2 |
Ecosystem Processes and Global Change |
| ROBERT G. WOODMANSEE | |
| Natural Resource Ecology Laboratory, | |
| Colorado State University, | |
| Fort Collins, CO 80523, USA |
Man's responsibility to evaluate and forecast his changing environment is being recognized as a major scientific and societal need for the coming decades. Not only small-scale or local environments, but the entire global environment requires evaluation. To accomplish the required evaluations, scientists and political decision-makers must acquire knowledge that will allow the integration of information ranging from small geographic scales (metres to hectares) and short temporal scales (minutes to hours) to large spatial scales (regions and the globe) and long temporal scales (decades to centuries). This integration is necessary to explain the role of biogenic processes in the changing global environment.
Discussions in this paper focus on problems and questions that relate ecological processes occurring at various hierarchical levels to the levels above and below that of current attention. The discussions emphasize that resolution of our environmental problems will require transdisciplinary team research. New institutional structures will probably be required to accommodate such research.
Humans, largely through scientific achievements, have developed into a species that has significantly and profoundly influenced their surrounding environment, from local to global levels of resolution. Humans have had a great effect on their fellow species of organisms, both directly, by harvest, and indirectly, by altering habitat. Yet, even though our conceptual, explanatory models of causality are complex and scientifically sophisticated, they are still simplistic and imperfect and, especially on a large spatial scale (regional or global), often are based on little more than dogma and intuition. With our best current scientific knowledge, we can explain only local environmental responses for short periods of time. When events occur that we do not understand, we declare them stochastic or too complex to explain and state that more data are needed to better understand the phenomena of interest. Our typical response is to conduct reductionist-oriented research about narrow but tractable topics. These topics rarely address problems of large-scale ecosystems such as landscapes, regions, and, indeed, the globe.
Because we have and will continue to alter our environment, we must accept the responsibility to explain rationally how we interact with it from local to global scales of resolution. We must learn to manage and direct those interactions so that we can maintain a hospitable environment. A major aim of research must be to gain the knowledge necessary to explain our environmental interactions at many spatial and temporal scales and to develop appropriate models to communicate that knowledge to decision makers and other scientists. Specifically, we must gain knowledge of biogeochemical cycles, primary production, and physical processes such as erosion and sedimentation, for time scales greater than a few years and spatial scales larger than a few square metres.
The primary goal of this chapter is to discuss examples of small-scale ecological phenomena and how they influence and interact with the broader environment, especially at the scale of landscapes and larger. This discussion will focus on the scientific process required to understand biospheric interactions. To accomplish this goal, I will:
Most ecological research has focused on problems that were perceived to be tractable in two- or five-year time frames because of funding constraints, and at spatial scales of a few square metres or, at most, a few hectares, because of our inability to deal effectively with heterogeneity. This focus is inadequate to assess the issues that truly concern large-scale ecosystem processes and dynamics. The scope of future research must extend to ecosystems dynamics requiring decades to centuries to be expressed and spatial scales that range from sub-continental to global in size.
Many large-scale (including global) phenomena such as increasing concentrations of greenhouse gases, erosion of agricultural lands, and loss of genetic diversity, are mediated by ecological processes that occur at small scales at the surface of the globe (Bowden and Bormann, 1986; Schimel et al. , 1986). Some of these problems are significant for the welfare of humans, and all are influenced by 'natural' and managed ecosystem behaviour (reviewed in Smil, 1985; Bolin and Cook, 1983). We currently have considerable information about important ecological processes that influence numerous problems at the small scale of spatial resolution and the short-term (less than one year) time scale. However, extrapolation from small scales to larger scales is tenuous at best. Among the most significant problems facing humans are increases in the concentrations of trace gases including CO2, erosion and sedimentation, and human caused changes in the composition of biotic communities.
