Proceedings: Alexander von Humboldt’s Natural History Legacy and Its Relevance for Today 2001 Northeastern Naturalist Special Issue 1:57-90 * Geographisches Institut, Universität Kiel, 24089 Kiel, Germany, fraenzle@geographie.uni-kiel.de. ALEXANDER VON HUMBOLDT’S HOLISTIC WORLD VIEW AND MODERN INTER- AND TRANSDISCIPLINARY ECOLOGICAL RESEARCH OTTO FRÄNZLE * ABSTRACT – The complex subject-matter of the 21st century world presents an enormous challenge to a discipline-based scientific system. Transdisciplinarity is demanded; it goes beyond interdisciplinarity and focuses both on the relevance of research to the problem at hand and on the feasibility of conducting and implementing it. Problem areas in which different sectors of society and academia may make effective contributions to include new technologies like genetic engineering, biotechnologies, energy, mobility, and nutrition; the creation, organization, and distribution of welfare and resources; human health, age, urban and regional development, and North-South cooperation; new modes of learning, new social systems and decision-making processes; and environmental issues like climate, biodiversity, soil, water, air, recycling, and waste. Real-world problems determine the kind of action to be taken and not, or at least to a distinctly lesser degree, the competence or the instruments available at any given time. Transdisciplinary research adopts an integrative approach to identifying such problems and working towards solutions. Industry, business, public administration, non-governmental organizations, and consulting firms all possess know-how which may be as important to developing new solutions as the knowledge generated and collected by universities or other scientific institutions. Thus, transdisciplinarity is a vital means of appropriately confronting many of the challenges of the present century. It also promises better and quicker solutions at lower costs, since its value lies not only in its potential for efficiently solving realworld problems but also in its ability to identify such problems at an early stage. It is the purpose of the present contribution to show, by way of example, how Alexander von Humboldt, who received his education in a time when the modern clear-cut distinction of science and art did not yet exist, aimed at an inter- and transdisciplinary comprehension of his World, and how his ideas became implemented in the second half of the past century in the form of application-oriented long-term ecosystem research. ALEXANDER VON HUMBOLDT’S HOLISTIC WORLD VIEW Considering the history of geography from its beginning in classical antiquity up to the 18th century, a series of brilliant stars stands out quite clearly: Herodotus, Strabon, Ptolemaios, Sebastian Münster, Bernhard Varenius, Anton Friedrich Büsching, Georg Forster, and Carl Ritter 58 Northeastern Naturalist Special Issue 1 (Beck 1982). When we ask what distinguished Humboldt from these master- minds one answer is clear: none of them had tried to understand the Earth or parts of it in their structural diversity and the complexity of biotic and abiotic interrelationships, nor did they include an impressive variety of societal aspects, as did Humboldt in his double capacity as an explorer and a geographer. The Roots of Alexander von Humboldt’s Holism While other contributions to the present Symposium have dealt in greater detail with Humboldt’s biography, the emphasis now is on of the mainstream of ideas which, together with contacts to leading scientific personalities of his early years of study, became the backbone of his holistic world view. Humboldt, who had been urged to study cameralistics at the Frankfurt/ Oder University, considers the visit of the young brilliant Berlin botanist Carl Ludwig Willdenow in 1788 as the decisive and most important for his whole life (Biermann and Lange 1969). The desire to accompany Willdenow on a scientific mission grew, and for preparation purposes Humboldt read “all available floristic literature on both Indies, i.e. the West and East Indies.” It should be noted in the light of present-day usage that in Humboldt’s time the notion “West Indies” meant the whole of the New World, and the tropical Americas in particular. A year later this longing for the tropics was still intensified by Georg Forster, whose acquaintance Humboldt made at Mainz in 1789, and whom he accompanied from March to July 1790 to Holland, England, and France. Back in Berlin, Humboldt managed to obtain permission for studying engineering in the oldest mining college of the world, now the Freiberg Institute of Technology in Saxony. Incorporating the previously acquired cameralistic knowledge, this paved the way to a brilliant career in the newly Prussian Margraviate of Ansbach-Bayreuth where he soon, and also owing to his nobility, became head of the mining industry, a position which he held with great professional enthusiasm and considerable social commitment. In addition he found time for scientific research (A. v. Humboldt 1797), and, what may appear more important in retrospect, for conceiving three major geoscientific research programs. The oldest program, dating from 1791/92, focused on a “History of Plants;” showing that botany has opened Humboldt’s way into the world of natural sciences, which clearly reflects the lasting influence that Willdenow had exerted. Willdenow had published an herbal book with a chapter on the history of plants, which meant in 18th century usage the attempt to trace the way of plants from their original habitat to the subsequent areas of cultivation. The second program established the corroboration of A.G. Werner’s law of strike and dip which is of lesser 2001 O. Fränzle 59 importance in the present context. The third dealt with the development of a so-called pasigraphy, i.e., a self-explanatory geological semeiotics, which led Humboldt to the fundamental distinction of hypsometric and formation maps. These programs summarize Humboldt’s most ambitious scientific aims; their importance is reflected in the fact that the pertinent publications envisioned in all of these fields led him to undertake comprehensive, comparative field research in different zones of the earth. Thus, in 1793, Humboldt started his unparalleled preparation for travel to the West Indies, which took him to northern Italy, Switzerland, and Austria, where he studied tropical plants in the greenhouses of Schönbrunn Castle, near Vienna, and met explorers who had already visited his dream-lands. Owing to his diplomatic skill he managed to obtain royal permission to explore the Hispano-American Colonies, accompanied by the talented French physician and botanist Aimé Bonpland. It is this expedition in the years 1799-1804 that personified him as an earth scientist who combined in a unique way the outstanding qualities of an enduring, daring explorer with those of a holistic geographer. Humboldt’s Physical Geography and the Sense of Unity The above characterization of Humboldt indicates that for him exploring travels and geography were two necessarily interdependent prerequisites for a deeper understanding of the world. On the one hand geographical key-notes made his expedition reports go far beyond a simple description of the routes followed. On the other, the enormous experience gained in the New World made his geography realistic and substantial. Obviously it was important for the conceptual development of Humboldt’s geography that in 1793 he had already applied systematics to his “Florae Fribergensis Specimen,” which was derived from the introduction to Immanuel Kant’s (1802, 1968) “Physical Geography,” marking an historical turning-point in so far as it moved geography from its classical incorporation into theological thought in general and physicotheology in particular (Beck 1982; Hard 1993; Vermij 1991, 1993). Here a complex organization of human experience is defined: (1) logical in the sense of Linné’s “Systema naturae” (1735), (2) historical, (3) geographical, i.e., in the sense of spatial differentiation. Kant’s classification, which implied a principal separation of geography from history irrespective of manifold links persisting, was transferred by Humboldt to the sciences of the earth, where he distinguished: (1) Physiographia (Histoire naturelle descriptive), (2) Historia telluris (Histoire du globe; closer to position 3 than 1), and (3) Geognosia (Théorie de la terre, Géographie physique, Physique du monde). Taken together these three dimensions of a comprehensive approach to a better understanding of the Earth, which combined both holistic and reduc60 Northeastern Naturalist Special Issue 1 tionist aspects and the experience gained during the six preparatory years of his expedition, brought about a geographical concentration still unparalleled many decades later. At this point, and referring to (3) above, it appears appropriate to add a few remarks on Humboldt’s equating “Geognosia” to “Erdkunde,” “Théorie de la terre” or “Géographie physique” and not simply to geography. The reason is, as an analysis of the “Relation Historique” shows (Beck 1982), that Humboldt considered himself more as a “physicien” than as a geographer in the habitual sense of the 18th century, i.e. a cartographer. Therefore, the term “physical geography” as synonymous to the terms enumerated above appeared more appropriate to describe the realm of analytical studies envisioned and the subsequent synthetic presentation of the results obtained. This type of physical geography includes consideration of the role of humans in shaping the face of the earth, which once more reflects Kant’s influence on the evolution of Humboldt’s notional concepts (Hard 1993). Moreover, it illustrates, furthermore, that his basic interests were not so much focused on the discovery of new species and the collection of biotic and abiotic specimens for banking purposes, as on the evaluative integration of known and new facts into a science which he finally no longer called “Physik” or “Theorie der Erde” but almost exclusively “physical geography.” A comparative analysis of the contemporary scientific literature indicates that notions like “Théorie de la terre” (Buffon, James Hutton, J. Forster), “Système de la nature” (Linné, Miraband), “Géographie physique” (Buffon, Buache, Desmarest, J.R. Forster, Kant), and “Philosophia naturae” (Linné, A.v. Humboldt) were common property of the scientific community. All of its leading representatives were in fact less interested in detailed