These two problems will lurk in the background of all the discussions here. There are also many other general philosophical problems about science that also occur in the ecological context, for instance, the role of idealizations, models, and so on. But ecology does not seem to make any unique contribution to these problems—they will be ignored here in the interest of focusing on ecology per se. The aim of this entry is to describe—in broad terms—the type of philosophical questions raised by different areas of ecology so as to encourage further philosophical work.
No general conclusions will be drawn from the various cases because none seem feasible at present. The golden age of theoretical ecology —to borrow the title of a book edited by Scudo and Ziegler—consisted primarily of population ecology. In recent years, interest has reverted to population ecology, sometimes in the form of metapopulation models consisting of a set of populations with migration between them. Models in population ecology are based on representing an ecological system as the set of populations of the same or different species it consists of.
Each population, in turn, consists of potentially interacting individuals of a species. Populations may be characterized by their state variables parameters representing properties of the population as a whole, for instance, size, density, growth-rate, etc. Classical population ecology is the part of ecology that is theoretically the most developed. Population ecology considers both deterministic and stochastic models. Much of philosophical writing on population ecology has been restricted to deterministic population models and this relatively large body of work will only be very briefly summarized.
More attention will be paid to stochastic models which raise much more interesting philosophical issues that have not been adequately explored. If population sizes are large, they can be studied using deterministic models, that is, fluctuations in populations sizes due to chance factors such as accidental births and deaths can be ignored. Usually a model considers members of a single or a very few interacting species, for instance, a few predator and a prey species.
The explanation of this phenomenon is straightforward: as prey populations increase, the increased availability of resources allows a rise in predator populations a little later in time. But the increase of predators leads to an increase of prey consumption and, consequently, a decrease in prey populations. But, now, the lack of resources leads to a decline of predator populations. As predator populations decline, prey populations increase initiating the cycle once again.
See Figure The Lotka-Volterra model mathematically predicts these cycles. As such, it exemplifies the explanatory ideal of ecology: not only is there a predictively accurate quantitative model, but the mechanisms incorporated in the model have a perspicuous biological interpretation.
Unfortunately, in ecology, because of the formalization and interpretation indeterminacy problems, the last condition is rarely satisfied. For the simpler case of single species, two standard models are that of exponential and logistic growth. The exponential growth model is supposed to capture the behavior of a population when there is no resource limitation; the logistic growth model is one of the simplest ways to try to capture the self-regulation of population sizes when there is such a limitation.
Then the population dynamics is described by the growth equation:. This is the exponential growth model. It assumes that no resource limitation constrains the "intrinsic growth rate", r. It can be solved to give:. Figure 2b shows how a population governed by the logistic equation grows in size.
At the level of individual behavior, this model does not have the kind of justification that the exponential growth model does in the sense that the logistic equation cannot be plausibly derived from the properties of individuals.
In this sense it is a purely "phenomenological" model. The exponential growth model appeals to only one essentially ecological parameter, the intrinsic growth rate r of a population, interpreted as the rate at which the population would grow if there were no external factor limiting growth; the logistic model also appeals to the carrying capacity K , interpreted as the maximum size of the population that can persist in a given environment.
See Figures 2a and 2b:. The figure on the bottom from Gause [], p. In general, biological experience suggests that all populations regulate their sizes, that is, they show self-regulation. Theoretical exploration of models has made it clear that a wide variety of mechanisms can lead to such self-regulation but it is usually unclear which models are more plausible than others thanks to the typical formalization indeterminacy of the field. Moreover, the precise mechanisms that are playing regulative roles in individual cases are often very hard to determine in the field, a classic case of partial observability.
Even parameters such as the intrinsic growth rate and carrying capacity are unusually difficult to estimate precisely. The last mentioned difficulties are perhaps most famously illustrated by the year cycle of snowshoe hares, muskrats and their predators in the North American boreal forests and, especially, the 4-year cycle of lemmings and, possibly, other microtines in the arctic tundra of Eurasia and North America.
In spite of almost seventy-five years of continuous research on these well-documented cycles the mechanisms driving them remain unresolved. Models producing such cycles abound, but the structural uncertainty of most of these models, coupled with partial observability of many of the parameters in the field have precluded resolution of the debate.