Concentrations of numerous biogenically controlled trace gases in the atmosphere are increasing. Among these are several gaseous species of nitrogen which are involved in complex chemical reactions which may ultimately influence the transmission of ultraviolet radiation to the earth's surface, influence atmospheric heating, or become involved in complex tropospheric chemical reactions which may profoundly influence the chemical environment of the biosphere (Crutzen, 1983). The influence can be direct, through absorption by organisms, or indirect, through deposition of chemical substances at the surfaces of ecosystems, resulting in complex reactions that alter our chemical environment (i.e. Strickland and Fitzgerald, 1984; Schindler et al., 1985; Olsen et al., 1985; Bowden, 1986).
Quantities of CO2 are increasing in the atmosphere because of industrialization and land-management activities (Smil, 1985; Bolin and Cook, 1983; Houghton et al., 1983). A consequence of higher CO2 levels may be altered patterns of plant productivity, with consequent effects on animal and microbial populations and functioning. Alteration of the biological components of ecosystems subsequently may alter the patterns of soil organic matter accumulation and biogeochemical cycling (Schlesinger, 1984; Schimel et al., 1985a). These system responses may result from either direct enhancement of CO2 levels on plants or indirect influences on the local environment caused by climate change (Strain, 1985).
Questions that must be addressed in the next decades include: If there is an increased incidence of ultraviolet radiation at the surface of the earth due to the chemical destruction of O3, as some scientists suspect, can biogenically produced gases such as N2O be shown to be significantly implicated in that destruction? Will there be a significant impact on the world's agricultural productivity that can be attributed to increased biogenic emissions of trace gases? Will a changing chemical composition of the atmosphere have detectable effects on the physiology of species that are deemed important to humans?
Erosion and sedimentation are linked processes which are extensively modified by man's activity (Schumm, 1977). Productivity of many terrestrial and aquatic ecosystems is being significantly altered by either increasing or decreasing rates of these processes (Schimel et al., 1985b, c). Even though erosion has been studied extensively and our knowledge of small scale dynamics is extensive, our understanding of these linked processes at scales larger than uniform field plots is limited. The public is told that enormous quantities of soil are lost from our farmlands each year, but little attention is paid to the amount of soil that is moved and redeposited within the same landscape. What is the ecological importance of this reworked soil material? What is the importance of differentially deposited soil sediments in aquatic ecosystems? These questions remain unanswered.
The patterns of biotic succession at many sites throughout the world are undergoing rapid change, largely because of major man-caused disturbance and greatly enhanced rates of exchange of propagules among sites (i.e. West et al., 1981; Houghton, et al., 1983; Detwiler, 1986). Implications of these changes include the possibility that large-scale genetic diversity will be significantly reduced and /or exotic species will significantly alter local ecosystem behaviour. Will changes in species composition of major biotic communities influence biogeochemical cycles locally or globally? Will there be a detectable change in the earth's biological productivity that can be attributed to human-caused loss of species?
These problems represent global-scale issues that will require information about ecological processes for solution. I contend that the global-scale problems cannot be solved until we learn to skilfully integrate appropriate smaller-scale information into the context of the defined large-scale concerns.
I view global-scale concerns as including two classes of problems. One class I call universal phenomena, and the other I call disjunct phenomena. Universal phenomena are experienced with relative uniformity throughout the globe. Examples of such phenomena are CO2 and N2O concentrations in the atmosphere. These gases are distributed throughout the atmosphere. Disjunct phenomena are local or regional, but they are experienced to some degree in many different locations in the world. Examples of disjunct phenomena are groundwater contamination, erosion and sedimentation, and biotic succession. Unfortunately, many important phenomena are expressed along a gradient of scales from global to local. These phenomena include increasing CH4, NH3, atmospheric particulate and aerosol concentrations and climate change. These latter phenomena show distinct patchiness around the globe, but they are present to some extent everywhere. We must learn to express what class of phenomena we are interested in, and clearly define problems associated with the phenomena.
The term ecosystem is used. herein to mean any level in an ecological hierarchy defined as an interconnected system of parts. Definitions of various ecosystems are developed and organized by humans, to evaluate a specific problem or enquiry. The word ecosystem has no distinct meaning itself and must be accompanied by an adjective to become descriptive.