generic analyses or unravelling specific microscale causalities, which became the explicit program of the socalled “organismic physics” of the 19th century (Fränzle 1998a, Markl 1995), but aimed primarily at defining interrelationships. At the same time it is an interesting illustration of continuing physico-theological thought that Linné saw many ecological relationships, e.g., food webs, in the light of Divine Providence and consequently described them as “oeconomia divina.” In the case of the majority of the above scientists, who preferred to speak of an “oeconomia naturae,” the basic world view and orientation of work was the same. Strangely enough, the fact that Humboldt began his great report on the 1799-1804 expedition with the “Essai sur la géographie des plantes” passed largely unnoticed. In the conceptual context of physical geography which had emerged out of the common ground of European Enlightenment, one reason is that Humboldt wanted to present an expression of the results obtained (Beck 1982). The other reason, equally important when speaking about Humboldt’s holistic world 2001 O. Fränzle 61 view, is deeply rooted in the fact that the 18th century had no clear distinction between art and science, as is the case in our time. On the contrary, it was a generally held belief that sciences were to reveal new phenomena to the artist, and to painters in particular, who were attributed with the ability to bring about the best and immediately comprehensible representation of nature (Trepl 1987). In the realm of science this unifying world view led to an emphasis on physiognomy, in particular with regard to man, plants, landscapes, and not least in landscape gardening (e.g., Bonnet 1764, Chambers 1772, Dezallier d’Argenville 1760, Hirschfeld 1779-85, Lavater 1775-78; see also Goethe’s “Urpflanze” and “Urphänomene” 1949). Physiognomy became the model concept for proceeding from the visible qualities of entities to their very essence, i.e., the “substantia” in the sense of scholastic philosophy. Embedded into this intellectual and methodological background, Humboldt’s “Physiognomik der Gewächse” (1806), aiming at the determination of the reciprocal interrelationships between site qualities and the “physiognomy” of individual plants, stands, or vegetation complexes, is rightly considered the starting point of an ecologically oriented plant geography. In view of the farreaching influence of this masterpiece, it appears that the background to Humboldt’s decisive break with the predominantly inventorying type of 18th century natural history, as reflected in the “Ideen zu einer Geographie der Pflanzen” (Humboldt 1807), is basically an estheticoemotional one. This is best illustrated by a literal quotation from the famous work Cosmos (1848, p. 22): “General views lead us habitually to consider each organism as a part of the entire creation, and to recognise in the plant or the animal, not merely an isolated species, but a form linked in the chain of being to other forms either living or extinct. They aid us in comprehending the relations that exist between the most recent discoveries and those which have prepared the way for them. Although fixed to one point of space, we eagerly grasp at a knowledge of that which has been observed in different and far distant regions.” This sentence reflects Humboldt’s conviction that only by means of a sufficiently differentiated analysis of the nested structure of reality which has to be coupled with a synthesizing “word picture,” the type of comprehensive world view could be achieved which is necessary for widening the intellectual horizon of man. In this context, a misunderstanding, or rather a spurious allegation, should be mentioned which concerns the term “landscape.” In geographical reasoning, where this term is frequently considered peculiar to defining the principal objective of analysis, landscape is not infrequently equated to the “total character of a region,” ascribed to Humboldt. The analysis of Humboldt’s works shows, however, that he only refers to the “global impact” that an observer has of a region: “for 62 Northeastern Naturalist Special Issue 1 appropriately describing the physiognomy of the vegetation cover of a country it is not necessary to specifically refer to taxonomic criteria (as is the rule in botany) but focus has to be on what determines the global aspect (Totaleindruck) of a region. The painter, whose artistic sentiment is of prime importance in this connection, distinguishes in his works, for instance, pine trees or palm stands from beech stands but not the latter from other deciduous species” (translated from p. 207 in: Zaunick 1958; emphasis by the author). The impressiveness of Humboldt’s ideas, in particular the “word pictures” he painted of his response to the tropical landscapes in South America, also had a tangible impact across the North Atlantic. His works, particularly the English translation of Cosmos that appeared in 1852, were widely read by many of the prominent writers of midcentury North America, such as Emerson, Thoreau, and the zoologist Agassiz (Edwards 1999). Humboldt’s emphatic exhortation to artists to visit the tropics and bring his perception of the grandeur of tropical life to a northern public bore fruit in the work of the great landscape painter Frederic Edwin Church. He did follow Humboldt’s footsteps and brought to an eager public the very unity in diversity of the tropics which may be seen as the fundamental message of Humboldt’s works. Church’s epic paintings such as “Heart of the Andes” or “Cotopaxi” had an extraordinary impact on the thousands who flocked to view them. Other landscape painters such as Bierstadt and Morau did the same for the great landscapes of the North American West. They contributed considerably to arousing public awareness of these marvels of nature, which in turn played its part in the politics leading to the creation of the first national parks in the world, in North America. Thus, Yellowstone, Sequoia, and Yosemite National Parks owe their preservation at least in part to the impact of Humboldt and his writings. In summary, it can be said that Humboldt’s enthusiastic zeal for a comprehensive understanding of nature by way of comparative analyses of its complex structures with their bewildering number of static and functional interrelationships was deeply rooted in his artistic genius. Science appears as a necessary element of aesthetics, and thus the demonstration of causalities or propensities by means of pertinent indicative phenomena was seen analogous to the revelation of the spiritual substance of a material entity. In the sense of Popper’s (1959) wellknown notional distinction, Humboldt appears more as a representative of the first type of holism, i.e., the artist’s view of the world. However, when analyzing interrelationships in the sense of functional networks, Humboldt basically tried to practice a holism of the second type, which established the basis for modern science in general and ecology in particular. It is characterized by defining the objectives of research in terms of open systems, i.e., sets of elements in dynamic interaction, 2001 O. Fränzle 63 exchanging energy, matter, and information with their environment, presenting import and export, building-up and breaking-down of their material components (Bertalanffy 1968). ECOSYSTEM STUDIES AND THEORY BUILDING Of Humboldt’s many profound legacies to science and art, perhaps the greatest was his sense of unity within the complexity of what we now call ecosystems. Clearly the ecosystem concept has been fundamental in the development of modern ecology, although its history shows that ecologists did not share a common understanding of this concept (Fränzle 1998, Golley 1993, Jax et al. 1992, McIntosh, 1985). Definitions involve basic assumptions on the “nature” or properties of ecosystems which occur in different combinations and thus account for differences among the definitions: • ecosystems are1 delimitable in space and time (e.g., Odum 1971, Rowe 1961, Streit 1980, Stöcker 1979) • ecosystems are mental constructs of the observer (e.g., Stöcker 1979, Tansley 1935) • ecosystems are real entities of nature (e.g., Leser 1984, Odum 1971) • ecosystems consist of interdependent components such as organisms and abiotic elements (e.g., Ellenberg et al. 1986, O’Neill et al. 1986) • ecosystems are in a state of dynamic equilibrium or evolve towards equilibrium (e.g., Jørgensen et al. 1992) • ecosystems are self-regulating (cybernetic) systems (e.g., Odum 1981, Patten and Salthe 1985) The purpose of this section is to outline the role of ecosystem research in the sequential refinement of ecosystem theories as reflected in model building. It ranges from empirical models for practical purposes to rather abstract ones, aiming at qualitative general insights. At one end of this spectrum are detailed and mechanistic descriptions of specific systems such as soil horizons or precisely defined soils interacting with one or a few pure chemicals in aqueous solutions. At the opposite end of the spectrum are relatively general models which must sacrifice numerical precision for the sake of general principles. Such conceptual models need not correspond in detail to any single “real world” process, but aim to provide a framework for the discussion of broad classes of phenomena or simply of contentious issues. Rationally handled, these different approaches mutually reinforce each other, thus providing reciprocally new and deeper insights. To illustrate this, an exemplary summary of ecosystem research orientations is presented in the framework of the International Biological Program (IBP) and UNESCO’s Man and the Biosphere Program 1 for the sake of brevity “are” always stands for “are considered as being...” 