History of Ecosystem Ecology 7 Context the a 1.2).Com the age rials that the at Ecosys land and oceans vide a cor text for under standing the broader significance of processe studied in a particular ecosystem.Latitudinal and seasonal patterns of atmospheric CO,con- centration,for example,help define the loca tions where carbon is absorbed or released from the land and oceans(see Chapter 15). Mechanism History of Ecosystem Ecology integrates the principles of several biological and Many early discoveries of bi ogy were moti provides the mechanistic egrated na system uropean se】 aterials four on the plant. idea that d。 nourishment f age structures as well as on community both air and water (Gorham 1991).Priesti extended this idea in the eighteenth century by esent and showing that plants produce a substance that is their rates of resource consumption.Ecosystem essential to support the breathing of animals.At ecology therefore depends on information about the same time MacBride and Priestley and principles developed in physiological,evo- showed that breakdown of organic matte lutionary,population,and community ecology caused the production of "fixed air"(carbon Fig.1.2). dioxide).which did not support animal life The supply of water and minerals from soils De Saussure,Liebig,and others clarified the to plants depends not only on the activities of roles carbon oxygen but t also on an I min among ninetee cn o and tv phoru osional t the wer the deta i Australia ent mechanism strains plant tha functions.Only in decades hav of nlants and a returned to the question that originally moti Principles of eco ystem ecology must therefore vated this research:How are biogeochemical rate the concepts and understand. processes integrated in the functioning of natural ecosystems? ogy,and climatology that focus on the physical Many threads of ecological thought have environment (Fig.1.2). contributed to the development of ecosystem Ecosystem ecology provides the mechanistic ecology (Hagen 1992).including ideas relating basis for understanding processes that occur to trophic interactions (the feeding relation at global scales.Study of Earth as a physical ships among organisms)and biogeochemistry system relies on information provided by (biological influences on the chemical processes
History of Ecosystem Ecology 7 on the population processes that govern plant, animal, and microbial densities and age structures as well as on community processes, such as competition and predation, that determine which species are present and their rates of resource consumption. Ecosystem ecology therefore depends on information and principles developed in physiological, evolutionary, population, and community ecology (Fig. 1.2). The supply of water and minerals from soils to plants depends not only on the activities of soil microorganisms but also on physical and chemical interactions among rocks, soils, and the atmosphere. The low availability of phosphorus due to the extensive weathering and erosional loss of nutrients in the ancient soils of western Australia, for example, strongly constrains plant growth and the quantity and types of plants and animals that can be supported. Principles of ecosystem ecology must therefore also incorporate the concepts and understanding of disciplines such as geochemistry, hydrology, and climatology that focus on the physical environment (Fig. 1.2). Ecosystem ecology provides the mechanistic basis for understanding processes that occur at global scales. Study of Earth as a physical system relies on information provided by ecosystem ecologists about the rates at which the land or water surface interacts with the atmosphere, rocks, and waters of the planet (Fig. 1.2). Conversely, the global budgets of materials that cycle between the atmosphere, land, and oceans provide a context for understanding the broader significance of processes studied in a particular ecosystem. Latitudinal and seasonal patterns of atmospheric CO2 concentration, for example, help define the locations where carbon is absorbed or released from the land and oceans (see Chapter 15). History of Ecosystem Ecology Many early discoveries of biology were motivated by questions about the integrated nature of ecological systems. In the seventeenth century, European scientists were still uncertain about the source of materials found in plants. Plattes, Hooke, and others advanced the novel idea that plants derive nourishment from both air and water (Gorham 1991). Priestley extended this idea in the eighteenth century by showing that plants produce a substance that is essential to support the breathing of animals.At about the same time MacBride and Priestley showed that breakdown of organic matter caused the production of “fixed air” (carbon dioxide), which did not