An ecological hierarchy is used here to indicate a ranking or ordering of ecosystems that is useful in describing and analysing a particular complex problem that requires information of varying resolution for evaluation. The concepts of ecosystem and ecological hierarchy are conveniences of the human mind that should be used as tools of logic (O'Neill, 1988).
A useful premise aiding analysis of hierarchical systems is that the understanding of one level in a hierarchy requires a thorough consideration of the level above and the level below (DeWit, 1970; Forrester, 1968). Pragmatically, questions guiding an analysis are developed from the level above the level of interest, and the specific analysis of system components and processes are conducted on the level below. I will develop the concept of ecological hierarchy below using an example of trace gas dynamics in the biosphere and atmosphere.
An ecological hierarchy for use as a conceptual device aiding analysis of trace gas dynamics in the terrestrial biosphere and atmosphere is shown in Figure 2.1. A guiding global question requiring analysis is: What is the relative contribution of biogenically produced trace gases in the total global budgets of the gases? Appropriate analysis of this question requires determining flux rates of these gases at their sources, that is, at the location where the biological activity responsible for their formation is prevalent. That location is, in reality, at the soil pore and individual bacterium, bacterial colony or soil enzyme scale of spatial resolution (a few mm3 ). However, for pragmatic reasons, most field studies are conceptualized at the patch scale of spatial resolution.
Figure 2.1 An ecological hierarchy useful for conceptualizing the levels of resolution necessary to relate small-scale trace gas emission phenomena to global-scale atmospheric chemistry.
A patch in the hierarchy of Figure 2.1 can be an ecosystem defined as a specific biotic community that resides on a definable soil type (Figure 2.2). An appropriate box and arrow representation of a biogeochemical cycle, such as N of a patch ecosystem, is shown in Figure 2.3. The size of a patch is typically a few square metres to a few hectares; it commonly has been described in units of an average square metre or an average hectare. A critical assumption that is made in defining a patch is that all organisms, physical features such as soil pores, soil particles, dead organic matter, nutrients, and water have known distributions within the patch. Thus, any square centimetre or square metre behaves like any other spatial unit within the patch.
Most ecological research has stressed spatial scales of patch size or smaller. Many ecological controlling influences that have been and are currently being addressed illustrate the importance of patch-scale processes in interpreting larger-scale dynamics. Water is probably the most widely studied control of trace gas emissions from patches. For example, the amount of water available for biological activity producing trace gas within a patch is a function of:
(1)
the amount that falls as precipitation
(2) the amount that runs onto the patch
(3) the amount that runs off
(4) the quantity lost via evaporation and transpiration
(5) the availability of subsurface water.
Figure 2.2 Representation of a patch ecosystem. Both the biotic
components and soil properties are assumed to be uniformly distributed and
describable with ordinary statistical procedures.
Figure 2.3 The major components and processes of typical
patch ecosystems.
The response of organisms responsible for gaseous emissions is controlled by the amount of water available for growth; the availability of nutrients, which are, in turn, largely controlled by water; and other factors, such as temperature and light. The amount and timing of water available in a patch controls the rates and, ultimately. the total amounts of gas produced. For example, saturated soils may favour biological denitrification but limit nitrification, while well-aerated soils may favour nitrification and completely retard denitrification. Additionally, frequently saturated soils may favour one suite of organisms, while well-aerated soils may support a completely different set of organisms. As we generalize about ecosystem behaviour, it is critical to recognize that adjacent patches often display very different properties, and, thus, very different responses.
The discussion in this section relates appropriately to patch-scale phenomena. Meaningful questions at this scale that are related to the global question above are: What is the contribution of trace gases of each patch (the level of field research on biological processes) to the air mass in the canopy boundary layer? To what extent does water derived from uphill patches control the biogeochemical cycles and consequently gas emissions from downslope patches? Simple generalization from this scale to larger scales is often misleading, if not foolhardy.