64 Northeastern Naturalist Special Issue 1 (MAB). This is followed by a detailed description of Ecosystem Research in the Bornhöved Lake District in Germany as a MAB pilot project. Next, focal aspects of current theoretical reflections on evolution, structure, and stability of ecosystems are demonstrated on the basis of inquiries into tropical rainforest stands, which had been a favorite topic of Humboldt’s research. Ecosystem Research in the Framework of IBP and MAB Programs The history of systems ecology illustrates the emergence of favorite orientations and foci of ecological interest as the result of an increasingly intensive feedback between theoretical reflection or modelling and theory-based interdisciplinary ecosystem research. By the mid- 1960s teams of researchers studied whole forests or lakes, while individual scientists focused on processes within systems, such as rates of primary production, material and energy transfers between trophic levels and populations, and the ways and rates of organic decomposition. Since only a largely unorganized body of theory was available at the time to stimulate research, ecosystems were viewed from a variety of perspectives, and frequently researchers reasoned analogically from physical, chemical, or biological systems to ecosystems (Fränzle 1998a, Golley 1993, McIntosh 1985). Thus, ecosystem theory was constructed from field natural history (Trepl 1987) or the type of physical geography conceived by Humboldt, from thermodynamics, from evolutionary theory, from information theory, and so on. As a consequence, part of the ecological community considered systems theory as the most appropriate paradigm to organize the information about ecosystems, since this approach provides definitions and general rules which allow very complex structures to be understood and predicted (Bertalanffy 1968). When allied to mathematical modeling techniques, system theory provides the conceptual framework for a highly effective general approach to the study of ecosystems. It ensues from the preceding summary of assumptions underlying definitions of ecosystems that all of them contain, at least implicitly, certain definitions of structure, order, complexity, or the like. These definitions presuppose, therefore, the existence of interrelated parts or system variables. Parts or elements can be defined qualitatively, variables quantitatively. Since a system is also characterized by function or activity, it is possible to ascribe function to the elements as well as to the relations connecting them; function is thus channeled and guided by the relations. By depending on these relationships (permanent or transient, weak or strong, etc.), certain alternatives of activity can be defined in a space of possibilities and actualities; this involves constraints (e.g., boundary conditions) that must become or be made manifest. Thus an Sstructure, dealing with the invariant couplings between the elements 2001 O. Fränzle 65 may be discerned from the P-structure, denoting the temporal behaviour or activity (Locker 1973). This problem of identifying parameters must be subsumed under the more general task of recognizing systems organization which is created by additional specifications, including recourse to specific theories of biology, geosciences, etc. It has been proposed, quite correctly and not only on the basis of a conventionalist attitude to reality, that order is a projection of the observer and its extent evidently depends on the subtlety of the description applied (Bennett and Chorley 1978, Waddington 1977). The balance between simplicity and complexity which would be both epistomologically trivial in their extreme forms, is determined by optimization considerations (e.g., with regard to stability or resilience problems). In regard to the former, a distinction between local, global, and structural stability is indicated. Local stability, or stability in the vicinity of an equilibrium point, describes the tendency of a community to return to its original state when subjected to a minor perturbation. Global stability describes this tendency when the community is subjected to a major disturbance. This involves an appropriate recognition of the fact that real-world environments are uncertain and stochastic, which means that the corresponding environmental parameters in the model equations exhibit random fluctuations (Begon et al. 1990, Gigon 1983, van der Maarel 1976). This leads to a yet more general meaning of stability, termed structural stability, which refers to the qualitative effects upon solutions of the model equations which are due to gradual variations in the model parameters. Thus, a system may be considered structurally stable if these solutions change in a continuous manner. Conversely, a system is structurally unstable if gradual changes in the system parameters (e.g., alterations in site factors of a community) produce qualitatively discontinuous effects (Fränzle 1993, Jørgensen 1990, van der Maarel 1976). The International Biological Program In practice, the application of the systems paradigm resulted in two complementary approaches, the first of which was characterized by ecosystem modeling from information about system components and linkages. The other, pioneered by Odum (1957) at Silver Springs, considered the ecosystem from the perspective of an object of research whose input-output relationships had to be determined in order to mechanistically explain the conversion of inputs into outputs by the system. The opportunity to test and further develop these alternatives was the primary research area of the International Biological Program (IBP) which had an overall focus on biological productivity as a basis for human well-being. Ultimately, it developed into a largely ecological 66 Northeastern Naturalist Special Issue 1 program, as the scientific director of the IBP, Worthington (1975, p. 64), stated in the synthesis volume: “It is this ecosystem approach which distinguishes much of the IBP research from what had dominated ecology before. Essentially it consists of the careful selection of a number of variables — biological, chemical and physical — about which data are collected, quantitatively as well as qualitatively. Thereby the ecosystem can be analyzed in order to ascertain which factors and processes are important in causing the dynamics of the whole. In this, the application of system analysis to biological systems has been one of the major innovations developed during IBP.” As regards terrestrial productivity, in the U.S.A. the program proposal was to study “landscapes as ecosystems,” with emphasis on production and trophic structure, energy flow pathways, limiting factors, biogeochemical cycling, and species diversity. An important feature was not to confine these studies to natural areas only; another was to use systems analysis as a mechanism for integrating the results of a study. In a comparative evaluation of the pertinent biome projects organized in the grasslands, tundra, deserts, coniferous forests, and deciduous forests, a critical American ecologist came to the conclusion that they furthered ecological knowledge but failed to contribute to the development of ecosystem theory: “The programs were not designed to sort out competing or contradictory ideas. Rather, they were driven, at least initially, by the idea that ecologists could construct a mechanical systems model built on the concepts of trophic levels, the food web, or the food cycle, and then represent the dynamic behaviour of the components by data from organisms or populations that are surrogates of the component. This ‘bottom-up’ or ‘design-up’ approach did not prove possible or useful. Further, the biome projects did not effectively promote landscape ecology, as Odum had hoped. The biome was the setting for site research but was not really addressed as such in an effective manner” (Golley 1993, p. 139). In comparison to the U.S. biome studies, the German IBP project, located in the Solling Mountains, was organized in a way similar to the Hubbard Brook project, located in the White Mountains of New Hampshire (Likens et al. 1977). Both approached ecosystem analyses from the components which could then be linked together systematically in a model-based theory or as a natural object (in Popper’s 1959 sense) studied by means of conventional scientific methods. Like the Hubbard Brook investigations, the Solling project also took a landscape approach from the beginning, based on an apriori ecological knowledge of the area which probably exceeded that in most other places where IBP research was undertaken. It focussed in a comparative way on natural-like ecosystems, nature-resembling ecosystems, 2001 O. Fränzle 67 transformed ecosystems, and degraded systems. To this end, the study sites included acidophilous beech forest and planted spruce forest stands at several different ages, along with permanent grassland and cultivated fields. Research proceeded from a description of climate and soils and the abundance and productivity of vegetation, animals, and microorganisms, to plant physiology, nutrient fluxes, and energetics (Ellenberg et al. 1986). In comparison to the majority of national IBP programs which focused on a single question or a few questions at best, and in comparison to studies undertaken in technologically advanced countries which mostly dealt with natural ecosystems, the Solling project was exceptionally successful. It provided a wealth of sound data on ecosystem structure and function, thus building the scientific concept of the ecosystem. Furthermore, it formed a promising basis to reason about the causes of forest dieback which was considered a serious problem in Germany in the 1980s. Unlike the biome programs of the United States, however, the final summary report of the Solling project did not attempt to force the results into a single synthesis on the basis of an abstract theoretical device or a model. Instead, each part was placed with a site-specific conceptual ecosystem model and developed a theme within its own logic. Owing to this methodology, the Solling project is rightly considered a milestone in the development of a theory-based ecosystem research. Both the conceptual framework and the practical experience gained in interdisciplinary research exerted a considerable influence on the conception of a comprehensive ecological surveillance system for Germany (Ellenberg et al. 1978). Composed of three interrelated components, namely an ecological monitoring network, comparative ecosystem research, and an environmental specimen bank, it is intended and largely implemented in Germany to promote both theoretical ecology and, in a transdisciplinary context, planning and policy. The integrative ecosystem research component of this German program has become an essential part of UNESCO’s Man and the Biosphere (MAB) Program which was developed out of the IBP experience. In the following section this will be considered in regard to its bearing on ecosystem research, theory, and modeling. Man and the Biosphere (MAB) Program The general objective of the MAB program, as launched by UNESCO in November 1971, endorsed by the UN Conference on the Human Environment (1972), and supplemented in 1986 and 1992, has been defined as: “... to develop within the natural and social sciences a basis for the rational use and conservation of the resources of the biosphere and for the improvement of the relationship between man and 68 Northeastern Naturalist Special Issue 1 the environment; to predict the consequences of today’s actions on tomorrow’s world and thereby to increase man’s ability to manage efficiently the natural resources of the biosphere” (Unesco 1998). The specific aims of the program are: • To assess the changes within ecosystems resulting from man’s activities and the effects of these changes on man. • To study and compare the structure, functioning, and dynamics of natural and modified ecosystems. • To study and compare the dynamic inter-relationships between “natural” ecosystems and socio-economic processes and especially the impact of changes in human populations, settlement patterns, and technology on these systems. • To define scientific criteria as a basis for rational management of natural resources. • To establish standard methods for acquiring and processing environmental data. • To promote the development of simulation and other techniques of prediction as tools for environmental management. • To foster environmental education in its broadest sense and encourage the idea of man’s responsibility for and personal fulfilment in partnership with nature. The MAB Program consists of fourteen international project areas which include the main ecological systems and physiographical units: tropical forests, mediterranean-type and temperate forests, grazing lands (savanna, grasslands, etc.), arid and semi-arid zones, lakes, marshes, rivers, deltas, estuaries and coastal zones, mountain and tundra lands, and island ecosystems. Man-made as opposed to natural ecosystems and man’s use or abuse of energy are also covered, as are four major fields of human activity or interaction with the biosphere: conservation of natural areas, effects of pesticides and fertilizers, major engineering works, and genetic and demographic changes. Ecosystem Research in the Bornhöved Lake District In the sense of methodology by example, a short description of the MAB pilot project “Ecosystem Research in the Bornhöved Lake District (Northern Germany)” illustrates the relevant approaches in the framework of this international program. The Bornhöved Lake District (Fig. 1), selected by means of comparative geostatistics as a core area of the German ecosystem research network (Fränzle et al. 1987), is situated some 30 km south of Kiel (Schleswig-Holstein) and covers about 50 km2 in terms of interrelated drainage basins. It is characterized by moraines of the Weichselian Pleniglacial with intercalated outwash sediments including kame deposits, dead-ice depressions with lakes, fens or wetlands, and Late Saalian moraines in the south. Thus, the assemblage combines the essential 2001 O. Fränzle 69 geomorphic features of northern Germany, which ensures a high degree of spatial representation in the research results obtained, which is essential for purposes of extrapolation (Fig. 2). Owing to the complicated geomorphic history (Fränzle 1981a, Garniel 1991, Piotrowski 1991, Stephan and Menke 1977), a great variety of natural ecosystems has evolved in the course of the last 12,000 years. Human impact began in the Late Mesolithic, reached a first maximum in the first millenium B.C., was minimum in the 7th, 8th, and 14th centuries A.D., and has increased thereafter to its present-day maximum (Fränzle 1990, Garbe-Schönberg et al. 1998, Schleuss 1992, Scholle and Schrautzer 1993). Thus, these ecosystems have faced longand short-term natural and anthropogenic changes which are quantitatively identified, on a comparative basis, in terms of basic processes Figure 1. Core areas of the German MAB Program. 70 Northeastern Naturalist Special Issue 1 such as energy, water, and nutrient cycles in order to improve ecosystem management. This involves cultivation experiments on arable land and pastures to define more precisely their susceptibility to different management practices and environmental chemical impact, comparable investigations on forest ecosystems and lakes, and the assessment of agricultural practices with regard to eutrophication and groundwater resources management. In this context, inquiries into the specific role of ecotones are particularly important. More specifically they relate to exchange and filter processes and the influence of ecotone structures on both stability and resilience of the adjacent aquatic or terrestrial ecosystems, respectively (Fig 3). In the framework of the socio-economic orientation of the Bornhöved Project, questions of environmental perception and economic expectations among farmers and fishermen, and the resultant private and public investment behaviour in the light of both national and international (EU) marketing policies, are also subject to assessment and modeling approaches (Fig. 4a,b). Figure 2. Saalian and Weichselian Glaciations in northern Germany (adapted from Ehlers 1983). 2001 O. Fränzle 71 Thus the general aims of the Bornhöved Project are: • defining and modelling structure, dynamics, and stability conditions of the interrelated terrestrial and aquatic ecosystems of the study area in terms of site characteristics and biocenotic diversity, natural and anthropogenic fluxes of energy and matter, productivity, and land use patterns; • determining and modeling environmental strains and resilience mechanisms of the ecosystem compartments affected by disturbances with particular reference to chemical impacts. In view of the specific structure of the study area, particular objectives of research are: • modeling of biotic, energetic, and material exchange processes between neighbouring ecosystems of different land-use and fishery patterns; • modeling site-dependent relationships between lakes and their drainage basins with focus on the role of different land/water interfaces; • modeling of agro and pasture ecosystems in the light of national and international production and marketing regulations; • ecotoxicological research on the fate and behaviour of environmental chemicals; • inquiries into the efficiency of environmental protection and conservation measures; • testing the validity of spatial extrapolations of simulation models by means of comparative site analyses and geographic information systems; Figure 3. Model of an ecosystem (adapted from Ellenberg et al. 1986). 72 Northeastern Naturalist Special Issue 1 • the paleoecological and historical reconstruction of ecosystem evolution in the study area since the end of the Weichselian Glaciation. In greater detail, these objectives are defined in a hierarchically structured system of several hundred working hypotheses on five levels of increasing generality which form the conceptual basis of interdisciplinary cooperation. It has grouped about 140 scientists from the following disciplines: agricultural management, bacteriology, chemistry, climatology, computer science, ecotoxicology, geobotany, geomorphology, human geography, hydrogeology, ichthyology, limnology, mathematics, meteorology, paleobotany, paleozoology, physical geography, physics, planktology, plant nutrition, plant physiology, soil microbiology, soil science, toxicology, and zoology (Fränzle 1990, 1998a; Müller 1992; Müller and Windhorst 1991). A particularly important type of change studied in the Bornhöved Project is adaptation, and the influence eliciting it can generally be considered as a stimulus and the adaptive process as a response, i.e., a Figure 4 a. Summary of ecological modelling approaches. 2001 O. Fränzle 73 stabilization. In greater detail, the behaviour of an adapting system can be classified into (1) goalseeking, (2) purposive, and (3) purposeful. The first adaptative mechanism consists in a one-to-one correspondence between stimulus and response, while a purposive system displays a one-to-many correspondence, which means that each stimulus normally elicits a number of responses. The term “purposeful” finally indicates an adaptive mechanism which brings about a change in the system’s functions. Since the system achieves, during the process of adaptation, a new kind of structural or functional order or organization, it undergoes selforganization. This process includes not only adaptation per se, but also cognition and learning with regard to constraints involving energy and matter. The exergy concept (Jørgensen 1992) is one basis of current attempts to develop thermodynamically founded models of ecosystem functioning which may, for example, make it possible to predict how the biota of an ecosystem might respond to specific environmental changes. Depending on the ratio of energy inflow to outflow, a nucleation (i.e., a clustering or aggregation of system elements) may occur. In these clusters, or regions in geographical terminology, a conversion of energy into signals is carried out, whereby the activity of the energy converter is decisive for the signal connectivity which in turn defines the interactions between elements. This is a principal statement connecting thermodynamics with information theory (e.g., Patten and Jørgensen 1995). Figure 4b. Spatio-temporal scales of models used in the framework of the Bornhöved Ecosystem Research Project (Schleswig-Holstein). 