support animal life. De Saussure, Liebig, and others clarified the explicit roles of carbon dioxide, oxygen, and mineral nutrients in these cycles in the nineteenth century. Much of the biological research during the nineteenth and twentieth centuries went on to explore the detailed mechanisms of biochemistry, physiology, behavior, and evolution that explain how life functions. Only in recent decades have we returned to the question that originally motivated this research: How are biogeochemical processes integrated in the functioning of natural ecosystems? Many threads of ecological thought have contributed to the development of ecosystem ecology (Hagen 1992), including ideas relating to trophic interactions (the feeding relationships among organisms) and biogeochemistry (biological influences on the chemical processes Earth system science Climatology Hydrology Soil science Geochemistry Physiological ecology Ecosystem ecology Population ecology Community ecology Context Mechanism Figure 1.2. Relationships between ecosystem ecology and other disciplines. Ecosystem ecology integrates the principles of several biological and physical disciplines and provides the mechanistic basis for Earth System Science
8 1.The Ecosystem Concep in ecosystems).Early research on trophic inter-that there were insufficient data to draw such actions emphasized the transfer of energy broad conclusions and that it was inappropriate among organisms.Elton (1927),an English to use mathematical models to infer general zoologist interested in natural history, relationships based on observations from a described the role that an animal plays in a single lake.Hutchinson,Lindeman's postdoc community (its niche)in terms of what it eats toral adviser,finally (after Lindeman's death) by.He in a foo matte om organisn to trophic s ructure ide on's concepts unc H T Odun of ems (see fur the oach to studvine Hutchinson.an American limnologist,was which mphasizes the e general properties of strongly influenced by the ideas of elton and osystems without docur enting all the under those of russian geochemist vernadsky who lving mechanisms and interactions.The Odum deseribed the movement of minerals from soil brothers used radioactive tracers to measure into vegetation and back to soil.Hutchinson the movement of energy and materials through suggested that the resources available in a lake a coral reef.These studies enabled them to doc must limit the productivity of algae and that ument the patterns of energy flow and metab algal productivity,in turn,must limit the abun- olism of whole ecosystems and to suggest dance of animals that eat algae.Meanwhile, generalizations about how ecosystems function Tansley (1935).a British (Odum 196 nc bee levelope strongly tha anc rec Ovin y1993 ment.e the te iding informa that is esser al for gen e t inorganic addressed by systems ecology include infor and organic cor omponents as well as among ansfer (Margalef 1968 )the ucture organisms food webs (Polis 1991).the hier rchical cha I indeman another limnologist was stron in ecosystem controls at different tem oral influenced by all these threads of ecological and spatial scales(O'Neill et al.1986),and the theory.He suggested that energy flow through resilience of ecosystem properties after distur- an ecosystem could be used as a currency to bance (Holling 1986) quantify the roles of organisms in trophic We now recognize that element cycles inter dynamics.Green plants (primary producers) act in important ways and cannot be under capture energy and transfer it to animals stood in isolation.The availability of water and (consumers)and At each trans nitrogen are important determinants of the rate energy Is ecosystem at which carbon cycles through the ecosystem through respira the productivity ntity he cycling rates of nitrogen ec t global cha ng in the made ecologist environ of the hic changes in e tha occur carbon arch on dis e or other the trophic-dynamic a nect of ecolooy was ini tal change ion the direct tially rejected for publication.Reviewers felt in ecosystem structure and functioning result
