Fisheries

The depletion of fish stocks on a global scale is well documented. The United Nations Food and Agriculture Organisation collects statistics on fisheries from all states and, despite obvious shortcomings in the data, a clear picture has been available for some time. Garcia and Grainger (2005) have succinctly documented the position from the latest available date: in 2003, only 3% of stocks were underexploited and 26% moderately exploited, while 52% were fully exploited, 16% were overfished, 7% were depleted, and 1% were recovering from earlier depletion. These global statistics mask two important phenomena. The first, highlighted by Pauly et al. (1998), is that fisheries are increasingly focusing on species lower down in the food-web and the second, highlighted by Myers and Worm (2003, 2005), is that large predatory fish have been particularly reduced in abundance. Both of these analyses are somewhat flawed. In the case of Pauly et al. there are two problems: the first is that the metrics used for the mean trophic level are presented as simple numbers with no estimates of error or indeed sensitivity. In such a situation, the changes in mean trophic levels are hard to interpret, particularly where the mean trophic level changes by at most around 10% over four decades. The second problem has been highlighted by a recent paper by Essington et al. (2006). They point out that in the periods when according to the analysis of Pauly et al. the mean trophic level was declining, in most cases catches of apex predators and indeed all upper trophic levels increased (an exception is the North Atlantic). In the case of the Myers and Worm analysis, they used the catch per unit of effort (CPUE) as an index of abundance. As discussed later in this chapter, there are problems with this, but more importantly for some key apex predators, in particular large tunas, the CPUE declines in the early stages of the fishery, where catches are small, but remains relatively stable under a regime of much higher catches. In such a situation, the interpretation that the CPUE reflects changes in abundance is clearly problematic.

Communities: patterns

In all areas of ecology, from studies of individual organisms through populations to communities and ecosystems, there have been huge empirical and theoretical advances over the past several decades. Our guess—a testable hypothesis—is that the worldwide research community of ecologists has grown by roughly an order-of-magnitude since the 1960s, as is of course true for other areas of the life sciences. One consequence is that it is harder to put together a book like the present one. And we find it especially hard when we compare this chapter on community patterns with the corresponding chapter in the second edition of Theoretical Ecology. For the chapters on single populations, for example, there has been growth both in understanding the nonlinear dynamical phenomena that can arise, along with a host of well-designed field and laboratory experiments which illustrate these processes. The narrative, however, retains a unifying central thread, and much of the task of overview and compression lies in choosing good examples from an increasing panoply of choice. For communities, on the other hand, we find so many different yet intersecting areas of growth, many of which have recently produced booklength collections of papers, that the task of choosing which topics to emphasize and which to elide is invidious. The result is necessarily quirky. Without further apology, here is an outline. One broad area of community ecology deals with models for the dynamical behaviour of collections of many interacting species—either within a single trophic level or more generally—essentially as a scale-up of models for single and pairwise- interacting populations. This was the subject of the preceding chapter. Here, we begin by emphasizing the importance of work which views communities from, as it were, a plumber’s perspective, looking at patterns of flow of energy or nutrients or other material. But we then move on quickly to other topics. These include: the network structure of food-webs (connectance, interaction strengths, etc.); what determines species’ richness (niche versus null models); relative abundance of species (observed patterns and suggested causes); succession and disturbance; species–area relations; and scaling laws (with suggested connections among some such laws).