74 Northeastern Naturalist Special Issue 1 Because the regions are characterized by their states, the process of self-organization can be considered as a transition between these states, undergoing new formations between the elements, provided the stimuli exceed certain threshold values. In addition to cybernetics (e.g., de Angelis 1995, Straskraba 1995), network theory (Higashi and Burns 1991, Patten 1992), catastrophe theory and bifurcation analysis have proven useful for formalized descriptions of such systems in transition. Catastrophe theory limits systems of interest to so-called gradient systems which arise from the minimization of some objective function and associated dynamics (or maximization of its negative). The latter kind of branching behaviour is relevant to systems of a more general type than the first, which cannot be appropriately characterized by the gradient of a potential function. Typically, differential equations of non-gradient systems have a small number of isolated stable equilibrium points and information about system behavior is presented as trajectories on state space (or phase) diagrams. The stable points act as attractors, and correspondingly unstable points as repellors, and these points shape the contours of the trajectories in state space accordingly. Bifurcation is reflected in the existence of critical parameter values at which the nature of the solution to the differential equation changes. However, there is no simple classification of possible “cases” as in catastrophe theory, and empirical inquiry into systems behaviour is essential to proceed in more precisely determining the realm of applicability of bifurcation methodology to generalized ecosystem analysis and modeling. Inquiries into self-organization processes involve comparative analyses of community and ecosystem succession and the chronosequential approach, or space-for-time substitution, is a traditional tactic applied. As a consequence of the probabilistic character of living systems, this approach may be misleading. Consequently, longterm studies were recognized by founders of ecology to be necessary for a reproducible understanding of succession (see Clements 1916). Despite the failings of vegetational or soil chronosequences, their value is clear if certain limits are kept in mind. In the framework of system evolutionary analyses of holarctic and tropical plant associations, space-for-time substitution has been employed to assess structural or compositional aspects (Fränzle 1994). The same applies to old-field successional studies which document trends in life history types, pathways of dominant species, convenient “stages,” and regional differences (Pickett 1989a). Many of these insights could, in principle, have come from long-term studies. However, the understanding that has emerged from the few long-term investigations of old-field succession is of a different sort, since they documented the nature of transitions, the role of year-specific conditions, the problems with end points, and the role of newly invaded species in succession (Asshoff 1997, Bobrowski 1982, Hemprich 1991). 2001 O. Fränzle 75 Biodiversity Analyses of Tropical and Holarctic Plant Communities An important issue of the self-organization paradigm is how biodiversity relates to community structure and which community-level properties emerge from the disparate interactions of organisms, populations, and site qualities. Possible examples of such “emergent properties” include trophic and guild structures, stability, resilience, and successional stages. A generalized offshoot of the concept of emergent properties is the notion that biological systems are hierarchically organized, with new properties at each level of the hierarchy (Allen and Starr 1982, O’Neill et al. 1986, Solbrig and Nicolis 1991). According to this view, diversity is better understood if ecosystems or biota are decomposed hierarchically so that each process can be viewed as a stabilizing or disruptive factor at each level in a hierarchy of time and space scales. Such hierarchical organization can be easily visualized or defined for the level of molecules through organisms where the hierarchies are nested, i.e., all lower-level hierarchies are included in one higher category. Species, however, not infrequently are part of more than one higher community or ecosystem. Thus, a mutation may be detrimental to the species living in one (adverse) environment but not when it lives in another. Therefore, the validity of hierarchy theory at the community and ecosystem levels is more controversial than at distinctly lower levels of organization. The following section on biodiversity analyses of tropical and ectropical plant communities is an attempt to combine viewpoints of hierarchy theory and thermodynamics of open systems, which leads to the following hypotheses: • Species diversity of climactic lowland rainforests is, ceteris paribus, for negentropic reasons inversely related to the nutrient content of soil. • Under conditions of long-lasting climatic and geomorphic stability ecosystem complexity evolves in a non-monotonous way with the nutrient status of soil. If soils form on basic or intermediate parent materials, pedogenic nutrient supply and species diversity may both increase during periods of 1 - 10 (possibly 100) ka, provided the water and energy factors are not limiting. Shorter fluctuations of variable but generally decreasing amplitude are likely to be superimposed on this long-term trend. In the following period of evolution, which can last several hundred or even thousand ka, soils degrade in regard to nutrient supply but species diversity keeps increasing for negentropic reasons. Biodiversity and Thermodynamics of Tropical Rainforests The verification of the first hypothesis is based on the comparative analysis of diversity and abundance spectra of neo- and paleotropical rainforest associations and their related soils. McIntosh’s (1967) diver76 Northeastern Naturalist Special Issue 1 sity index is applied, as it appears better suited for comparison with the literature. The indices of stands 30-62 are derived from primary data kindly provided by Professor Brünig (Hamburg), the others are based on species lists from Aubréville (1961), Davis and Richards (1933, 1934), and Richards (1963). In order to better compare with similar figures in the literature, it is noted that values in the abscissa are standardized by introducing a relative abundance parameter Â, defined as number of species per 100 individual plants (Fig. 5). Figure 5 clearly shows that in most cases rainforest stands on highly nutrient-depleted ferralsols, acrisols, and podzols of tropical lowlands attain diversity indices above 90% of the theoretical maximum. However, further differentiation according to local nutrient status and water budget is not possible, given the available data. Stand 38 (Borneo) is of particular interest since it represents a forest which developed on a tropical bog. Its exceptionally low index value allows the conjecture that its nutrient supply has fallen below the critical value defined below. On a very general level, the apparently unusual inverse relationship between nutrient status of soils and diversity of related plant communities may be interpreted in terms of biological thermodynamics (Prigogine 1976). Biotic communities are open systems in exchange of energy and matter with their environment; their entropy production, dS, comprises the terms diS and deS, where diS denotes the entropy produc- Figure 5. Diversity indices of paleo- and neotropical rainforest stands. 2001 O. Fränzle 77 tion within the system while deS is a flux term describing entropy “export” into the environment: dS = diS + deS. (1) Only diS > 0, but deS can be negative. By identifying entropy with disorder, it follows from Eq. 1 that an isolated system can only evolve towards greater disorder. For an open system, however, the “competition” between deS and diS permits the system, subject to certain boundary conditions, to adopt new states or structures. These are stationary if dS = 0 (2.1) or deS = -diS < 0 , (2.2) respectively. diS can be expressed in terms of thermodynamic “forces,” Xi, and rates of irreversible phenomena, Ji (Prigogine 1976). Xi may be gradients of temperature or concentration; the corresponding “rates” are then heat flux and chemical reaction rate. Hence, diS n (3) — = Σ JiXi dt i=1 Provided the reservoirs of energy and matter in the environment of the open system are sufficiently large to remain essentially unchanged, the system can tend to a non-equilibrium stationary state far beyond the domain of linear thermodynamics. This state may be associated with dissipative structures (Glansdorff and Prigogine 1971), i.e., structures resulting from a dissipation of energy rather than from conservative molecular forces. Considering phytocenoses from the point of view of stationary dissipative structures, the relationship of diS and deS as expressed in Eq. 2 and the specific boundary conditions controlling entropy production and flux rates appear to be particularly important. It is a consequence of Eq. 2.2 that, thermodynamically speaking, stability or the capacity to maintain a non-equilibrium steady state is coupled with a (relative) minimum of total entropy production dS. Clearly this can be accomplished by either minimizing diS or maximizing deS, or a combination of both strategies. Concentration processes involved in the normal metabolic activities of living systems play an important role in this connection, as can be seen from the following equation. C2 ΔG° = R·T·ln — , (4) C1 78 Northeastern Naturalist Special Issue 1 where ΔG° = difference in standard free energy, R = 8.31 J· mol-1 · K-1, T = temperature in Kelvin, and C2, C1 = higher or lower thermodynamic concentrations, respectively. Changes in concentration are a physical prerequisite for the production of a great many compounds, and an absolutely cogent one if substances are produced whose free energy is higher than that of the corresponding “raw materials.” In Amazonia, for example, forest stands with their mosaic-like structure owing to the patchiness of natural rejuvenation processes, developed specific adaptations facilitating these concentration processes. A dense root mat is formed over the soil in intimate contact with litter, and root tips growing upwards into the fallen litter (Klinge 1983). Mycorrhizal associations, as the most common plant root-soil microorganism symbioses, benefit their host plants by effectively increasing the absorptive surface area of the root system, thus providing for increased uptake of nutrients (Harley and Smith 1983), and by biological nitrogen fixation in the root-humus-soil interface (Dommergues et al. 1985; Herrera et al. 1981; Fittkau 1983, 1991). The dense fabric of predominantly fine roots also plays an important part in exchange and adsorption of nutrients from throughfall water. In addition, the structure of the foliage favors the use of nutrients by a long active life, the retransport of certain nutrients before leaf shedding, and a high polyphenol content and coriaceous nature which reduce herbivory. Another factor of high importance is the multilayered structure of the forest and the activities of epiphytes and microorganisms on much of the exposed surfaces, which together form highly efficient filtering systems scavenging nutrients from rainwater. Research in New Guinea and the Ivory Coast has shown that in the rainforests approximately two-thirds of the potassium input of about 100 kg / (hectare * year) and one-third of the magnesium input of around 30 kg / (hectare * year) came from rainfall and leaching from tree crowns. The remainder was due to litter fall (Schultz 1988). The effectiveness of these negentropic processes is further enhanced by most efficient entropy fluxes related to the transpiration and nocturnal respiration of plants. The molal entropy of H2O increases from 63 J x mol-1 x K-1 (liquid) to 189 J x mol-1 x K-1 (gas) in the course of evaporation, and CO2 has a molal entropy of no less than 214 J x mol-1 xK-1. Thus, the reverse process, the photosynthetic fixation of CO2, is also of comparable importance for the negentropy balance in the light of Eqs. 2.2 and 4. Even in the simplest cells the normal metabolic pathways imply several thousand complex chemical reactions which must be coordinated by means of an extremely complicated functional network; this means that hierarchical order in both functional and spatio-temporal respect constitutes a further highly important negentropic factor. It characterizes every living system from the sub-microscopic level to 2001 O. Fränzle 79 rainforest communities like those of the Amazonian Hylaea which also comprise extremely diversified faunal assemblages (e.g., 80 ant species on one tree). The evolution of such a nested system is a multiple-level process, in which patterns generated at one level determine the evolutionary dynamics and evolutionary potential at other levels (Hogeweg 1994). It is also a particularly good example of May’s (1972) cybernetic model of community behaviour after small disturbances, which suggests that increases in species number lead to lesser stability unless compensatory decreases occur in the degree of web connectance (i.e., the fraction of all possible pairs of species interacting directly) or interaction strength (Fränzle 1998b). Summarizing, it may be said that the combination of these highly efficient negentropic community structures and processes is apt to overcompensate the high entropy status of the related soils as defined in terms of nutrient supply, which brings about the long-term stability of tropical rainforest ecosystems under the stable climatic and geomorphic boundary conditions assumed above. Species richness and site analyses of ectropical plant communities A verification of the second hypothesis involves a basically analogous approach with regard to ectropical phytocenoses as the above diversity and site analyses of tropical lowland rainforest communities. For the sake of brevity, Figure 2 summarizes the results of comprehensive cluster (centroid sorting) and regression analyses of 108 phytocenoses from northern Germany, in the form of a matrix defining the absolute frequencies of stands characterized by N ⊕ M nutrient and soil moisture contents and H ⊕ Â combinations. It shows that highly diversified plant communities of medium to low abundance (classes 12- 16) are most frequent on sites characterized by high nutrient but medium to low soil moisture contents (classes 1-5). Phytocenoses of both high diversity and abundance (i.e., classes 5-7), however, are relatively rare on sites of this type (Fränzle 1994). Thus, at the sub-continental scale, in northern Germany species diversity tends to increase with pedogenic nutrient supply, and is largely independent of differences in soil moisture regime. A comparable analysis of North American plant communities (Fränzle 1981b) permits, at the continental scale, a more precise definition of the qualitatively wellknown dependence of plant growth on the water factor. In statistical terms, the plant-available amount of soil moisture is the major controlling factor of species richness (Fig. 6). A triphase model of ecosystem evolution In conclusion, the inverse species richness/nutrient relationship of tropical and ectropical plant communities leads to the question of whether a unifying interpretation of the above results is possible. The 80 Northeastern Naturalist Special Issue 1 affirmative answer can be summarized in form of a triphase evolutionary model (Fig. 7). (1) Phase I is characterized by a parallel increase in pedogenic nutrient supply and species richness owing to accelerated immigration of plants, provided that water and energy factors are not limiting. Spatial heterogeneity of sites is reflected in a corresponding variability in numbers of species and plant individuals. With this variability, plant associations may simultaneously retain genetic and behavioural types that can maintain their existence in low populations together with others which can capitalize on opportunities for dramatic increase. Ecosystems are consequently characterized by a (comparatively) high resilience but much lesser stability. Figure 6. Absolute frequencies of clustered stand characteristics (H ⊕ Â) on different sites (N ⊕ M). 2001 O. Fränzle 81 (2) Phase II can last for several hundred or even thousand ka; and the ecosystems as complex hierarchial structures are characterized by marked soil degradation in terms of nutrient supply on the one hand, while species richness keeps increasing for negentropic reasons on the other. The more homogeneous the climatic and edaphic environment in space and time, the more likely is the macrosystem to display low overall fluctuations; i.e., its overall stability has attained a maximum, while resilience is distinctly less than during phase I. Tropical rainforests represent such buffered and (largely) self-contained systems with relatively low (natural) variability on highly nutrient depleted soils of the ferralsol, acrisol, and podzol classes. Temporary and larger-scale disturbances are likely to affect a proportionally higher number of species than in phase-I populations, where the interspecific differences in ecological valence are usually distinctly higher. At a lower level of the hierarchically differentiated analysis; i.e., in higher spatial resolution, however, the patchiness of the vegetation cover as a result of interspersed small-scale rejuvenation processes is an expression of shortterm localized instabilities out of phase, which efficiently contribute to enhance the overall long-term stability of the macrosystem. (3) Phase III marks the eventual and comparatively rapid decrease in species richness of phytocenoses, once the nutrient status of soil has fallen below a critical level which can no longer be compensated by the set of negentropic mechanisms described. Consequently plant communities may be highly resilient, although unstable. In the light of this model it seems appropriate to finally restate the three ecological principles of Thienemann (1956), two of which were first proposed in 1920 but repeatedly discovered and put forward as new ideas: • The greater the diversity of site conditions, the larger is the number of species which make up the biotic community. • The longer a site experiences the same evolutionary conditions, the richer and more stable becomes its biotic community. Figure 7. Model of pedogenic nutrient supply and species diversity in the course of ecosystem evolution. 82 Northeastern Naturalist Special Issue 1 • The more site qualities deviate from normal, and hence from the normal optima of most species, the smaller is the number of coexisting species while their abundances are proportionally increased. The evolutionary model suggested, allows for more precise definition of the realms of validity of these principles: in the above sequence of quotation they refer to phase I, II, and III or I populations, respectively. UNITY IN COMPLEXITY – QUEST FOR A UNIFIED ECOSYSTEM THEORY The analysis of real communities with their site characteristics and the comparative assessment of ecological models have provided a better insight into their bewildering complexity which clearly fascinated Humboldt. Population dynamics of real-world ecosystems are spatially distributed in general, associated with a marked temporal variation, providing a multitude of ways in which the probability of coexistence is enhanced and biodiversity increased. Coexistence under stochastic, non-equilibrium conditions as described by patch-dynamics models at different scales can be just as strong and stable as that occurring under a deterministic, niche-differentiation model. In a single patch, species extinction can occur as a result of competitive exclusion, overexploitation, and other destabilizing interspecific interactions or they are due to environmental instability, e.g., unpredictable disturbances or changes in conditions. In this connection it is worth remembering that soil conditions are largely controlled by pH and redox systems. Since both are nonlinear interrelated buffering systems, soil stability with regard to sorption and mobilization processes is liable to unpredictable, chaotic changes. An important parallelism seems to exist between the properties of a community and the properties of its component populations which will be subject to a (relatively) high degree of K selection in stable environments; r selection predominates in variable environments. K-selected populations with their high competitive capacity, high inherent survivorship but low reproductive rates, are normally resistant to disturbances. However, once perturbed they have little possibility to recover, i.e., low resilience. The r-selected populations, by contrast, have less resistance but distinctly higher resilience. This inverse relationship between resistance and resilience is of particular importance in ecotoxicological respect. In conclusion it may be said that most communities are probably organized by a temporally and spatially varying mixture of “forces,” namely competition, predation, and disturbances, with competition and predation being presumably less important in more disturbed environments. Thus, there is no such thing as a single stability or sensitivity 2001 O. Fränzle 83 measure for a community. Stability varies with the aspect of the community or ecosystem under study and the nature of the disturbance. Without pertinent semantic specifications relating to defined temporal and spatial scales (duration and dimensionality of observations), indicator variables, and the nature of perturbation, statements about systems stability and sensitivity would be meaningless in the ecological as in many other respects (Fränzle 1998b). This (necessarily incomplete) summary of focal points of ecological interest may indicate that the present theoretical background to ecosystem research has not yet reached the level of a comprehensive unified theory. There exist, however, a commendable number