Ecosystem Structure & Function Research
By analyzing the structure and function of ecosystems, we hope to gain a greater understanding of the dynamic systems and relationships, enabling better long-term management. Research in this field covers a wide array of topics, ranging from studies of biological diversity to methods for monitoring change. Some of the past and ongoing Ecosystem Structure and Function Research Projects at SNRE include:
Ecosystem Structure & Function Research Projects
Biological Diversity of Andean Rivers (David Allan)
Research under this proposal takes place in the northern Andes of Venezuela, and examines how the ecology and biological diversity of tropical rivers are altered by changes in landuse at the catchment (watershed) scale. A detailed investigation will be conducted along the length of one relatively undisturbed river from its origin at more than 4000 meters on Pico Humboldt, one of the highest mountains in Venezuela, until it reaches the great lowland savannahs (llanos) at approximately 200 meters in elevation. Using this river as a benchmark, we will compare a series of piedmont (foothill) rivers that have experienced differing degrees of landscape transformation. Also to be addressed are fundamental issues of structure and function of tropical river systems via specific hypotheses concerning altitudinal trends in: (1) species richness and range, (2) size of individuals and biomass of populations, (3) resource availability and guild structure, and (4) habitat structure. The biodiversity of these rivers will be examined by assessing local (alpha) diversity by collecting that includes adults and seasons, and determining local species richness using species accumulation curves. Examination of the habitat (beta) diversity among collection locations will also be done. The long range goals of this project are to expand scientific knowledge of the structure and function of tropical rivers, and to examine how riverine ecology may be altered in response to human impacts and changing land use.
Development of a Nonwadable Rivers Assessment Protocol for Michigan Rivers (David Allan, Richard W. Merritt and Kenneth W. Cummins)
The sampling of nonwadable rivers to etermine macroinvertebrate abundance and diversity, and to assess physical habitat quality, has been neglected relative to small, wadable streams and rivers. This lack of suitable techniques and protocols has severely hampered our ability to assess the effects of environmental stressors on larger rivers (order 5 and above). We propose to develop a Manual of Procedures that will be suitable for general use by field monitoring crews and laboratory technicians who are tasked with conducting assessments of river ecological health. The research plan includes five phases: (1) Develop a protocol for nonwadable river biological integrity and physical habitat assessment suitable for Michigan rivers of order 5 and greater based on studies of large rivers elsewhere. (2) Produce a draft Manual of Procedures based on this protocol.(3) Use the draft Manual of Procedures to conduct an intensive field test of the protocol in three selected reaches of the Kalamazoo River. (4)Use the draft Manual of Procedures to conduct an extensive field test of the protocol in five additional nonwadable rivers in Michigan, selected to be representative of geographical distribution and ecological condition. (5) Use analysis of the intensive and extensive field datato refine the Manual and produce a final document. For more information, please visit the project website.
Recent evidence suggests that the frequency and intensity of El Niño Events are increasing as a result of global climate change. With this change, the frequency and intensity of tropical storms may also increase. This is long-term research that examines the effect of a large and catastrophic hurricane that affected the lowland tropical rain forest of Nicaragua. Hurricane Joan was a large storm that severely damaged about 500,000 ha of rain forest when it struck the Atlantic Coast of Nicaragua in 1988. Studies of post-hurricane succession were initiated a year later and permanent plots were established. These plots have been monitored at least once a year since 1990 for tree growth (dbh and height), mortality and recruitment. We have documented the different phases of the regeneration of the forest and an increase in tree diversity after the hurricane. This research has also contributed to our understanding of the mechanisms that maintain species diversity in tropical rain forests. Until recently, the debate was centered on the role of tree-fall gaps as a primary source of disturbance. This type of disturbance was thought to set back the process of competitive exclusion, thus conforming to the intermediate disturbance hypothesis: neither very large nor very small disturbances can deter the eventual extinction of species, so intermediate disturbance was thought to be critical for the maintenance of biodiversity. Our research strongly suggest that large storms are indeed diversity-preserving and the mechanism for this diversity-preservation is a severe recruitment limitation of the pioneer species that normally arrive in a tree fall gap and other disturbed areas. Our knowledge about the way that the rain forest responds to natural disturbance is contributing to the development of more sustainable methods of selective logging that can benefit the local communities. This research represents a collaborative effort between CIDCA (Center for Research and Documentation of the Atlantic Coast of Nicaragua), Universidad de las Regiones de la Costa Atlantica de Nicaragua and the University of Michigan.