The models discussed so far are continuous-time models, that is, the temporal or dynamic parameter is assumed to be a continuous variable. However, discrete-time models have also been used to study population processes. A discrete analog of the logistic growth model was one of the first systems in which chaotic dynamic phenomena were discovered. If population sizes are small, then models should be stochastic: the effects of fluctuations due of population size must be explicitly analyzed.
Stochastic models in ecology are among the most mathematically complex models in science. Nevertheless they have begun to be systematically studied because of their relevance to biological conservation—see the entry on conservation biology.
They also raise philosophically interesting questions because they underscore the extent to which the nature of randomness and uncertainty remains poorly explored in biological contexts. What has, by and large, become the standard classification of stochasticity goes back to a dissertation by Shaffer.
The context of that dissertation provides a striking exemplar of the social determination of science. For small populations, even if it is increasing in size on the average, a chance fluctuation can result in extinction. Stochastic models are necessary to predict parameters such as the probability of extinction within a specified time period or the expected time to extinction.
In his dissertation, Shaffer attempted such an analysis for the grizzly bears Ursus arctos of Yellowstone which were believed to face the prospect of stochastic extinction. Shaffer distinguished four sources of uncertainty that can contribute to random extinction:. Shaffer went on to argue that all these factors increase in importance as the population size decreases—a claim that will be questioned below—and, therefore, that their effects are hard to distinguish.
Leaving aside the conventional elements of the definition given above, even for the same species, populations in marginally different habitat patches often show highly variable demographic trends resulting in highly variable MVP estimates for them, with each estimate depending critically on the local context. This should not come as a surprise: what would have been more surprising is if legislative fiat had identified a scientifically valuable parameter.
After the demise of the concept of the MVP, PVA began to be performed largely to estimate other parameters, especially the expected time to extinction of a population the estimation of which does not require any conventional choices—see the entry on conservation biology.
The first point to note is that genetic stochasticity is not even the same type of mechanism as the other three: its presence makes Shaffer's classification oddly heterogeneous. The reason for this is that genetic stochasticity is a consequence of demographic stochasticity: in small populations, a particular allele may reach fixation purely by chance reproductive events.
It is even possible that stochasticity increases the rate at which a beneficial allele may go to fixation in a small population provided that the initial frequency of that allele is already high. Leaving genetic stochasticity aside, do the other three categories provide a good classification of stochasticity, or is the classification more like that of animals in Borges' notorious Chinese encyclopedia?
Relevance is determined contextually by the possibility of there being a coherent account of how those categories came to be defined. Including genetic stochasticity in the classification leads to a lack of such coherence—hence, its exclusion above; ii the categories should be jointly exhaustive, able to subsume all cases of the relevant phenomena; and iii the categories should be mutually exclusive, that is, no phenomenon should be subsumed under more than one category.
It is the third criterion that is often called into question by Shaffer's classification of stochasticity primarily because of formalization indeterminacy. Consider some small reptile that fails to breed because a flood creates a barrier across its habitat that it cannot cross and there are no available mates on its side of the barrier. Is this environmental or demographic stochasticity?
On the one hand, it is obviously environmental because floods are precisely the type of mechanism by which environmental stochasticity is expressed. On the other hand, it is equally obviously demographic because the failure to reproduce is due to the chance unavailability of a mate on the appropriate side of the barrier.
Ultimately, as will be discussed below in some detail, whether this is a case of demographic or environmental stochasticity depends on how it is modeled. There is an important philosophical lesson here: especially when a new discipline is being formed, the structure of the phenomena—how they are distinguished and classified—are in part determined by the models used to represent them.
Consequently, classification is not theoretically innocent. This is as true in ecology as in any other scientific context. Thus, as a prelude to modeling, Lande et al. The impact of environmental stochasticity is roughly the same for small and large populations.
Random variation in individual fitness, coupled with sampling effects in a finite population, produces demographic stochasticity. Moreover, Lande and others regard random catastrophes as extreme cases of environmental stochasticity.
Consequently, it requires explicit mathematical models in which these distinctions are made exact through formal definitions. If it does, the model in question is one of demographic stochasticity; if it does not, it is one of environmental stochasticity. This choice captures the intuition mentioned earlier that the effect of the former depends on the population size whereas the effect of the latter does not.
The mathematical analysis of these models is non-trivial. The most general and uncontroversial theoretical result to date is that progressively larger populations are required for safety in the face of demographic, environmental, and random catastrophic stochasticity.