8 1. The Ecosystem Concept in ecosystems). Early research on trophic interactions emphasized the transfer of energy among organisms. Elton (1927), an English zoologist interested in natural history, described the role that an animal plays in a community (its niche) in terms of what it eats and is eaten by. He viewed each animal species as a link in a food chain, which described the movement of matter from one organism to another. Elton’s concepts of trophic structure provide a framework for understanding the flow of materials through ecosystems (see Chapter 11). Hutchinson, an American limnologist, was strongly influenced by the ideas of Elton and those of Russian geochemist Vernadsky, who described the movement of minerals from soil into vegetation and back to soil. Hutchinson suggested that the resources available in a lake must limit the productivity of algae and that algal productivity, in turn, must limit the abundance of animals that eat algae. Meanwhile, Tansley (1935), a British terrestrial plant ecologist, was also concerned that ecologists focused their studies so strongly on organisms that they failed to recognize the importance of exchange of materials between organisms and their abiotic environment. He coined the term ecosystem to emphasize the importance of interchanges of materials between inorganic and organic components as well as among organisms. Lindeman, another limnologist, was strongly influenced by all these threads of ecological theory. He suggested that energy flow through an ecosystem could be used as a currency to quantify the roles of organisms in trophic dynamics. Green plants (primary producers) capture energy and transfer it to animals (consumers) and decomposers. At each transfer, some energy is lost from the ecosystem through respiration.Therefore, the productivity of plants constrains the quantity of consumers that an ecosystem can support. The energy flow through an ecosystem maps closely to carbon flow in the processes of photosynthesis, trophic transfers, and respiratory release of carbon. Lindeman’s dissertation research on the trophic-dynamic aspect of ecology was initially rejected for publication. Reviewers felt that there were insufficient data to draw such broad conclusions and that it was inappropriate to use mathematical models to infer general relationships based on observations from a single lake. Hutchinson, Lindeman’s postdoctoral adviser, finally (after Lindeman’s death) persuaded the editor to publish this paper, which has been the springboard for many of the basic concepts in ecosystem theory (Lindeman 1942). H. T. Odum, also trained by Hutchinson, and his brother E. P. Odum further developed the systems approach to studying ecosystems, which emphasizes the general properties of ecosystems without documenting all the underlying mechanisms and interactions. The Odum brothers used radioactive tracers to measure the movement of energy and materials through a coral reef. These studies enabled them to document the patterns of energy flow and metabolism of whole ecosystems and to suggest generalizations about how ecosystems function (Odum 1969). Ecosystem budgets of energy and materials have since been developed for many fresh-water and terrestrial ecosystems (Lindeman 1942, Ovington 1962, Golley 1993), providing information that is essential for generalizing about global patterns of processes such as productivity. Some of the questions addressed by systems ecology include information transfer (Margalef 1968), the structure of food webs (Polis 1991), the hierarchical changes in ecosystem controls at different temporal and spatial scales (O’Neill et al. 1986), and the resilience of ecosystem properties after disturbance (Holling 1986). We now recognize that element cycles interact in important ways and cannot be understood in isolation. The availability of water and nitrogen are important determinants of the rate at which carbon cycles through the ecosystem. Conversely, the productivity of vegetation strongly influences the cycling rates of nitrogen and water. Recent global changes in the environment have made ecologists increasingly aware of the changes in ecosystem processes that occur in response to disturbance or other environmental changes. Succession, the directional change in ecosystem structure and functioning result-
History of Ecosystem Ecology 9 ing from biotically driven changes impor ien under tems Early Am sts such as Co rties of herelatively pr prop veg pe nded s (Dok 1879.0 dictable patterns of vegetation develo 1941.Ellenberg 1978).Pro ess-based studies of after osure of un getated land surfaces. vided insight into many Sand dunes on Lake Michigan,for example,are initially colonized by drought-resistant herba- of organisms and soils along these gradients ceous plants that give way to shrubs,then small (Billings and Mooney 1968,Mooney 1972 trees and eventually forests (Cowles 1899). Larcher 1995.Paul and Clark 1996).These Clements(1916)advanced a theory of commu- studies also formed the basis for extrapolation nity development,suggesting that this vegeta- of processes across complex landscapes to char tion succession is a predictable process that acterize large regions (Matson and Vitousek eventually leads,in the absence of disturbance 1987,Turner et al.2001).These studies often to a stable community type characteristic of a relied on field or laboratory experiments climax).He sug- that manipulated some ecosyst m property o munity is like proc ess ve