Ecological Changes in the Northern Lake Huron Biosphere Preserve (James Diana and Paul Webb)
This project seeks to understand ecological change in the Les Cheneaux Islands area, brought on by human use of the area. Part of the project, funded by The Nature Conservancy, examines the function of Great Lake wetlands in this area and attempts to evaluate human influences (through road building, draining, and sewage disposal) on biodiversity of these wetlands. Another part of the project examined interactions between double-crested cormorants (a bird that has recently expanded its abundance in this area), yellow perch, and humans. Cormorants eat perch and other prey fish, which influence abundance of the forage fish community. Yellow perch are targeted both by anglers and cormorants, and their overall mortality rate is high enough to severely limit angler harvest. However, estimates in 1995 were similar for mortality of perch due to cormorant consumption (<4%) and due to fishing (2.4%), and both were small compared to annual total mortality (45%). Recent efforts in this project have been to establish a community-based monitoring system for wetland status and trends in organisms in the area.
Effects of Dams, Reservoirs and Watershed Characteristics on Riverine Processes and Fish Populations in the Au Sable River (James Diana, David Clapp, Michele DePhilip, John Hudson, Jennifer Abdella)
This study is focused on stream and landscape modifications and their effects on interaction between two predatory fish: brown trout and walleye. Both species in the Au Sable River system make extensive long-range movements. Brown trout in both the Mainstream and South Branch regularly moved 300-3,000 m each night during the summer. The extent of these movements was correlated to gradient. A third study in Mainstream between Mio and Alcona Pond indicated that brown trout there showed much lower daily movements than in upstream reaches. Brown trout remained within 25 km of Mio Dam (out of a possible 50 km) and during summer remained relatively inactive. Their low rates of activity were correlated again to gradient, but also to presence of dams, which may have interfered with longer upstream movement. Walleye commonly used the same stretch of river as brown trout, as well as Alcona Pond on the downstream end. They regularly moved between the pond and river system, and passed over the lower 25 km. The upper reach of this study area was also heavily influenced by Mio Pond, which combined with local development to reduce large woody debris in the river, to increase water temperature, and to decrease temperature flux. Interactions in this dammed reach include predation by walleye on small brown trout. This interaction may result in a lack of recruitment for brown trout between Mio and Alcona Pond. Dams and impoundments on the AuSable River increase walleye populations and interactions between walleye and brown trout, both to the detriment of the brown trout fishery. They also largely influence stream conditions and aquatic communities within the study reaches.
Effects of Soil Temperature and Nitrate on Fine Root Construction and Maintenance Costs in Northern Hardwood Forests (Donald Zak)
More than half of all carbon fixed in the leaves of forest trees during photosynthesis is allocated to the production of fine roots, but the effects of pollution and climate change on the production and longevity of these roots is unknown. This project is studying the relationship between soil temperature and soil nitrate (a common pollutant in the Great Lakes region) and the production and death of fine roots in sugar maple forests, to better predict what might happen if pollution increases and climate warms. The study sites are separated by several hundred kilometers, spanning a temperature gradient equivalent to the predicted warming that might occur from climate change (i.e., 2 degrees). Nitrate inputs to these forests have also been increased through fertilization. Using small video cameras first developed for arthroscopic survey, the birth, growth, and death of fine roots are being observed via small tubes permanently installed in the soil. Using this approach, root activity between colder and warmer forests, and between forests subject to high and low levels of nitrate pollution can be compared. In addition, root and microbial respiration within the soil is being quantified. Because the metabolic activity of plants and soil microorganisms increases with rising temperature, the respiratory activity of these organisms should be greatest in the warmest forests. These processes should have important implications for understanding how climate change could alter the cycling and storage of carbon and nitrogen within forested ecosystems.
Elevated Atmospheric CO2 and Feedback between Carbon and Nitrogen Cycles (Donald Zak)
The response of plants to atmospheric carbon dioxide has received considerable attention because carbon dioxide is used by plants for growth; it is the key molecule necessary for photosynthesis. The concentration of carbon dioxide in the Earth's atmosphere will likely double in the next century as a result of fossil fuel burning. Although the extent and duration of such an increase is debatable, it is likely to have important consequences for the rate at which organic matter and associated plant nutrients are cycled within terrestrial ecosystems. This project examines the possibility that increases in atmospheric carbon dioxide will accelerate plant growth and the cycling of carbon within terrestrial ecosystems. To test this idea, trembling aspen were grown under different conditions of ambient carbon dioxide concentrations. Researchers observed that increases in atmospheric carbon dioxide accelerate plant growth, especially the growth of roots in soil. However, although plants grow larger at higher than normal atmospheric carbon dioxide concentrations, there was no evidence that such a growth increase will alter the activities of decomposing organisms in the soil. To understand the response of soil microorganisms, a conceptual model depicting the influence of elevated atmospheric carbon dioxide was tested. The model was based on the premise that above and below-ground plant production provide the primary link between rising atmospheric carbon dioxide concentrations and any associated changes in the cycling of C and N within terrestrial ecosystems. To date, data support the hypothesis that, at least for one growing season, increased root growth can elicit a positive feedback response by microbial populations and N dynamics within the soil.