Moreover, because of the structural uncertainly of these models, apparently slight differences in assumptions and techniques routinely lead to widely divergent predictions. This can be illustrated using the well-studied example of the Yellowstone grizzlies. In Foley constructed a model for this population incorporating environmental stochasticity alone and depending on the intrinsic growth rate of the population and the carrying capacity of the environment.
In Foley constructed another model incorporating both demographic and environmental stochasticity but with the option of setting either part equal to 0. It does not. It predicts a much lower expected time to extinction. The interactive definition given above is attractive for two reasons: a mere association leaves little of theoretical or practical interest to study, while requiring some specified elevated levels of interaction introduces an unnecessary arbitrariness in a definition of community; and b the former would make any association of species a community [ 34 ] whereas the latter would typically introduce so much structure that virtually no association would constitute a community.
Community models can be conveniently represented as loop, diagrams [ 35 ] generalized graphs that have each species as a vertex and edges connecting these vertices when the species interact. The edges indicate whether the relevant species benefit or are harmed by the interaction, that is, whether they tend to increase or decrease in abundance, by an interaction.
As with population ecology, what is of most interest is are the changes in a community over time. This brings us to one of the most interesting—and one of the most vexed—questions of ecology: the relationship between diversity and stability. A deeply rooted intuition among ecologists has been that diversity begets stability.
If this claim is true, it has significant consequences for biodiversity conservation—see conservation biology. Both communities have the same richness because they both have two species; however, there is a clear sense in which the first is more diverse—or less homogeneous—than the second. Moreover, the difference is likely to be relevant. If diversity does beget stability in these communities, then that stability must be a result some interaction between the two species. If species B comprises only 0.
There have been several attempts to define and quantify diversity beyond richness; one of them is described in Box 2. This is a measure of the diversity of a community in the same way that the Shannon measure of information content is a measure of the variety in a signal. Unfortunately, though, there has been little success in tying these concepts to theoretical rules or even empirical generalizations.
Stability turns out to be even more difficult to define. At the practical level, this definition faces the problem of vacuous scope: almost no natural community satisfies such a strict requirement of equilibrium.
Moreover, almost every community experiences significant disturbances. With this in mind, stability has been variously explicated using a system's response to disturbances or its tendency not to change beyond specified limits even in the absence of disturbance. Boxes 3 and 4 see below list some of the definitions of stability that have been in vogue and how they may be measured in the field.
How do any of these measures of stability relate to diversity? The only honest answer is that no one is sure. If diversity is interpreted as richness, traditionally, it was commonly assumed that diversity is positively correlated with at least persistence.
However, there was never much hard evidence supporting this assumption. If stability is interpreted as a return to equilibrium, mathematical models that should answer questions about stability are easy to construct but hard to analyze unless the system is already close to equilibrium. This is called local stability analysis.
The most systematic analyses performed so far give no straightforward positive correlation. The traditional assumption of a general positive correlation between diversity as richness and stability has been seriously challenged on both theoretical and empirical grounds since the s. Meanwhile, Pfisterer and Schmid have produced equally compelling empirical evidence that richness is inversely correlated with stability, interpreted as resilience and resistance.
All that is certain is that McCann's confident verdict in favor of a positive diversity-stability relationship was premature. At least at the theoretical level, this remains an open field for philosophers. Clear formal results would not go unnoticed by ecologists.
Within community ecology, philosophers have lately paid considerable attention to the theory of island biogeography and the controversies surrounding its relevance for the design of networks of biological reserves. But, what is the form of this relationship? Moreover, what is the mechanism responsible for it? In spite of sporadic work over almost an entire century, these remain open questions. Larger areas were presumed to have greater habitat heterogeneity and could, therefore, host a larger number of species each with its own specific needs.
In recent years the relation is more often attributed to the belief that larger areas can support larger populations of any species. Consequently, on the average, more species are likely to be present in larger areas than smaller ones even if both started with the same species richness. Whether the species-area curve rather than the mere qualitative relation has any empirical support remains a matter of contention.
There is a dynamic equilibrium in the sense that this number does not change over time though there is a turnover of species which changes the composition of the community. The immigration rate varies inversely with the degree of isolation while the extinction rate decreases with area. Thus, this theory incorporates the second mechanism for the species-area relation mentioned in the last paragraph.
While some initial experimental evidence seemed to support the theory, by the mids its status had become controversial. Nevertheless, in the s, island biogeography began to be viewed as a model for biological reserves which, by being surrounded by anthropogenically transformed lands, were supposed to be similar to islands—see the entry on conservation biology.