studies acting part tudies of an od Thi 977 Sch ecological comn it efully designed gradient studic organism laid the groundwork for pts of al.1988 ecosystem physiology (for example.the net Ecosystem experiments have provided both ecosystem ex ofCO、and water vano basic understanding and information that are between the ecosystem and the atmosphere) critical in management decisions.The clear The measurements of net ecosystem exchange cutting of an experimental watershed a are still an active area of research in ecosystem Hubbard Brook in the northeastern United ecology.although they are now motivated by States,for example.caused a fourfold increase different questions than those pose by in streamflow and stream nitrate concentra Clements.His ideas were controversial from tion-to levels exceeding health standard such as Gleason vater (Likens et al.1977).These not a emonstrate the key role o as ents ha lied Ins geta di the pers taedpey debate lands led to large-scal sible for mncrease supp (see Chapter 13). s in the Ex mental Lakes Are eral approach to ecosystem of southern canada she owed that pho limits the productivity of many lakes(Schindler tem 1985)and that pollution was sponsible studies of ecosystem components.This interest for algal blooms and fish kills that were originated in studies by plant geographers and common in lakes near densely populated areas soil scientists who described general patterns of in the 1960s.This research provided the basis variation with respect to climate and geological for regulations that removed phosphorus from substrate (Schimper 1898).These studies detergents. showed that many of the global patterns of Changes in the Earth System have led to plant production and soil development vary studies of the interactions among terrestrial
History of Ecosystem Ecology 9 ing from biotically driven changes in resource supply, is an important framework for understanding these transient dynamics of ecosystems. Early American ecologists such as Cowles and Clements were struck by the relatively predictable patterns of vegetation development after exposure of unvegetated land surfaces. Sand dunes on Lake Michigan, for example, are initially colonized by drought-resistant herbaceous plants that give way to shrubs, then small trees, and eventually forests (Cowles 1899). Clements (1916) advanced a theory of community development, suggesting that this vegetation succession is a predictable process that eventually leads, in the absence of disturbance, to a stable community type characteristic of a particular climate (the climatic climax). He suggested that a community is like an organism made of interacting parts (species) and that successional development toward a climax community is analogous to the development of an organism to adulthood. This analogy between an ecological community and an organism laid the groundwork for concepts of ecosystem physiology (for example, the net ecosystem exchange of CO2 and water vapor between the ecosystem and the atmosphere). The measurements of net ecosystem exchange are still an active area of research in ecosystem ecology, although they are now motivated by different questions than those posed by Clements. His ideas were controversial from the outset. Other ecologists, such as Gleason (1926), felt that vegetation change was not as predictable as Clements had implied. Instead, chance dispersal events explained much of the vegetation patterns on the landscape. This debate led to a century of research on the mechanisms responsible for vegetation change (see Chapter 13). Another general approach to ecosystem ecology has emphasized the controls over ecosystem processes through comparative studies of ecosystem components. This interest originated in studies by plant geographers and soil scientists who described general patterns of variation with respect to climate and geological substrate (Schimper 1898). These studies showed that many of the global patterns of plant production and soil development vary predictably with climate (Jenny 1941, Rodin and Bazilevich 1967, Lieth 1975). The studies also showed that, in a given climatic regime, the properties of vegetation depended strongly on soils and vice versa (Dokuchaev 1879, Jenny 1941, Ellenberg 1978). Process-based studies of organisms and soils provided insight into many of the mechanisms underlying the distributions of organisms and soils along these gradients (Billings and Mooney 1968, Mooney 1972, Larcher 1995, Paul and Clark 1996). These studies also formed the basis for extrapolation of processes across complex landscapes to characterize large regions (Matson and Vitousek 1987, Turner et al. 2001). These studies often relied on field or laboratory experiments that manipulated some ecosystem property or process or on