Unanswered questions and why they matter

The earlier chapters in this book could be thought of as travel notes from an intellectual journey across the landscape of ecological science. In particular, the previous five chapters, 10–14, implicitly or explicitly indicate some of the unintended consequences of the growth in numbers of people and in their environmental impacts. In this final chapter, I begin with a survey of some quantitative measures of the scale of human impacts. Emphasizing the many lamentable uncertainties in our knowledge base, I focus especially on the rising rates of extinction of plant and other animal species. Why should we care about such impoverishment of our planet’s biological diversity? I outline three kinds of possible reasons, under the headings of narrowly utilitarian, broadly utilitarian, and ethical. Each of these is then discussed, with emphasis on ways in which current lack of knowledge—lack of data and/or lack of theoretical understanding—is a handicap. In places, this carries the discussion into areas not commonly found in ecology texts (ethical, economic, and political questions, for instance). In other places, there is the more familiar exhortation for more research on this or that topic. Contrary to some impressions, human population growth has been far from simply exponential. Broadly speaking, humans have been around for a couple of hundred thousand years (Deevey, 1960; Cohen, 1995). For essentially all this time, they were small bands of hunter-gatherers, with the total human population being variously estimated at around 5–20 million people. With the benefits of the invention of agriculture, roughly simultaneously in various parts of the world around 10 000 years ago, things started to change. Denser aggregations of people became possible, and villages began their journey to cities. Following the advent of this agricultural revolution, human populations arguably grew more rapidly in the first 5000 years than in the more recent 5000, up to the beginning of the Scientific- Industrial Revolution around the 1600s. This relative slowing of population growth is almost surely associated with infectious diseases which were not sustainable at the low population densities associated with hunter-gatherers.

A Doubly Green Revolution: ecology and food production

Ecology has informed and underpinned agricultural production since the first faltering steps in domestication and cultivation. When someone (probably a woman) living in the Fertile Crescent carried seeds of wild wheats and barleys from the great natural cereal stands of the region and sowed them near her house she initiated the process of domestication. She also began the process of crop cultivation, creating what were to become ecologically complex, home gardens. Similarly swidden agriculture was based on imitations of ecological processes that would create a sustainable form of agriculture. The first articulation of this concept was not for many thousands of years later. The great Roman writer and agriculturalist of the first century bc, Marcus Terentius Varro, wrote as follows (Hooper and Ash, 1935) : . . . Agri cultura est ‘Non modo est ars, sed etiam necessaria ac magna; eaque est scientia, quae sint in quoque agro serenda ac facienda, quo terra maximos perpetuo reddat fructus’ . . . Agriculture is ‘not only an art but an important and noble art. It is, as well, a science, which teaches us what crops are to be planted in each kind of soil, and what operations are to be carried on, in order that the land may regularly produce the largest crops.’ (Varro, Rerum Rusticarum I, III) Not only does Varro place crops in their environment but the phrase quo terra maximos perpetuo reddat fructus (which can be translated as ‘that the land yields the highest in perpetuity’) struck me, when I first came upon it in one of the little red Loeb Classical Library translations, as an extraordinarily clear, elegant, and concise definition of sustainability. In this chapter I want to illustrate how ecological concepts illuminate the building blocks of agriculture—gardens, swiddens, pastures, orchards, and fields—and provide a basis for the continuing challenge of feeding everyone in an increasing population. The transformation of an ecosystem into an agroecosystem involves a number of significant changes. The system itself becomes more clearly defined, at least in terms of its biological and physico-chemical boundaries. These become sharper and less permeable, the linkages with other systems being limited and channeled.

Diversity and stability in ecological communities

How the diversity of an ecological community affects its stability is an old and important question (Forbes, 1887; Elton, 1927; Nicholson, 1933). The science of ecology grew out of the study of natural history in the nineteenth century, when nature was viewed as wondrous, mysterious, complex, and largely in balance (even if murderous to experience from an individual’s point of view; Forbes, 1887). Whereas our current scientific view is more textured and guarded, the ‘balance of nature’ still permeates the popular press. Some vestiges also remain in the scientific literature. Over the last 100 years, conclusions about the relationship between ecological diversity and stability have varied wildly (May, 2001; Ives, 2005). The goal of this chapter is to show that these wildly varying conclusions are due largely to wildly varying definitions of both stability and diversity. To do this, I will take two tacks, one for stability and the other for diversity. For stability, I will give an abbreviated history of the changing definitions of stability, merging both empirical and theoretical studies. I make no pretence of being comprehensive, but will instead pick highlights that show how the definition of stability often changes from one study to the next. For diversity, I will present a theoretical model to illustrate how different ‘diversity effects’ on stability can be parsed out. This model shows in a concrete way how any theoretical study (and, for that matter, empirical study) necessarily makes a long list of assumptions to derive any conclusion about diversity and stability. The multiple definitions of stability, and the multiple roles of diversity, argue against any general relationship between stability and diversity. In the final section of the chapter, I will argue that understanding the relationship between diversity and stability requires the integration of theory and experiment. Theory is needed to define in unambiguous terms the meanings of stability and diversity. Experiments are needed to ground theory in reality. Unfortunately, rarely is this done. To present an abbreviated history of the changing definitions of stability, I will discuss theoretical and empirical studies side by side.