A flowpath is a connected and interrelated series of patches. At least two concepts of flowpaths are important in consideration of trace gas dynamics. An example of one, a product flowpath shown in Figure 2.2, is an air mass that moves along wind pathways within or directly above vegetation canopies of adjacent patches. As the air mass moves along its trajectory, the gases within not only are subject to chemical transformation in the air, but some also are subject to reabsorption by the vegetation, soils, or free water along the ground surface. Simple examples of control flowpaths are toposequences or soil catenas, where water or wind are the vectors of transport; small streams; and groundwater channels. More complex examples of flow paths exist, but they will not be discussed here (see Woodmansee and Adamsen, 1983, for discussion).
The flowpath is a concept that embraces the influences of one patch on another. However, some patches may be isolated and thus are not interactive with other patches. Questions important at the flowpath scale are beginning to receive scientific attention. For example, do patches that occur in downwind positions significantly alter the chemical composition of canopy air masses derived from upwind patches? To what extent are trace gases emitted from one patch absorbed, or chemically transformed, by soils or biota of adjacent or downwind patches? What are the rates of chemical transformations of various trace gases as they move along atmospheric flowpaths?
Few examples exist that can be used to answer these questions about trace gas behaviour in canopy boundary layers. One notable exception is found in Hutchinson and Viets (1969). They showed biogenically evolved gases (NH3) from one patch to be absorbed or adsorbed by adjacent or downwind patches along product flowpaths. Simple calculations of gas losses from individual patch ecosystems may be inadequate for estimating gaseous inputs to the atmosphere.
Increasing attention is being devoted to the influence of adjacent patches on each other in relation to surface and subsurface drainage pathways or control flowpaths (Correll, 1983). The focus of much of this research is directed towards determining the degree to which the ecological processes in one patch mediate the responses (i.e. nutrient output) of other patches. Recent research results indicate that even though large amounts of fertilizer nutrients may be lost in drainage waters from up slope farm fields, the same nutrients are taken up by forest ecosystems in down slope positions (Correll, 1983). These uptake processes result in the transfer of much higher-quality water out of the watershed than if the slopes are cultivated to the streamside. This example illustrates that knowledge about either of the patch ecosystems alone is insufficient to explain larger-scale system behaviour. Comparable examples need to be developed for atmospheric flowpaths.
Current ecological research is beginning to focus on information that will help explain the responses of landscape ecosystems, the scale important to man's physical, economic, or emotional welfare (i.e. pastures, farm fields, and habitat for wildlife). At this scale the focus is on the interaction of flowpaths (and isolated patches) and how they integrate into complex, whole-system responses, such as introduction of chemicals to the planetary boundary layer of the atmosphere, as water quality, beef production, farm-product yields, or sediment yields to waterways. Landscape is a term used by many disciplines, but it has no universal definition. Unfortunately, I know of no word that better describes this level in the ecological hierarchy. I use the term to be a geographic unit that contains complexes of interacting flow paths that are conceptually related in an analysis of trace gas dynamics, the linked processes of erosion and sedimentation, and biotic community dynamics. The geographic unit may be a watershed, a farm or farms, a lake, or a pasture. Using the trace gas example, an air mass that has a characteristic chemical composition above a mosaic of patches and flowpaths could define the boundaries of the landscape. A critical assumption, using this logic, is that even though emission and absorption rates from patches within the landscape may greatly vary, the mixing processes and chemical reactions within the air mass in the canopy boundary layer are distributed in a relatively uniform way. Neighbouring landscapes would yield air masses with different characteristics. Exchange processes occur between the canopy boundary layer and the planetary boundary layer at this level of resolution. A scientific hope would be that appropriate analyses of regions, the next level above landscapes in the hierarchy, would reveal repeating and predictable patterns of landscapes.
A fundamental need in environmental sciences is to describe better the dynamics of interacting flowpath-scale phenomena, but to do so we need to examine some fundamental facts related to the formation of landscapes. One fact is that the landscape is the level at which the products of biogenic gas formation are admixed into the planetary boundary layer. The landscape is the smallest geographic unit that can pragmatically and routinely be monitored using current remote sensing technology. It is also the level in which control pathways interact and form unique conditions that may favour or inhibit biogenic formation of trace gases.