This research examines the function of biological diversity in the coffee agroecosystem in Southern Mexico. In particular, this research focuses on top-down processes associated with reduction of herbivory by vertebrate and invertebrate predators, and their impact on coffee yield. We examine the regulation of herbivory using a comparative approach along an agricultural intensification gradient. Two general hypotheses guide this research: 1) that top down control in agroecosystems can limit herbivory, and 2) that diversity and abundance of predators (vertebrates and invertebrates) affect the degree of this limitation. Using coffee as a model system with varying degrees of diversity, our group predicts that as the diversity of these predators is reduced along the agricultural intensification gradient, so is their ability to control herbivores, with a resultant decline in yield. By means of exclusion experiments we will quantify the effects of insectivorous birds, ants, spiders and parasitoids on the herbivores of coffee. Once these effects are established, we will perform another exclusion experiment to examine the interactions among these predators. Preliminary results show an interesting trait-mediated indirect effect of ants on herbivores. This research is part of a larger project that examines the benefits of traditional agroforestry systems for the conservation of biodiversity. It has guided efforts to certify shaded coffee so that the producers can be compensated for the ecological services that they provide with their diverse agroforestry systems.
Impact of Silvicultural Treatments on Biological Diversity in Northern Hardwood Ecosystems (Donald Zak)
This project is researching the problem of how the function of old-growth northern hardwood forests differ from the function of second-growth forests that now cover much of the Lake States. The basic goal is to increase our basic understanding of how widespread deforestation in the late nineteenth century altered the composition, structure, and function of Lake State forests. There are three main research objectives: (1) Compare old-growth northern hardwood forests with compositional, structural, and functional characteristics of second-growth northern hardwood forests that are unmanaged, or have received various management treatments; (2) relate differences in the composition and structure of second-growth northern hardwood forests brought about by current management practices, to the ecological processes in old-growth northern hardwood forests; and (3) explore ways to create old-growth functional characteristics in managed second-growth northern hardwood forests. To accomplish these objectives, ecosystem level Carbon and Nitrogen budgets were constructed for one old-growth and three different second-growth (even-aged, uneven-aged, and unmanaged) northern hardwood ecosystems in the Ottawa National Forest. Two stands in each of these forests were selected. Within each stand, three plots were randomly located to measure the storage of Carbon and Nitrogen in overstory trees, understory vegetation, and soil.
Land Use and Ecological Change in Lakes of the Northern Highland Lake District, Wisconsin (James Diana, Sarah Zorn, Ashley Johnson)
The Northern Highland Lake District is an area of extreme lake density, with a large number of lakes protected in national and state forests. The area is very scenic and is heavily used as second-home and recreational areas for a number of people in the Midwest region. These lakes are also home to some unusual fish assemblages, particularly those containing muskellunge in small inland lakes. Muskellunge are large, predatory fish, popular as game fish, and often reach sizes up to forty or more pounds. This unique species has created a demand for sport fishing in the area and for management of the sport fishery as well. We have conducted a series of studies evaluating spawning habitat, spawning success, and the relationship between the land use in the area and spawning of muskellunge.
The first study focussed on early life history of muskellunge in four lakes; two with good reproduction and two with limited reproduction. Spawning areas in naturally reproducing lakes had variable dissolved oxygen at the sediment water interface (0.5 - 9.6 mg/L), while stocked lakes had low dissolved oxygen. Spawning habitat was identified by visual assessment and fish tended to use areas with emergent vegetation and organic substrates. Fallen logs, stumps, and other wood in spawning areas likely increases egg survival. Egg survival in controlled habitats on natural substrates was very low, even for self-sustaining populations. Collections of eggs and observations of fry indicated that the major mortality occurred after egg deposition, but prior to swim-up by muskellunge in both stocked and natural lakes.