of unifying concepts and integrative approaches with regard to ecosystem research. Here it may suffice to quote the following, which not infrequently exhibit a considerable amount of convergence. Catastrophe theory (Thom 1975, Wilson 1981) concerns both the stability and creation of forms, and each of the pertinent concepts also implies a general concern with dynamical analysis. Jørgensen (1992, 1995) introduced the exergy principle and deduced a theory of ecosystem evolution on this basis. Exergy may be associated with holistic indicator variables such as entropy, emergy, and ascendency (Jørgensen 1997 a, b). Schneider and Kay (1994 a, b) identified thermodynamic non-equilibrium, exergy gradients and flows as equally useful indicators of ecosystem functioning. Following Odum’s (1969) succession concept and touching upon cybernetic system analysis, Patten (1998) ascribed to certain state functions the intrinsic quality of attractors; in Bossle’s (1992) terminology they constitute ecological orientors or goal functions (Müller and Leupelt 1998). Epistemologically speaking, all of these concepts are derived from, or associated with, systems theories in general and, more specifically, with theories of self-organizing ecological systems. The latter, in turn, amalgamate components of the thermodynamic theory of irreversible processes (Nicolis and Prigogine 1977; Prigogine 1967, 1976, 1985) with elements of the above catastrophe theory, furthermore with deductions from the information and network theories (Margalef 1995; Ulanowicz 1986, 1995), the theory of games (McMurtrie 1975, Pimm 1982, Ulanowicz 1986), and hierarchy theory (Allen and Hoekstra 1992, Allen and Starr 1982, O’Neill et al. 1986). One of the most conspicuous representations of self-organization processes in open systems is the formation of gradients whose structural and functional analysis provides for a particularly integrative aspect in ecosystem studies (Fränzle 1994; Müller 1998; Prigogine 1976; Schneider and Kay 1994 a, b). In an open system the “competition” between internal entropy production and entropy “export” into the environment permits the system, subject to certain boundary conditions, to 84 Northeastern Naturalist Special Issue 1 adopt new states or structures. Thus, gradients with their specific coefficients, e.g., coefficients of thermal conductivity or diffusion, are related to, and an expression of, exergy flows in ecological systems. For evolving biotic networks primarily isolated notional characterizations like homogenization, amplification, synergism (Patten 1992), ascendency (Ulanowicz 1986), power, or energy (Odum 1983) can be conveniently associated with energetic or nutritional gradients. Also ecological hierarchies can, for the sake of a unifying generic characterization, be described in terms of scale-dependent hierarchies of gradients. Gradients are the most conspicuous reflection of a system’s heterogeneity which is, in the operational sense of the term, a scale-dependent outcome of the regionalization procedures adopted. The generic definition and analysis of ecologically relevant fluxes of energy, matter, and information, including the determination of thermodynamic “forces” and gradients, is therefore intimately linked with the appropriate definition of spatial structures which frequently undergo seasonal or other temporal variations. This fact assumes a particular quality when considering ecosystems or their major compartments from the viewpoint of self-organized dissipative structures (Fränzle 1994, Prigogine 1976). To most effectively maintain the whole set of intra- and intersystemic fluxes essential for the negentropy-related stability, the systems must continuously degrade and re-establish a whole network of gradients (Schneider and Kay 1994a). This implies that concentration processes are necessarily coupled with entropy-exporting dissipative processes, such that the succession of gradient formation and degradation can be characterized by a temporal cyclicity as suggested by Holling (1986). A final point to be made with regard to all concepts of ecosystem analysis and modelling is the fact that the analyst in most cases has incomplete and partial knowledge of the system which he seeks to understand and control. The nature of his knowledge is often incremental in that greater insight is gained as analysis or control continue. This is the elementary basis for claiming long-term commitments for ecological studies (Bennett and Chorley 1978, Pickett 1989b). However, even under such favourable technical boundary conditions, elements in the analysis of ecosystems will normally be necessarily indeterminate, both because of the complexity of such systems and because of the intimacy of man-environment interrelations or subjective constraints on understanding. The major limitations are due to methods or measurement, the presence of stochastic variation and the methods of system analysis adopted, and finally the manner in which systems knowledge is obtained, accumulated, and applied. Each of these points is of crucial importance, and each constitutes a major challenge to both theoretical reflection and systematic empirical research. 2001 O. Fränzle 85 LITERATURE CITED ALLEN, T.F.H., and T.W. HOEKSTRA. 1992. Toward a Unified Ecology. Columbia University Press, New York, NY. ALLEN, T.F.H., and T.B. STARR. 1982. Hierarchy: Perspectives for Ecological Complexity. University of Chicago Press, Chicago, IL. ASSHOFF, M. 1997. Die Erschließung und Modellierung ökologischen Wissens für das Management von Feuchtwiesenvegetation. Ph.D. Thesis, Universität Kiel, Kiel, Germany. AUBRÉVILLE, A. 1961. Etude écologique des principales formations végétales du Brésil. Orstrom. Nogent-sur-Marne, France. BECK, H. 1982. Große Geographen. Pioniere – Außenseiter – Gelehrte. Reimer, Berlin, Germany. BEGON, M., J.L. HARPER, and C.R. TOWNSEND. 1990. Ecology: Individuals, Populations, and Communities. Blackwell, Boston, MA. BENNET, R.J., and R.J. CHORLEY. 1978. Environmental Systems. Methuen, London, UK. BERTALANFFY, L.v. 1968. General System Theory. Braziller, New York, NY. BIERMANN, K.R., and G.R. LANGE. 1969. Alexander von Humboldts Weg zum Naturwissenschaftler und Forschungsreisenden. Alexander von Humboldt. Festschrift aus Anlaß seines 200. Geburtstages. Pp. 87-102. Akademie der Wissenschaften, Berlin, Germany. BOBROWSKI, U. 1982. Pflanzengeographische Untersuchungen der Vegetation des Bornhöveder Seengebietes auf quantitativ-soziologischer Basis. Kieler Geographische Schriften 56. 175 pp. BONNET, C. 1764. Contemplation de la nature. Rey, Amsterdam, Holland. BOSSLE, H. 1992. Real-structure process description as the basis of understanding ecosystems and their development. Ecological Modelling. 63:261-276. CHAMBERS, W. 1772. A Dissertation on Oriental Gardening. London, UK. CLEMENTS, F.E. 1916. Plant Succession: An Analysis of the Development of Vegetation. Carnegie Institute, Washington, DC. DAVIS, T.A.W., and P.W. RICHARDS. 1933. The vegetation of Moraballi Creek, British Guiana: an ecological study of a limited area of tropical rainforest. Part. I. Journal of Ecology 21:350-384. DAVIS, T.A.W., and P.W. RICHARDS. 1934. The vegetation of Moraballi Creek, British Guiana: an ecological study of a limited area of tropical rainforest. Part. II. Journal of Ecology 22:106-155. DeANGELIS, D.L. 1995. The nature and significance of feedback in ecosystems. Pp. 450-467, In B.C. Patten and S.E. Jørgensen (Eds.). Complex Ecology: The Part-whole Relation in Ecosystems. Prentice Hall. Englewood Cliffs, NJ. DEZALLIER D’ARGENVILLE, A.-J. 1760. La théorie et la pratique du jardinage. Paris, France. DOMMERGUES, Y., B. DREYFUS, H.G. DIEM, and E. DUHOUX. 1985. Fixation de l’azote et agriculture tropicale. La Recherche 162:22-31. EDWARDS, J.S. 1999. Humboldt’s South America today. Alexander von Humboldt Stiftung - Mitteilungen 73/1999:47-52. ELLENBERG, H., O. FRÄNZLE, and P. MÜLLER. 1978. Ökosystemforschung im Hinblick auf Umweltpolitik und Entwicklungsplanung. Umweltforschungsplan des Bundesministers des Innern - Ökologie-Forschungsbericht 78-101 04 005. Bonn, Germany. 86 Northeastern Naturalist Special Issue 1 ELLENBERG, H., R. MAYER, and J. SCHAUERMANN. 1986. Ökosystemforschung. Ergebnisse des Sollingprojekts 1966-1986. Ulmer, Stuttgart, Germany. FITTKAU, E.J. 1983. Flow of nutrients in a large open system: the basis of life in Amazonia. The Environmentalist 3, Supplement No. 5:41-49. FITTKAU, E.J. 1991. Tropische Regenwälder - Ökologische Zusammenhänge. Bensberger Protokolle 66:27-63. FRÄNZLE, O. 1977. Biophysical aspects of species diversity in tropical rain forest ecosystems. Biogeographica 8:69-83. The Hague, Netherlands. FRÄNZLE, O. 1981a. Erläuterungen zur Geomorphologischen Karte der Bundesrepublik Deutschland 1:25,000, GMK 25 Blatt 8, 1826 Bordesholm. Geo Center, Stuttgart, Germany. FRÄNZLE, O. 1981b. Vergleichende Untersuchungen über Struktur, Entwicklung und Standortsbedingungen von Biozönosen in den immerfeuchten Tropen und der gemässigten Zone. Aachener Geographische Arbeiten 14:167-191. FRÄNZLE, O. 1990. Ökosystemforschung und Umweltbeobachtung als Grundlagen der Raumplanung. MAB-Mitteilungen 33:26-39. FRÄNZLE, O. 1993. Contaminants in Terrestrial Environments. Springer, Berlin, Germany. FRÄNZLE, O. 1994. Thermodynamic aspects of species diversity in tropical and ectropical plant communities. Ecological Modelling 75/76:63-70. FRÄNZLE, O. 1998a. Grundlagen und Entwicklung der Ökosystemforschung. Chapt. II-2.1: 1-24, In O. Fränzle, F. Müller, and W. Schröder (Eds.). Handbuch der Umweltwissenschaften. ecomed. Landshut/Lech, Germany. FRÄNZLE, O. 1998b. Sensivity of ecosystems and ecotones. Pp. 75-115, In G. Schüürmann and B. Markert (Eds.). Ecotoxicology. John Wiley & Sons, New York, NY. FRÄNZLE, O., D. KUHNT, G. KUHNT, and R. ZÖLITZ. 1987. Auswahl der Hauptforschungsräume für das Ökosystemforschungsprogramm der Bundesrepublik Deutschland. Umwelforschungsprogramm der Bundesrepublik Deutschland. Umweltforschungsplan des BMU. Forschungsbericht 101 04 043/02. Kiel, Germany. GARBE-SCHÖNBERG, C.-D., J. WIETHOLD, D. BUTENHOFF, C. UTECH, and P. STOFFERS. 1998. Geochemical and palynological record in annually laminated sediments from Lake Belau (Schleswig-Holstein) reflecting paleoecology and human impact over 9000 a. Meyniana 50:47-70. GARNIEL, A. 1991. Weichselzeitliche Morphogenese im nördlichen Mittelholstein unter besonderer Berücksichtigung der Eisabbauvorgänge. Schriften des Naturwissenschaftlichen Vereins für Schleswig-Holstein 61:25-54. GIGON, A. 1983. Über das biologische Gleichgewicht und seine Beziehungen zur ökologishen Stabilität. Berichte des Geobotanischen Institutes ETH. Stiftung Rubel. Zürich. 