comparative studies across environmental gradients. This approach was later expanded to studies of intact ecosystems, using whole-ecosystem manipulations (Likens et al. 1977, Schindler 1985, Chapin et al. 1995) and carefully designed gradient studies (Vitousek et al. 1988). Ecosystem experiments have provided both basic understanding and information that are critical in management decisions. The clearcutting of an experimental watershed at Hubbard Brook in the northeastern United States, for example, caused a fourfold increase in streamflow and stream nitrate concentration—to levels exceeding health standards for drinking water (Likens et al. 1977). These dramatic results demonstrate the key role of vegetation in regulating the cycling of water and nutrients in forests.The results halted plans for large-scale deforestation that had been planned to increase supplies of drinking water during a long-term drought. Nutrient addition experiments in the Experimental Lakes Area of southern Canada showed that phosphorus limits the productivity of many lakes (Schindler 1985) and that pollution was responsible for algal blooms and fish kills that were common in lakes near densely populated areas in the 1960s. This research provided the basis for regulations that removed phosphorus from detergents. Changes in the Earth System have led to studies of the interactions among terrestrial
10 1.The Ecosystem Concept ecosystems,the atmosphere,and the oceans.oxidation of hydrogen sulfide(HS)to produce The dramatic impact of human activities on the organic matter.Decomposer microorganisms Earth System (Vitousek 1994a)has led to the (microbes)break down dead organic material urgent necessity to understand how terrestrial releasing CO2 to the atmosphere and nutrients ecosystem processes affect the atmosphere in forms that are available to other microbes and oceans.The scale at which these ecosystem and plants.If there were no decomposition enects are largeha large accumulations of dead organic equester ent glob.em support pl nt grow critica of atm ponent they trans models are in an Igy an s The tial abiotie nts o CO,and pollutants in the atmosphere.for an ecosystem are water;the atmo which example,provide telltale evidence of the majo scarbon and nitrogen:and soil minerals locations and causes of global problems (Tans which supply other nutrients required by et al.1990).This gives hints about which ecosys- organisms tems and processes have the greatest impact on An ecosystem model describes the major the earth system and therefore where research pools and fluxes in an ecosystem and the factors and management should focus efforts to under that regulate these fluxes.Nutrients,water.and stand and solve these problems(Zimov et al. energy differ from one another in the relative 1999)- mportance of ecosystem inputs and outputs vs The of systems approaches internal recycling(see Chapters 4 to 10).Plants process understan ing.an d glo or exa mple,acquire arbon prim ing I atmosphere,an in the on ret to the atmo cycling th gh tems Is therefor eimegrowd th la nerties influe nce the Earth System?The ools of carbon sto red in stems.so s that a rino in e ctivities of animals and m rohes are have blurred any previous distinction between what buffered from variations in carbon up basic and applied research.There is an urgen take by plants.The water cycle of eco stem need to understand how and why the ecc SVS- is also relatively open with water entering tems of Earth are changing. primarily by precipitation and leaving by evap oration,transpiration,and drainage to ground- water and streams In contrast to carbon Ecosystem Structure most ecosystems have a limited capacity to store water in plants and soil,so the activity of Most ec cosystem energy from the sun and to water inputs or cks.transf In contra nong compe en withi sten and phosphorus th recycled ith 11 tha re sma relat ntit in s in the 8 These differ pe ss of bri and sshuff "of the few ec cycles fundar s such as doe tally infuence the cont ols n thermal vents.have no ants but instead patterns of the cycling of mate rials through have bacteria that derive energy from the