Metapopulations and their spatial dynamics

The study of metapopulation dynamics has had a profound impact on our understanding of how species relate to their habitats. A natural, if naïve, set of assumptions would be that species are to be found wherever there is suitable habitat that they can get to; that species will rarely, if ever, be found in unsuitable habitat; that they will be most abundant in their preferred habitat; that species can be preserved as long as a good-size chunk of suitable habitat is conserved for them; and that destruction of a species’ habitat is always detrimental for its abundance. We will see that none of these reasonable-sounding assumptions is necessarily true. Metapopulation biology is a vast field, so to focus this chapter I will be guided partly by questions relevant to conservation biology. There are two important kinds of metapopulation. The so-called Levins metapopulation idea (Levins, 1970) is illustrated in Figure 4.1. It is imagined that patches of habitat suitable for a species are distributed across a landscape. Over time, there is a dynamical process of colonization and extinction: the colonization of empty patches by occupied patches sending out colonizing propagules and the extinction of local populations on occupied patches. This extinction can occur for a number of reasons. Small populations are prone to extinction just by the chance vagaries of the environment, reproduction, and death—environmental and demographic stochasticity (May, 1974b; Lande et al., 2003). An example of a species for which this is important is the Glanville fritillary butterfly (Melitaea cinxia), which has been extensively studied by Hanski and colleagues (Hanski, 1999). This Scandinavian butterfly lives in dry meadows which are small and patchily distributed. Another reason for local population extinction is that the habitat patch itself may be ephemeral. For example, wood-rotting fungi will find that their patch ultimately rots completely away (Siitonen et al., 2005) and epiphytic mosses will ultimately find that their tree falls over (Snall et al., 2005). The second type of metapopulation consists of local populations connected by dispersal, but without the extinction of the local populations.

Conservation biology: unsolved problems and their policy implications

A plot of the number of parks and other terrestrial protected areas established around the world over the past 100 years exhibits near-exponential growth, with marine parks following a similar trend. This is a testament to the growing recognition of the importance of sustaining natural systems worldwide. Yet, at the same time an expanding human population and the desire of all people for a more prosperous life have resulted in unprecedented rates of deforestation and habitat conversion. Accompanying these changes has been the spread of invasive, non-native species (including new disease organisms) to virtually all parts of the globe. With recent assessments placing 12% of the world’s birds, 23% of mammals, and 32% of amphibians in danger of extinction (Baillie et al., 2004), conservationists feel a justifiable sense of panic. Any attempt to measure the full extent of the current biodiversity crisis is made immensely more difficult by our astounding lack of knowledge about the species that share this planet with us. For example, we do not know within an order of magnitude the number of species currently present on Earth (May, 1988, 1992; Novotny et al., 2002); estimates range from 3 to more than 30 million species, of which only 1.5–1.8 million have been described to date. Not surprisingly, our inventory of the more charismatic groups of organisms, such as birds, mammals, and butterflies, is vastly more complete than our inventory of insects, arachnids, fungi, and other less conspicuous but no less important groups. If we ask the logical follow-up question—what proportion of known (described) species is in danger of extinction?—we run into a similar barrier. While organizations like the World Conservation Union (IUCN) have prepared reasonably complete assessments for a few groups, notably the charismatic vertebrates, most species are too poorly known to assess. Even within the USA only about 15% of the species catalogued to date are sufficiently known to be given any sort of conservation rank, such as endangered or not endangered (Wilcove and Master, 2005); among invertebrates that value drops to less than 5%. Compounding this shortfall of data is an equally serious shortfall of money.