The landscape-scale concept in the ecological hierarchy discussed here greatly expands the descriptive meaning of the term currently being used by many geographers and, indeed, many ecologists. The meaning here includes the descriptive aspects, but the emphasis is on the processes that occur within landscapes that lead to the exchange of matter, radiation, or information (sensu controlling factors for biological and physical processes). I believe that the first two are the vital issues in the context of global change.
A region has no consistent biological meaning, since larger geographic units are admixtures of landscapes ranging from urban-industrial to relatively natural. However, regional air masses are formed by the coalescence of air masses derived from landscapes. The boundaries of regions are highly dynamic because of varying wind-directed air mass movement. The chemical constituents of specific regional air masses ultimately mix as they migrate along global circulation pathways. Excellent, but terrifying, hypothetical examples of this type of mixing in the troposphere can be seen in the SCOPE nuclear winter scenarios (Harwell and Hutchinson, 1985; Pittock et al., 1986).
Even though regional ecosystems are heterogeneous and complex arrays of landscapes, they do have many characteristics that distinguish them from one another and from their constituent components. An important question to be asked at the regional level is: Can generalizations about the chemical composition of air masses be made for different regions of the world, or is each region of the world unique because of its own peculiar combination of chemical constituents? The chemical composition of air masses seems to be a function of many interacting sources and processes that typically occur in patch-scale ecosystems. Emissions of gases from one landscape, whether an industrial complex or a grouping of farms, is typically inadequate to explain the chemical nature of the atmosphere above a region. The air mass above a region is a complex mixture of all sources combined with continuous transformations therein. Our scientific hope is that patches and flowpaths within landscapes form definable and repeatable patterns, and thus the landscape can be considered a homogeneous unit. Within this context, we can integrate information from landscape units into broader regional contexts.
Our knowledge of the interaction of patch-, flowpath-, or even landscape- scale ecosystem charactetistics with one another to form regional ecosystems concepts is sketchy at best. Likewise, our knowledge of the interactions of regional climates with those of smaller scale is deficient.
Understanding the behaviour of ecological processes, especially in terrestrial ecosystems, requires a thorough knowledge of the interactions of the hydrologic cycle with dominant biogeochemical cycles, and biotic community composition appropriate to each scale within the ecological hierarchy described above. Because of the importance of these interactions in ecosystem functioning, they are emphasized in the following discussion.
The discussion above emphasized spatial patterning and relationships. Temporal dynamics and scales also are critical, as they relate to hydrologic and other physical processes. Physical attributes of water control the dynamics of important ecological phenomena at many spatial scales. For example, at the time scale of seconds to minutes, erosion and sedimentation are profoundly influenced by the energy associated with moving water. This energy is, in turn, influenced by rainfall intensity (or snowmelt rate), runoff and runon rates, the angle of slope of the flowpath, the size of particles, and surface roughness. If we assume that each patch along the flowpath has its own characteristics with respect to these controlling influences, then each patch displays not only its own response but also contributes to the collective behaviour of the larger flowpath and the even larger landscape during intense erosion and sedimentation events.
Examples of ecological responses appropriately evaluated during time scales of hours to days are the formation, evolution, and transformation of biogenically derived cases, such as NH3, NO, CH4 and N2O. Conditions that lead to the formation of these compounds are frequently linked to wetting, aeration, or freezing cycles of soils.
Whereas the ecological processes that yield numerous trace gases can begin to operate in hours, the combination of processes that lead to plant biomass production generally require days to weeks to become apparent. Water must be present to support root and stem growth and to allow microbial mineralization to supply nutrients required by the plants. Thus, plant production should be envisioned as an integration of many ecological processes that must occur in synchrony to yield significant plant growth (Woodmansee, 1984).