10 1. The Ecosystem Concept ecosystems, the atmosphere, and the oceans. The dramatic impact of human activities on the Earth System (Vitousek 1994a) has led to the urgent necessity to understand how terrestrial ecosystem processes affect the atmosphere and oceans. The scale at which these ecosystem effects are occurring is so large that the traditional tools of ecologists are insufficient. Satellite-based remote sensing of ecosystem properties, global networks of atmospheric sampling sites, and the development of global models are important new tools that address global issues. Information on global patterns of CO2 and pollutants in the atmosphere, for example, provide telltale evidence of the major locations and causes of global problems (Tans et al. 1990).This gives hints about which ecosystems and processes have the greatest impact on the Earth System and therefore where research and management should focus efforts to understand and solve these problems (Zimov et al. 1999). The intersection of systems approaches, process understanding, and global analysis is an exciting frontier of ecosystem ecology. How do changes in the global environment alter the controls over ecosystem processes? What are the integrated system consequences of these changes? How do these changes in ecosystem properties influence the Earth System? The rapid changes that are occurring in ecosystems have blurred any previous distinction between basic and applied research. There is an urgent need to understand how and why the ecosystems of Earth are changing. Ecosystem Structure Most ecosystems gain energy from the sun and materials from the air or rocks, transfer these among components within the ecosystem, then release energy and materials to the environment. The essential biological components of ecosystems are plants, animals, and decomposers. Plants capture solar energy in the process of bringing carbon into the ecosystem. A few ecosystems, such as deep-sea hydrothermal vents, have no plants but instead have bacteria that derive energy from the oxidation of hydrogen sulfide (H2S) to produce organic matter. Decomposer microorganisms (microbes) break down dead organic material, releasing CO2 to the atmosphere and nutrients in forms that are available to other microbes and plants. If there were no decomposition, large accumulations of dead organic matter would sequester the nutrients required to support plant growth. Animals are critical components of ecosystems because they transfer energy and materials and strongly influence the quantity and activities of plants and soil microbes. The essential abiotic components of an ecosystem are water; the atmosphere, which supplies carbon and nitrogen; and soil minerals, which supply other nutrients required by organisms. An ecosystem model describes the major pools and fluxes in an ecosystem and the factors that regulate these fluxes. Nutrients, water, and energy differ from one another in the relative importance of ecosystem inputs and outputs vs. internal recycling (see Chapters 4 to 10). Plants, for example, acquire carbon primarily from the atmosphere, and most carbon released by respiration returns to the atmosphere. Carbon cycling through ecosystems is therefore quite open, with large inputs to, and losses from, the system. There are, however, relatively large pools of carbon stored in ecosystems, so the activities of animals and microbes are somewhat buffered from variations in carbon uptake by plants. The water cycle of ecosystems is also relatively open, with water entering primarily by precipitation and leaving by evaporation, transpiration, and drainage to groundwater and streams. In contrast to carbon, most ecosystems have a limited capacity to store water in plants and soil, so the activity of organisms is closely linked to water inputs. In contrast to carbon and water, mineral elements such as nitrogen and phosphorus are recycled rather tightly within ecosystems, with annual inputs and losses that are small relative to the quantities that annually recycle within the ecosystem. These differences in the “openness” and “buffering” of the cycles fundamentally influence the controls over rates and patterns of the cycling of materials through ecosystems
Controls over Ecosystem Processes 11 The pool sizes and rates of cycling differ variations in climate explain the distribution of substantially among ecosystems (see Chapter biomes(types of ecosystems)such as wet trop rger po ore mperate rct dra (see Within each bi pails tha bor stem als ation in expla sub at eg pools.for re asons that will be explored in later P climate t at a local chapters. scale.The potential biota s s and diversity of organisms that actually ar ●ontrols over Ecosystem Processes mainland ecosystems because new species reach islands less frequently and are more likely to go extinct than in mainland locations (MacArthur and Wilson 1967).Time influences the development of soil and the evolution of organisms o material (i.e.,the rocks that give rise to soils) incorporates the ecosysten pas disturb a wide and enviro occupy a site).and time (Fig.1.3)(Jenny 1941 are Amundson and Jenny 1997).Together these five factors set the bounds for the characteris s state facto ach was a maior tics of an ecosystem. vstem ecolog On broad geographic scales,climate is the state factor that most strongly determines First it emphasized the controls over proces ecosystem processes and structure.Global