and ecosystem ecology. In our original (1981) LTER proposal we said:
Population and community ecology can be greatly strengthened by consideration of the long-term effects, the indirect effects and the feedback effects that the ecosystem approach emphasizes. Similarly, ecosystem ecology can be strengthened by detailed studies of the dynamics of interactions among individual species that play such a key role in the processes of productivity, energy flow and nutrient cycling. We believe this synthesis will only be possible when long-term experimental research combines population, community and ecosystem perspectives.

This philosophy has led us to simultaneously collect, from the same observational and experimental sites, long-term, detailed information on the abundances, dynamics and diversity of species on several trophic levels, on the controls of the pool sizes and dynamics of such ecosystem traits as soil C and N, on above and below ground primary productivity, on N mineralization, and on disturbance regimes. It has led us to seek ways to scale up from ecophysiology and interspecific interactions to their effects on ecosystem stability, productivity, nutrient dynamics, and susceptibility to invasion. It has led to long-term experiments that explore the impacts on species abundances, diversity, and ecosystem functioning of the major driving variables at Cedar Creek: soil N, trophic structure, fire frequency, and human land use. The results of these on-going experiments have, themselves, led to new long-term experiments that explore the effects of species composition and biodiversity on ecosystem functioning, and the joint effects of biodiversity, elevated CO2, and N deposition on population, community and ecosystem processes. In all this work, we seek to identify potentially generalizable ecological principles through the synthesis of population, community and ecosystem processes, patterns, and mechanisms.

Context:
All work is performed within the context of our 22 km2 site, Cedar Creek Natural History Area. Cedar Creek is unique among the LTER sites because of its location on the climatically, edaphically, and disturbance controlled boundary between North America’s prairie and forest (Fig. 8 [pdf]). As Curtis (1959) suggested, the oak savanna vegetation that characterizes this "tension zone" between forest and prairie is highly sensitive to climatic variation (Tilman and El Haddi 1992, Faber-Langendoen and Tester 1993, Tilman and Downing 1994, Tilman 1996, 1999). It is also strongly impacted by edaphic factors, especially soil N availability (Tilman 1987, 1988, 1993; Inouye et al. 1987c; Hairston and Grigal 1991), by fire frequency (White 1983, Tester 1989, Faber-Langendoen and Tester 1993, Peterson and Reich 2000a, Reich et al. 2000) and by herbivores and predators (Inouye et al. 1987a, 1987b, Huntly and Inouye 1984, 1987, 1988, Ritchie and Tilman 1992, 1993, Ritchie et al. 1998, Ritchie 2000). Cedar Creek contains large tracts of native oak savanna, of prairie and successional grasslands, and of oak forest, and smaller stands of basswood-sugar maple, white pine, jack pine, and red pine. This diverse mosaic of upland ecosystems has been shaped by the soil parent materials, by a variety of landscape features that influenced fire frequency, by the pattern and history of physical disturbances, and by European settlement and agriculture. These upland ecosystems also are species rich. A m2 of savanna contains 10-40 plant species, and a hectare about 200 plant species. Over 1200 insect species, including 577 species of herbivores, 226 species of predators, 281 species of parasites, and 141 species of detritivores, were found in a quantitative survey of successional grasslands and savanna (Siemann et al. 1996, 1999a). About 4000 insect species are in our on-site collection, which, like our herbarium, is accessible at www.lter.umn.edu. The species-rich successional grasslands and savannas of Cedar Creek are the focus of the LTER program. Our research explores the various physical factors (climate and its variation, soil N, fire, and various disturbances), species interactions, and feedback effects that have shaped various aspects of these ecosystems, including their species diversity, species composition, productivity, nutrient dynamics and stability.

Research Themes:
The Cedar Creek LTER has evolved since its founding to address both elaborations on its initial themes and new themes inspired by our results and their relevance to the work of others. We were initially interested in the causes of high plant and insect diversity, in the controls of community assembly, and in processes controlling the responses of species and ecosystems to disturbances. These interests were linked by our search for underlying mechanisms, and by the hypothesis that similar mechanisms would be involved in all three. After all, successional dynamics are the dynamics of community assembly after disturbance, and diversity is one component of the type of ecosystem created by the assembly process. Our long-term studies continue to address succession and community assembly, the controls of diversity, and the impacts of disturbances on population, community and ecosystem traits, processes and dynamics. Our studies of succession continue to give us insights into community assembly. The process of community assembly is of interest in its own right and is directly relevant to an emerging area of interest for many Cedar Creek researchers and collaborators – the processes controlling abundances of exotic or invasive species. Moreover, such questions are deeply intertwined with the mechanisms of multi-species coexistence because species persistently coexist only if each species can invade communities from which it is absent.