The initially prevalent view, based on island biogeography theory, was that reserves should be as large as possible. Meanwhile the species-area curve also began to generate serious skepticism. Important early criticism of the use of island biogeography theory for reserve network design came from Margules and several collaborators in By , it became clear that there would be no winner in the SLOSS debate; since then there has been no unequivocal role for island biogeography theory to play in the design of biodiversity reserve networks.
Should ecosystem ecology, then, be regarded as an instance of the unification of the physical and biological sciences? There has been so little philosophical attention to ecology that this question does not appear ever to have been broached. They also demanded that other specialists, including geochemists and soil scientists, be brought in so that all the relevant physical parameters of ecosystems, besides the biological ones, could be tracked simultaneously.
This study constituted the biologists' attempt to engage in publicly-funded Big Science, initiated by the physicists during the Manhattan Project, and subsequently profitably exploited by social scientists since the s. The trouble was that, at this level of analysis, very few general claims could be sustained. Those that could—for instance, that Sun is ultimately the source of all energy in biological systems or that primary producers have to contain chlorophyll or some other such molecule—were usually trivial and well-known long before the initiation of systematic large-scale ecosystem studies in the s.
Usually ecosystem studies produced detailed analyses of nutritional or climatic requirements of particular communities. But the details of nutritional requirements were either so general as to be almost irrelevant, or so specific that they were rarely transportable from one ecosystem to another.
Almost all of what is known about climatic requirements of vegetation types and other communities was known to biogeographers long before the invention of ecosystem studies. The carbon and nitrogen cycles had also been worked out long before the advent of ecosystem studies as an organized discipline. However, the physical characteristics of habitats do matter to organisms living in them.
Moreover, physical changes on a global scale, for instance, climate change through global warming, have serious long-term implications for biota. In one interesting analysis—one among many—Ryan has used a complex model tying physiological processes to the physical environment to suggest that increased temperature will make maintenance respiration which represents the physiological costs of protein synthesis and replacement, membrane repair, and the maintenance of ion gradients in cells for plants more difficult.
This definition has become somewhat of a rallying cry for community and population centered ecology. Clearly, this definition has not stimulated exploration of the frontier of ecology with the sciences of the physical environment. Odum began with the Haeckelian definition, but his desire to establish a new kind of ecology -- ecosystem ecology -- led him further from that cornerstone than most.
He provided several statements of the scope of ecology, including the difficult-to-interpret statement that ecology was simply environmental biology.
Truest to his brand of ecosystem thinking was his definition of ecology as the study of the structure and function of nature. Although Odum's extreme reliance on emergent properties and resuscitation of superorganismic thinking have proven problematic to many ecologists, his loosening of the bonds of Haeckel's focus on the organism is useful.
The positive side of the first definition is that it is simple and it emphasizes both biotic and abiotic aspects of nature. On the negative side is its overemphasis on the organism as the focus. Haeckelian statements should always be cast as the study of relationships rather than the study of organisms in relation to environment.
The difference in emphasis may appear to be minor, but it indicates the deficiency of Haeckel's definition. The second definition is positive in its emphasis on quantifiable and unambiguous parameters, but it falls short because it omits a range of critical ecological subjects.
To its credit, the third definition is not restricted to patterns or organisms and recognizes that ecology is about processes. All of the definitions take organisms as their starting point. Upcoming SlideShare. Like this presentation?
Why not share! Embed Size px. Start on. Show related SlideShares at end. WordPress Shortcode. Next SlideShares. Download Now Download to read offline and view in fullscreen. Download Now Download Download to read offline. M Nadeem Akram Follow. Impact of exotic Species On Local Climate.
Rabbit Colour Genetics. Gene and Human Clonning. Vegetation of Pakistan. Ethnobotany - Relation between Plants and Human. Replication In Eukaryotes and Prokaryotes.
Population Structure. Insectivorous Plants. Protein sequencing. Plant anatomy presentation. Related Books Free with a 30 day trial from Scribd.
Elizabeth Howell. Ralph Eubanks. Logicomix: An epic search for truth Apostolos Doxiadis. Related Audiobooks Free with a 30 day trial from Scribd. Jen Gunter. A Wild Idea Jonathan Franklin. Midhun ScherZinger.
Tahira Rai.
0コメント