rather than simply descriptions of patterns Second,it suggested an experimental approach to test the importance and mode of action of Climate Time each control.A logical way to study the role of each state factor is to compare sites that are as Disturbance regime similar as possible with respect to all but one a chronosequence 1s8 Modulators factor.For example. Ecosystem series of sites of different ages with simila climat processes. potent al parent ma e c same orgar Resources pos er et al Poleoal material allow us to study the impact of these state factors on ec (Vitousel FIGuRE 1.3.The relationship between state factors et al.1988.Walker et al.1998).Finally.a com parison of ecosystems that differ primarily in side the potential biota,such as the mediterranean shrublands that have developed on west coasts permission from American Naturalist.Vol.148 of California,Chile,Portugal.South Africa,and 1996 University of Chicago Press Chapin et al.1996.) Australia,illustrates the importance of evolu-
Controls over Ecosystem Processes 11 The pool sizes and rates of cycling differ substantially among ecosystems (see Chapter 6). Tropical forests have much larger pools of carbon and nutrients in plants than do deserts or tundra. Peat bogs, in contrast, have large pools of soil carbon rather than plant carbon. Ecosystems also differ substantially in annual fluxes of materials among pools, for reasons that will be explored in later chapters. Controls over Ecosystem Processes Ecosystem structure and functioning are governed by at least five independent control variables. These state factors, as Jenny and co-workers called them, are climate, parent material (i.e., the rocks that give rise to soils), topography, potential biota (i.e., the organisms present in the region that could potentially occupy a site), and time (Fig. 1.3) (Jenny 1941, Amundson and Jenny 1997). Together these five factors set the bounds for the characteristics of an ecosystem. On broad geographic scales, climate is the state factor that most strongly determines ecosystem processes and structure. Global variations in climate explain the distribution of biomes (types of ecosystems) such as wet tropical forests, temperate grasslands, and arctic tundra (see Chapter 2). Within each biome, parent material strongly influences the types of soils that develop and explains much of the regional variation in ecosystem processes (see Chapter 3). Topographic relief influences both microclimate and soil development at a local scale. The potential biota governs the types and diversity of organisms that actually occupy a site. Island ecosystems, for example, are frequently less diverse than climatically similar mainland ecosystems because new species reach islands less frequently and are more likely to go extinct than in mainland locations (MacArthur and Wilson 1967). Time influences the development of soil and the evolution of organisms over long time scales. Time also incorporates the influences on ecosystem processes of past disturbances and environmental changes over a wide range of time scales. These state factors are described in more detail in Chapter 3 in the context of soil development. Jenny’s state factor approach was a major conceptual contribution to ecosystem ecology. First, it emphasized the controls over processes rather than simply descriptions of patterns. Second, it suggested an experimental approach to test the importance and mode of action of each control. A logical way to study the role of each state factor is to compare sites that are as similar as possible with respect to all but one factor. For example, a chronosequence is a series of sites of different ages with similar climate, parent material, topography, and potential to be colonized by the same organisms (see Chapter 13). In a toposequence, ecosystems differ mainly in their topographic position (Shaver et al. 1991). Sites that differ primarily with respect to climate or parent material allow us to study the impact of these state factors on ecosystem processes (Vitousek et al. 1988, Walker et al. 1998). Finally, a comparison of ecosystems that differ primarily in potential biota, such as the mediterranean shrublands that have developed on west coasts of California, Chile, Portugal, South Africa, and Australia, illustrates the importance of evoluTime Topography Climate Parent material Potential biota Ecosystem processes Modulators Resources Biotic community Disturbance regime Human activities Figure 1.3. The relationship between state factors (outside the circle), interactive controls (inside the circle), and ecosystem processes. The circle represents the boundary of the ecosystem. (Modified with permission from American Naturalist, Vol. 148 © 1996 University of Chicago Press, Chapin et al. 1996.)