50:149-177. GLANSDORFF, P., and I. PRIGOGINE. 1971. Thermodynamic Theory of Structure, Stability and Fluctuations. Wiley, New York, NY. GOETHE, J.W.v. 1949. Die Schriften zur Naturwissenschaft, hg. im Auftrag der Dt. Akademie der Naturforscher (Leopoldina) zu Halle, Germany. GOLLEY, F.B. 1993. A History of the Ecosystem Concept in Ecology. Yale University Press, New Haven, CT. 2001 O. Fränzle 87 HARD, G. 1993. Kant’s “Physische Geographie”, wiedergelesen. In Kattenstedt, H. (Hg.): «Grenz-Überschreitung». Abhandlungen zur Geschichte der Geowissenschaften und Religion/Umwelt-Forschung 9:51-72. HARLEY, J.L., and S.E. SMITH. 1983. Mycrorrhizal symbiosis. Academic Press, New York, NY. HEMPRICH, G. 1991. Landschaftsökologische Untersuchungen im Bereich des Belauer Sees und Schmalensees. Diplom Arbeit. Universität Kiel, Germany. HERRERA, R., E. MEDINA, and H. KLINGE. 1981. How human activities disturb the nutrient cycles of tropical rainforest in Amazonia. Ambio 10:109-114. HIGASHI, M., and T.P. BURNS. 1991. Theoretical Studies of Ecosystems. The Network Approach. Cambridge University Press, Cambridge, UK. HIRSCHFELD, C.C.L. 1779-85. Theorie der Gartenkunst. 5 Bde. Leipzig, Germany. HOGEWEG, P. 1994. Multilevel evolution: replicators and the evolution of diversity. Physica D 75:275-291. HOLLING, C.S. 1986. The resilience of terrestrial ecosystems: Local surprise and global change. In W.M. Clark and R.E. Munn (Eds.). Sustainable Development of the Biosphere. Oxford, UK. HUMBOLDT, A.v. 1797. Versuche über die gereizte Muskel und Nervenfaser. Posen, Berlin, Germany. HUMBOLDT, A.v. 1806. Ideen zu einer Physiognomik der Gewächse. Cotta, Tübingen, Germany. HUMBOLDT, A.v. 1807. Ideen zu einer Geographie der Pflanzen, nebst einem Naturgemälde der Tropenländer, auf Beobachtungen und Messungen gegründet [aus dem Französischen übersetzt]. Cotta, Tübingen, Germany. HUMBOLDT, A.v. 1848-58. Cosmos: A Sketch of a Physical Description of the Universe. Vol. I-V. Bohn, London UK. JAX, K., G.-P. ZAUKE, E. VARESCHI. 1992. Remarks on terminology and the description of ecological systems. Ecological Modelling 63:133-141. JØRGENSEN, S.E. 1990. Modelling in Ecotoxicology. Elsevier, Amsterdam, Holland. JØRGENSEN, S.E. 1992. Development of models able to account for changes in species composition. Ecological Modelling 62:195-208. JØRGENSEN, S.E. 1995. The growth rate of zooplankton at the edge of chaos: Ecological models. Journal of Theoretical Biology 175:13-21. JØRGENSEN, S.E. 1997a. Thermodynamik offener Systeme. Kap. III-1.6. In O. Fränzle, F. Müller, and W. Schröder (Eds.). Handbuch der Umweltwissenschaften. ecomed. Landshut/Lech, Germany. JØRGENSEN, S.E. 1997b. Möglichkeiten zur Integation verschiedener theoretischer Ansätze. Kap. III-1.8. In O. Fränzle, F. Müller, and W. Schröder (Eds.). Handbuch der Umweltwissenschaften. ecomed. Landshut/ Lech, Germany. JØRGENSEN, S.E., B.C. PATTEN, and M. STRASKRABA. 1992. Ecosystems emerging toward an ecology of complex systems in a complex future. Ecological Modelling 62:1-28. KANT, I. 1968. (First edition 1802). Physische Geographie, Vol. IX. Pp 151-436, In F.T. Rink. Kants Werke, Akademie Textausgabe. Berlin, Germany. KLINGE, H. 1983. Forest structures in Amazonia. The Environmentalist 3, Supplement No. 5:13-23. 88 Northeastern Naturalist Special Issue 1 LAVATER, J.K. 1775-78. Physiognomische Fragmente zur Beförderung der Menschenkenntnis und Menschenliebe. Weidmann, Reich. Leipzig- Winterthur, Germany. LESER, H. 1984. Zum Ökologie-, Öcosystem-, und Öcotopbegriff. Natur und Landschaft. 59:351-357. LIKENS, G.E., F.H. BORMANN, R.S. PIERCE, J.S. EATON, and N.M. JOHNSON. 1977. Biogeochemistry of a Forested Ecosystem. Springer. New York, NY. LINNÉ, C. von. 1735. Systema Naturae. Halae Magdeburgicae. LOCKER, A. (Ed.). 1973. Biogenesis, Evolution, Homeostasis. Springer, Berlin, Germany. MARGALEF, R. 1995. Information theory and complex ecology. Pp. 40-50, In B.C. Patten and S.E. Jørgensen (Eds.). Complex Ecology: The Part-whole Relation in Ecosystems. Prentice Hall, Englewood Cliffs, NJ. MARKL, H. 1995. Physik des Lebendigen. Alexander von Humboldt Stiftung - Mitteilungen 65:13-24. MAY, R.M. 1972. Will a large complex system be stable? Nature 238:413-414. McINTOSH, R.P. 1967. An index of diversity and the relation of certain concepts to diversity. Ecology 48:392-404. McINTOSH, R.P. 1985. The Background of Ecology. Cambridge University Press, Cambridge, UK. McMURTRIE, R.E. 1975. Determinants of stability of large randomly connected systems. Journal of Theoretical Biology 50:1-11. MÜLLER, F. 1992. Hierarchical approaches to ecosystem theory. Ecological Modelling 63:215-242. MÜLLER, F. 1998. Gradients in ecological systems. Ecological Modelling 108:3-21. MÜLLER, F., and LEUPELT. (Eds.) 1998. Eco Targets, Goal Functions, and Orientors. Springer, Berlin, Germany. MÜLLER, F., and W. WINDHORST. 1991. Die Modellierungsstrategie des FE-Vorhabens “Ökosystemforschung im Bereich der Bornhöveder Seenkette.” Berichte des Forschungszentrums Waldökosysteme, Reihe B 22:75-93. NICOLIS, G., and I. PRIGOGINE. 1977. Self-organization in Non-equilibrium Systems. Wiley. New York, NY. ODUM, E.P. 1957. Trophic structure and productivity of Silver Springs, Florida. Ecological Monographs 27:55-112. ODUM, E.P. 1969. The strategy of ecosystem development. Science 164:262- 270. ODUM, H.T. 1983. Maximum power and efficiency: A rebuttal. Ecological Modelling 20:71-82. O’NEILL, R.V., D.L. DeANGELIS, J.B. WAIDE, and T.F.H. ALLEN. 1986. A Hierarchical Concept of Ecosystems. Princeton Monographs in Population Biology 23. Princeton University Press, Princeton, NJ. PATTEN, B.C. 1992. Energy, emergy, and environs. Ecological Modelling 62:29-70. PATTEN, B.C. 1998. Steps towards a cosmography of ecosystems: 20 remarkable properties of life in environment. Pp. 137-160. In F. Müller and M. Leupelt (Eds.). Eco-targets, Goal functions, and Orientors. Springer. Berlin, Germany. 2001 O. Fränzle 89 PATTEN, B.C., and S.E. JØRGENSEN (Eds.). 1995. Complex Ecology: The Part-whole Relation in Ecosystems. Prentice Hall. Englewood Cliffs, NJ. PATTEN, B.C., and E.P. ODUM. 1981. The cybernetic nature of ecosystems. American Naturalist 118:886-895. PICKETT, S.T.A. 1989a. Space-for-time substitution as an alternative to longterm studies. Pp. 110-135, In G.E. Likens (Ed.). Long-term Studies in Ecology. Springer. Berlin, Germany. PICKETT, S.T.A. 1989b. Long-term studies: Experience from the Institute of Ecosystem Studies and Cary Conference II. MAB-Mitteilungen 31:116-141. PIMM, S.L. 1982. Food Webs. Chapman Hall, London, UK. PIOTROWSKI, J.A. 1991. Quartär- und hydrogeologische Untersuchungen im Bereich der Bornhöveder Seenkette, Schleswig-Holstein. Berichte. Geologisch-Paläontologisches Institut, Universität Kiel 43. POPPER, K.R. 1959. The Logic of Scientific Discovery. Harper and Row, New York, NY. PRIGOGINE, I. 1967. Thermodynamics of Irreversible Processes. Wiley. New York, NY. PRIGOGINE, I. 1976. Order through fluctuation: Self-organization and social system. Pp. 93-133, In E. Jantsch and C.H. Waddington (Eds.). Evolution and Consciousness. Addison-Wesley, Reading, MA. PRIGOGONE, I. 1985. Vom Sein zum Werden. Zeit und Komplexität in den Naturwissenschaften. Piper, München, Germany. RICHARDS, P.W. 1963. Ecological observations on the rain forest of Mount Dulit, Sarawak. Journal of Ecology 24:1-37 and 340-360. ROWE, J.S. 1961. The level-of-integration concept and ecology. Ecology 42:420-427. SALTHE, S.N. 1985. Evolving Hierarchical Systems. Columbia University Press. New York, NY. SCHLEUSS, U. 1992. Böden und Bodenschaften einer Norddeutschen Moränenlandschaft. EcoSys - Beiträge zur Ökosystemforschung, Suppl. Bd. 2. Kiel, Germany. 185 pp. SCHNEIDER, E.D., and J. KAY. 1994a. Life as a manifestation of the second law of thermodynamics. Mathematical Computer Modelling 19:25-48. SCHNEIDER, E.D., and J. KAY. 1994b. Complexity and thermodynamics: Towards a new ecology. Futures 26:626-647. SCHOLLE, D., and J. SCHRAUTZER. 1993. Zur Grundwasserdynamik unterschiedlicher Niedermoor-Gesellschaften Schleswig-Holsteins. Zeitschrift für Ökologie und Naturschutz 2:87-98. SCHULTZ, J. 1988. Die Ökozonen der Erde. Fischer, Stuttgart, Germany. SOLBRIG, O.T., and G. NICOLIS. 1991. Biology and complexity. Pp. 1-6, In O.T. Solbrig and G. Nicolis (Eds.). Perspectives in Biological Complexity. IUBS, Paris, France. STEPHAN, H.-J., and B. MENKE. 1977. Untersuchungen über den Verlauf der Weichselkaltzeit in Schleswig-Holstein. Zeitschrift für Geomorphologie N.F. 27:12-28. STÖCKER, G. 1979. Ökosystem - Begriff und Konzeption. Archiv für Naturschutz und Landschaftsforschung 19:157-176. 90 Northeastern Naturalist Special Issue 1 STRASKRABA, M. 1995. Cybernetic theory of ecosystems. Umweltwissenschaften 6:31-52. STREIT, B. 1980. Ökologie. Thieme, Stuttgart, Germany. TANSLEY, A.G. 1935. The use and abuse of vegetational concepts and terms. Ecology 16:284-307. THIENEMANN, A.F. 1956. Leben und Umwelt. Rohwolt, Hamburg, Germany. THOM, R. 1975. Structural Stability and Morphogenesis. Benjamin, Reading, MA. TREPL, L. 1987. Geschichte der Ökologie. Athenäum, Frankfurt am Main, Germany. ULANOWICZ, R.E. 1986. Growth and Development: Ecosystems Phenomenology. Springer, Berlin, Germany. ULANOWICZ, R.E. 1995. Ecosystem integrity: A causal necessity. Pp. 77-87, In J. Lemons and L. Westra (Eds.). Perspectives on Implementing Ecological Integrity. Kluwer, Dordrecht, Holland. UNESCO. 1998. Man belongs to the Earth. International Co-operation in Environmental Research. UNESCO, Paris, France. VAN DER MAAREL, E. 1976. On the establishment of plant community boundaries. Berichte der Deutschen Botanischen Gesellschaft 89:415-443. VERMIJ, R.H. 1991. Secularisering en natuurwetenschap in de zeventiende en achttiende eeuw: Bernard Nieuwentijt. Amsterdam, Holland. VERMIJ, R.H. 1993. The Beginnings of Physico-Theology. Pp. 173-184, In M. Kattenstedt, «Grenz-Überschreitung». Abhandlungen zur Geschichte der Geowissenschaften und Religion/Umwelt-Forschung 9. WADDINGTON, C.H. 1977. Tools for Thought. Cape, London, UK. WILSON, A.G. 1981. Catastrophe Theory and Bifurcation. Croom Helm, London, UK. WORTHINGTON, E.B. (Ed). 1975. The Evolution of IBP. Cambridge University Press, Cambridge, UK. ZAUNICK. R. 1958. Alexander von Humboldt - Kosmische Naturbetrachtung. Kröner, Stuttgart, Germany.