The results of this study encouraged us to begin further work on a larger selection of lakes. This study focused on 22 lakes, with varying rates of natural reproduction, from nonexistent to excellent. The lakes also varied in land use and other characteristics. Again, in all lakes, muskellunge habitat was evaluated visually during spawning, and the characteristics of that habitat were measured. Land use in the region was measured both on the lake front by direct means, and in the watershed by GIS maps. Various measures of development in the watershed, as well as on the lake shoreline, were significantly related to variation in spawning success of muskellunge. In addition, some lake characteristics, such as total phosphorus, were also related to muskellunge spawning success. The habitat deterioration during spawning in the impacted lakes included increases in sediment softness, decreases in dissolved oxygen at the sediment water interface and lack of woody material in spawning habitats. These changes were induced by local land use change.
While these studies are still in progress, we hope that the large-scale landscape variables influencing spawning success can be used in a predictive fashion to both determine lakes where spawning success is likely to be high, as well as to target lakes where remedation might occur. In addition, we hope that these studies will help in the management of development in the Northern Highland Lakes region. We hope to involve both sport fishing groups and management agencies in the evaluation of spawning success in a larger number of lakes, and in some mitigation activities to try to reestablish muskellunge as a keystone predator in these lakes.
Landscape Ecology and the Occurrence of the Kirtland's Warbler (Burton Barnes)
Kirtland's Warbler is a rare and endangered species whose summer breeding grounds are only in northern lower Michigan. The species is dependent on jack pine-oak forest landscapes, which require fire to regenerate. In 1980, a prescribed burn got out of control and burned 23,000 acres. It provided a unique opportunity to study the response of the warbler over much more diverse landscapes than had previously been available for study. The research during the 1980s was the first of its kind demonstrating that an understanding of whole ecosystems can explain the warbler's interactions with the landscape, and highlighted the limitations of research that focus solely on the organism itself. The current Kirtland's warbler research has three parts: (1) monitor the pattern of Kirtland's warbler occurrence in landforms and ecosystem types of the Mack Lake burn, (2) study past and present areas colonized by the warbler in nine counties, and (3) study the effect of different ecosystems on the height growth and crown development of jack pine. The result of this study will be a detailed understanding of good versus poor warbler habitat in terms of the ecosystem components of physiography, microclimate, soil, and vegetation. An important application of this knowledge will be to develop guidelines that assist the DNR and the US Forest Service to better select landscapes to designate as Kirtland's warbler management areas for maximum duration of warbler occupancy.
Since 1986, three major stress factors have affected Michigan forests. These factors are insect defoliation, drought, and frost. To understand the effects of these interactive stress factors on Michigan forests and their healthy maintenance, a set of solid baseline data on Michigan's most susceptible forest types was needed and in 1991 the Michigan Impact Monitoring System (MIMS) was established. The monitoring system is composed of 566 plots measured on a regular basis to track changes in conditions and understand these changes in relation to the stress factors present in the stands. To date, this research has shown that overall hardwood forests in Michigan are in good condition with relatively few dead trees. However, much variation in tree vigor, mortality, and regeneration occurs among individual stands and ecosystem units within each forest type and also among tree species within a forest type. We are continuing to study whole-ecosystem sustainability in these systems and the extent to which different ecosystem types are vulnerable to reduction in health. We are also continuing to examine regeneration , particularly in various ecosystem units of the oak forest type. Benefits of this research include aiding land managers in understanding stress impacts and enabling them to better manage forest areas affected by various biotic and abiotic stresses. Results from this study have been used to help the Michigan DNR justify a lack of pesticide spraying for gypsy moth control in rural forest areas.
National River Restpration Science Synthesis Project (David Allan)
The National River Restpration Science Synthesis Project is a unique undertaking by a team of eminent freshwater scientists and American Rivers that will evaluate the state of river restoration in the United States - what is being done, what is 'working' from an environmental and conservation perspective, and what the costs and benefits are for various types of projects.
This project continues the examination of the relationship between non-sustainable land-use practices and nonpoint source pollution, and the widespread decline of the biological and functional integrity of aquatic ecosystems of the River Raisin watershed in southeast Michigan. It is believed that protection of riparian corridors can help in the development of sustainable agricultural practices that also maintain high water quality. However, the benefits conveyed by a vegetated riparian area are not well understood. Based on preliminary evidence, land use at the watershed scale is a better predictor of stream ecosystem degradation than are riparian measures. The goals of this study are to determine the characteristics and extent of riparian vegetation throughout the watershed, and to evaluate the protective benefits of various riparian configurations, from the perspectives of soil loss and biological integrity. By achieving these goals, the researchers expect to be able to answer specific questions and test hypotheses concerning the effects of land use, management practices, riparian conditions and instream physical habitat on the function and biotic integrity of stream ecosystems, and develop a predictive model relating these elements and contribute to USDA's ability to assess the effects of agricultural practices on the functional integrity of aquatic ecosystems.