Biotic succession and soil development are complex ecological phenomena that are integrations of myriad biological, chemical, and physical processes that require years to decades, and often centuries, to be expressed. The processes of succession and soil formation may never come to completion because they are often closely correlated to climate and climate changes. A major unanswered question in many patch- to landscape-scale ecosystems is the extent to which the biota and soils are interrelated with climate in a dynamic sense. Superimposed upon this uncertainty is the dramatic influence of humans, both as agents of climatic change and as a major force in disturbance of ecosystems.
Position within the landscape is vital to the expression of ecological responses within patches. These responses are often strongly influenced by water availability, erosive potential, seasonality and duration. However, the configuration of the landscape itself is often the result of erosion and sedimentation. Water can be the direct geomorphic agent that shapes the landscape; or, in its absence, wind can be the major agent. The time scale for landscape shaping is centuries to millennia.
Thus, landscapes, which are composites of patches linked into flowpaths of varying relationships, form units that can be described and explained and whose behaviour can be predicted with some reasonable degree of confidence. Global scale problems can be addressed if patches and flowpaths can be described and their composite behaviours integrated at the landscape-scale, and if landscapes can be shown to form repeatable patterns. If these requisites can be met, prediction of regional-scale phenomena may be achievable and a goal of global-scale understanding may be within our grasp.
The discussion above describes examples of the importance of phenomena that occur at small spatial scales, especially the patch, in influencing the behaviour of larger-scale phenomena. The dimension of time is inextricably interwoven with the dimension of space. However, our ability to extend knowledge from small spatial scales and short time scales to evaluation of problems of regional and global importance is not well developed. A major challenge facing humans who perform science in the next decade and beyond is to develop our knowledge base to adequately evaluate large-scale problems.
Limitations on development of knowledge are:
We know that large-scale system characteristics such as the trace-gas composition of the global atmosphere, the sediment loads of waters at the mouths of major rivers, contamination of groundwater in aquifers, and other phenomena of vital importance to humans, are greatly influenced by patch-, flowpath- and landscape-scale ecosystem behaviour. As scientists, we need a means of organizing our knowledge into comprehensible forms. I contend that a working acceptance of a holistic systems approach to the evaluation of regional and global problems is necessary for this integration. A hierarchical view of the world is probably necessary. In fact, we probably will need to develop the ability to envision several levels of any hierarchy simultaneously. Rigorous philosophical structures such as 'hierarchy theory' should prove valuable (O'Neill, 1988).
Regardless of the intellectual approach, integration of knowledge across disciplines will be vital because no single individual or discipline holds the key to understanding regional- and global-scale behaviour. Our disciplines have evolved to universally sophisticated states using advanced models of reality, state-of-the-art chemical analyses, and remote-sensing capabilities; and, indeed, most have adopted systems analysis as a powerful research approach. However, most of this evolution has taken place within the disciplines themselves. Our next step must be to cross disciplinary lines and describe the world as being organized around problems rather than disciplinary viewpoints.
The continuing instrumentation, analytical and computer revolution has enhanced our ability to acquire, manage, manipulate and report data at tremendous rates. Unfortunately, researchers often develop unique data management facilities, thus limiting exchange. As our conceptual models of large-scale ecosystems become more reliable, and as we gain the wisdom to evaluate the environment in the context of human needs, our ability to meaningfully analyze data and formulate information should make great progress, especially if exchange is made easier.
The need for well-conceived and tightly focused long-term research addressing transient phenomena is beginning to be recognized as vital in evaluating ecological systems behaviour (Callahan, 1984). In the past, long-term data gathering or monitoring was viewed with skepticism by many scientists and funding agencies alike. There are many reasons for this skepticism, some reasonable and some not. Regardless of past naiveté, or even abuses and neglect of records, we must establish sites and networks of sites to evaluate long-term trends of ecological phenomena. Without these evaluations we will have no means of placing our ephermeral observations in historical context.