As our understanding of the ecosystems of Cedar Creek has grown, we have broadened our perspectives and used long-term studies to address a variety of additional questions. One of these – the consequences of the loss of biodiversity for ecosystem functioning – has grown into a major theme. We proposed two long-term experiments on the ecosystem effects of biodiversity and functional group composition in our 1994 LTER renewal, and have since started (mainly with funds from another grant) a third experiment that explores the joint and interactive effects of biodiversity, N deposition, and elevated atmospheric CO2. Moreover, our ongoing work is raising many new questions about the consequences of N deposition, elevated atmospheric CO2, changes in fire frequency, agricultural land use and abandonment, loss of biodiversity, and changes in trophic structure. These disturbances are human-caused. As such, our initial and continuing interest in and focus on disturbance and diversity, combined with the expanding global impacts of human society, have resulted in an increasing focus on the effects of human ecosystem domination (e.g., Vitousek 1994, Vitousek et al. 1997). In total, we now have four major questions, and a host of subsidiary issues, that provide the interdependent themes for Cedar Creek research:

Theme 1: What are the impacts of major perturbations -- especially climatic variation, N deposition, land use history, changes in fire frequency, elevated CO2, exotic species, and changes in trophic structure --on species composition, diversity and ecosystem functioning?
Theme 2: What processes, interactions and positive and negative feedbacks control species abundances, community assembly, and community composition, diversity and dynamics in Cedar Creek grasslands and savanna?
Theme 3: How do composition and biodiversity directly and indirectly impact ecosystem functioning?
Theme 4: What general principles allow integration across scales ranging from ecophysiological and population processes to ecosystem functioning; from single trophic levels to whole foodwebs; from single plots to landscapes; and from snapshots in time to long time series?

We are pursuing these four themes in five inter-related types of long-term studies that form the heart of the Cedar Creek LTER. Each is guided by our research philosophy and each addresses several themes. LTER funding supports this core long-term work and the research infrastructure of Cedar Creek (computer network, analytical chemistry laboratory, herbarium and insect collections, data management and software development, and shared research equipment). Additional grants allow deeper pursuit of questions. For instance, L. Kinkle and colleagues are working on soil microbial dynamics in the long-term N addition experiments using their NSF ‘Microbial Observatory’ grant. A DOE-funded grant is allowing us to expand work on the effects of biodiversity to include an experiment in which we factorially vary N deposition, CO2, and biodiversity. A grant from the Andrew Mellon Foundation is supporting parallel work on community assembly and invasibility both at Cedar Creek and, in collaboration with Jim Reichman, in the oak savannas of U. C. Santa Barbara’s Sedgwick Ranch. Nancy Johnson of Northern Arizona University has an NSF grant to work on the interactions of plant diversity and mycorrhizal fungi at Cedar Creek. The British NERC is supporting work by J. P. Grime and Graham Burt-Smith to test the ability of Grime’s theory to predict the observed effects of plant diversity on productivity. Indeed, there are currently 17 grants from diverse funding sources supporting work related to LTER by 25 researchers, many of whom are not directly supported by LTER. Thus, by establishing, maintaining, and collecting core data in long-term studies, the LTER provides a data-rich environment that attracts additional researchers. Just as greater plant diversity increases primary productivity, so may the greater intellectual diversity that results from having many others work at Cedar Creek lead to greater scientific creativity and productivity.

Our five types of long-term studies, which are described in detail below, are: (1) experimental manipulations of plant diversity; (2) N addition experiments; (3) fire frequency experiments; (4) experimental manipulations of herbivore and predator trophic levels; and (5) long-term observations of soils, plants, insects and mammals in a successional chronosequence and savanna (Table 1). Each of these five types of studies is designed to give insights into two or more of our four themes. For instance, the diversity experiments directly illustrate the ecosystem consequences of plant diversity (Theme 3), and also address both Theme 1 via long-term observations of the effects of climatic variation on species composition, diversity and ecosystem functioning, and Theme 2 through the observed process of species coexistence and community assembly (reflected in how the abundance of each plant species depends on the identities and diversity of the other species planted in the same plot). The N addition experiments and the burning experiments are directly relevant to Theme 1 because N deposition and fire suppression are perturbations and their long-term data allow determination of the effects of climatic variation. Moreover, they also are directly relevant to Theme 2 because N is the major limiting resource of Cedar Creek and its availability influences plant, herbivore and predator composition and diversity, just as does fire frequency. Moreover, by influencing diversity, N addition also gives insights into the effects of diversity on ecosystem functioning. The experimental manipulations of higher trophic levels demonstrate direct effects of these on plant community composition, diversity and dynamics (Theme 2), and indirect effects mediated through their impacts on ecosystem processes. Finally, our long-term observations describe successional and community assembly patterns and processes that are directly relevant to Themes 1 and 2. Because sites within and among fields differ in diversity and composition, they give insights into Theme 3. They also provide data on our broadest spatial and temporal scales, and thus provide opportunities to test alternative ways to scale up from the smaller plots, shorter time scales, and underlying mechanisms to larger temporal and spatial patterns (Theme 4). In total, our overarching philosophy of synthesis of population, community and ecosystem perspectives guides our work in these five types of long-term studies, as do our four unifying themes. In the following sections we describe in more depth these five types of long-term studies.