Institutions ranging from large, international and national funding agencies to academic departments and small research units are ill-equipped to organize cross-disciplinary research, even within a single scale of resolution. Many current and future problems of urgent human concern will require not only cross-disciplinary research team building to deal with small-scale concerns, but also research that integrates knowledge from the patch to the global scale. Among the most important problems facing the research community will be the education, legitimization, funding, and rewarding of scientists who devote their energies to research of global change and problem solving.
Educational institutions will need to expand traditional views of disciplinary graduate training to provide the curricula and research opportunities for students to develop broad systems-oriented backgrounds. Additionally, the traditional view of graduate students will need to be expanded to include post graduate retraining in a science that integrates across disciplines.
Academic, public, and private employing institutions need to openly accept participation in team-oriented, integrative research as a legitimate activity at critical junctures during career development, such as promotion, tenure evaluation, and pay rises. Dogmatic adherence to the traditional review criteria of single- or dual-authored papers published in disciplinary based journals, individual achievements in 'grant getting', and overemphasis on the importance of holding office in scientific organizations all may be counterproductive for the development of this new practice of science.
Funding agencies will need to develop review and funding criteria that can accommodate cross-disciplinary, multi-investigation, and, often, novel research proposals. Clearly, review of such proposals is difficult because few or no individuals will be able to review them competently in their entirety. However, review difficulty will be unacceptable as an excuse for not developing an acceptable review mechanism. Overcoming the obstacles mentioned above will not be easy; however, the current scientific community has some experience with large (e.g. IBP, IGY, WCRP, and others) programme science and imbedded in that experience are some valuable precedents. We are now ready to build on the past and embark on a quest for knowledge about the global environment and how humans can maintain an acceptable existence on earth beyond the next decade.
Preparation of this chapter was supported by the National Science Foundation Grant #BSR-8114822 for the Central Plains Experimental Range, Long-term Ecological Research Program at Colorado State University and the Colorado State Experiment Station. I thank Dr. D. S. Schimel for his help with the literature survey. Drs D. S. Schimel, C. V. Cole, J. A. Logan and Peter Vitousek made excellent comments about the manuscript.
Bolin, B., and Cook, R. B., (1983). The Major Biogeochemical Cycles and Their Interactions. John Wiley and Sons, New York: 532 pages.
Bowden, W. B. (1986). Gaseous nitrogen emissions from undisturbed terrestrial ecosystems: An assessment of their impacts on local and global nitrogen budgets. Biogeochemistry, 2, 249-280.
Bowden, W. B., and Bormann, F. H., (1986). Transport and loss of nitrous oxide in soil water after forest clearcutting. Science 223, 867-869.
Callahan, J. T. (1984). Long-term ecological research. Bioscience, 34, 363-367.
Correll, D. E. (1983). N and P in soils and runoff in three coastal plain land uses. In Lowrance, R. R., Todd, R. L., Asmussen, L., and Leonard, R. A., (Eds.) Nutrient Cycling in Agricultural Ecosystems pp. 207-224. Spec. Publ. No.23, Univ. Georgia, College of Agric. Exp. Sta., Athens, GA.
Crutzen, P. J. (1983). Atmospheric interactions-homogeneous gas reactions of C, N and S containing compounds. In Bolin, B., and Cook, R. B. (Eds.), The Major Biogeochemical Cycles and Their Interactions. John Wiley and Sons, New York: 532 pages.
Detwiler, R. P. (1986). Land use change and the global carbon cycle: The role of tropical soils. Biogeochemistry, 2, 67-94.
DeWit, C. T. (1970). Dynamic concepts in biology. In Sellik, I. (Ed.) Prediction and Measurement of Photosynthetic Productivity, pp. 17-23, Centre for Agric. Publ. Docu. Wageningen, The Netherlands.
Forrester, J. W. (1968). Principles of Systems. Wright-Allen Press, Cambridge, MA. 400 pp.
Harwell, M. A., and Hutchinson, T. C. (1985). Environmental Consequences of Nuclear War: Ecological and Agricultural Effects. SCOPE 28. John Wiley and Sons, Chichester: 523 pages.
Houghton, R. A., Hobbie, J. E., Melillo, J. M., Moore, B., Peterson, B. J., Shaver, G. R., and Woodwell, G. M. (1983). Changes in the carbon content of terrestrial biota and soils between 1860 and 1980: A net release of CO2 to the atmosphere. Ecol. Mongr., 53, 235-262.
Hutchinson, G. L., and Viets, F. G., Jr. (1969). Nitrogen enrichment of surface water by absorption of ammonia volatilized from cattle feedlots. Science, 166, 514-515.
Olsen, R. K., Reiners, W. A., and Lovett, G. M. (1985). Trajectory analysis of forest canopy effects on chemical flux in throughfall. Biogeochemistry, 1, 361-374.
O'Neill, R. V. (1988). Hierarchy theory and global change. (Chapter 3, this volume.)
Pittock, A. B., Akerman, T. P., Crutzen, P. J., MacCracken, M. C., Shapiro, C. S., and Turco, R. P. (1986). Environmental Consequences of Nuclear War: Physical and Atmospheric Effects. SCOPE 28. John Wiley and Sons, Chichester: 359 pages.
Schimel, D. S., Coleman, D. C., and Horton, K. A., (1985a). Soil organic matter dynamics in paired rangeland and cropland toposequences in North Dakota. Geoderma, 36, (1985), 201-214.
Schimel, D. S., Stillwell; M. A., and Woodmansee, R. G. (1985b). Biogeochemistryof C, N and P in a soil catena of the shortgrass steppe. Ecology, 66, 276-282.
Schimel, D. S., Kelly, E. F., Yonker, C., and Heil, R. D. (1985c). The effects of erosional processes on nutrient cycling in semiarid landscapes. In VI International Symposium on Environmental Biogeochemistry, pp. 571-580. Van Nostrand Rein- hold, New York.
Schimel, D. S., Parton, W. J., Adamsen, F. J., Woodmansee, R. G., Senft, R. L. and Stillwell, M. A., (1986). The role of cattle in the volatile loss of nitrogen from a shortgrass steppe. Biogeochemistry, 2, 39-52.
Schindler, D. W., Turner, M. A., and Hesslein, R. H. (1985). Acidification and alkalinization of lakes by experimental addition of nitrogen compounds. Biogeo- chemistry, 1, 117-134.
Schlesinger, W. H. (1984). Soil organic matter: A source of atmospheric CO2. In Reichle, D. E. (Ed.). The Role of Terrestrial Vegetation in the Global Carbon Cycle. Springer-Verlag, Berlin and New York.
Schumm, S. A. (1977). The Pluvial System. John Wiley & Sons, New
York: 338 pages. Smil, V. (1985). Carbon-Nitrogen-Sulfur: Human Interference
in Grand Biospheric Cycles. Plenum Press, New York: 459 pages.
Strain, B. R. (1985). Physiological and ecological controls on carbon
sequestering in ecosystems. Biogeochemistry, 1, 219-232.
Strickland, T. C., and Fitzgerald, J. W., (1984). Formation and mineralization of organic sulfur in forest soils. Biogeochemistry, 1, 79-95.
West, D. C., Shugart, H. H., and Botkin, D. B., (Eds.) (1981). Forest Succession: Concepts and Application. Springer-Verlag, Berlin and New York.
Woodmansee, R. G. (1984). Comparative nutrient cycles of natural and agricultural ecosystems: A step toward principles. In Lowrance, R., Stinner, B., and House, G. J., (Eds.) Agricultural Ecosystems: Unifying Concepts, pp. 145-156. John Wiley & Sons, New York.
Woodmansee, R. G., and Adamsen, F. J., (1983). Biogeochemical cycles and ecological hierarchies. In Lowrance, R. R., Todd, R. L., Asmussen, L., and Leonard, R. A., (Eds.) Nutrient Cycling in Agricultural Ecosystems. Spec. Publ. No. 23, Univ. Georgia, College of Agric. Exp. Sta., Athens, Georgia.
|
|
|
The electronic version of this publication has been
prepared at |