Dynamics of infectious disease

Host–pathogen associations continue to generate some of the most important applied problems in population biology. In addition, as foreshadowed in Chapter 5 of this volume, these systems give important insights into the dynamics of host– natural enemy interactions in general. The special place of pathogens in the study of host–natural enemy dynamics arises partly from excellent longterm disease-incidence data, reflecting the public health importance of many infections. However, we argue that host–pathogen dynamics are also distinctive because the intimate association between individual hosts and their pathogens is often reflected with particular clarity in the associated population dynamics. Throughout this chapter we focus in parallel on the population dynamics of host–pathogen interactions and the insights that host–pathogen dynamics can provide for population biology in general. Population-dynamic studies of infectious disease have a long history, which predates the modern foundations of ecology (Bernoulli, 1760). During the twentieth century, the preoccupation of population ecologists with the balance between extrinsic and intrinsic influences on population fluctuations and the role of nonlinearity and heterogeneity (Bjørnstad and Grenfell, 2001) find strong parallels in epidemiological studies of human diseases (Bartlett, 1956; Anderson and May, 1991). In terms of the ecological effects of parasitism, the traditional view held that ‘welladapted’ parasites would not have a consistent impact on the ecology of their hosts (Grenfell and Dobson, 1995). The 1970s saw a new departure, when Anderson and May pointed out the potential of infectious agents to exert nonlinear—regulatory or destabilizing—influences on the population dynamics of their hosts (Anderson and May, 1978, 1979; May and Anderson, 1978, 1979). There has since been an explosion of work on the population biology of human, animal, and plant pathogens. This work spans a huge range: from highly applied to basic theoretical work; from within-host to the metapopulation scale; from short-term population dynamics to long-term evolutionary processes. In this chapter we first outline the simple theory of epidemiological models; we then refine this picture to illustrate the potential impact of pathogens on the population dynamics of their hosts, as well as aspects of host–pathogen interactions which provide important insights into more general ecological dynamics.

Interspecific competition and multispecies coexistence

Interspecific competition is an interaction in which species inhibit each other such that increased abundance of one species leads to lower growth rates of the other species. Numerous field studies have shown that interspecific competition is a major force determining species abundances for a wide variety of taxa in many different ecosystems (Harper, 1977; Tilman, 1982; Connell, 1983; Schoener, 1983; Aarssen and Epp, 1990; Goldberg and Barton, 1992; Casper and Jackson, 1997; Miller et al., 2005). Predator–prey interactions can also be of simultaneous importance in determining the abundances and dynamics of species (e.g. Sih et al., 1986), as can host–pathogen interactions (e.g. Hassell and Anderson, 1989; Hochberg et al., 1990; Dobson and Crawley, 1995; Mitchell and Powers, 2003) and mutualistic interactions (e.g. Kawanabe et al., 1993; Richardson et al., 2000; Stachowicz, 2001). Although this chapter focuses on competition, all types of interaction operate simultaneously in nature. Much of the early and continuing interest in competition has centered on how so many competing species coexist. G.E. Hutchinson (1959, 1961) posed the paradox of the plankton, asking how 30 or more species of algae could coexist in a few milliliters of lake or ocean water when there were only one, two, or three limiting resources and when the open waters of lakes and oceans were so homogeneous because of wind-driven mixing. Theory predicted that no more species could coexist than there were limiting factors or resources (e.g. MacArthur and Levins, 1964; Levin, 1970; Armstrong and McGehee, 1980). The same paradox occurred for terrestrial plants and animals. The Earth’s 250 000 species of vascular plants compete for a few limiting factors (usually a subset of nitrogen, phosphorus, potassium, calcium, water, and light).Alarge part of their diversity can, of course, be explained by the heterogeneity seen along major continental-scale and smaller-scale spatial gradients (Tilman, 1988). Expressed another way, these 250 000 vascular species are spread among perhaps 50 different biomes that occur in each of the five major biogeographic realms of Earth. One might expect different species in different biomes because of their differing climates.

Plant population dynamics

Plants exhibit an extraordinary range of sizes and generation times, from single-celled algae with body sizes of the order of 5 mm and generation times of the order of 1 day, to massive forest trees more than 50 m tall that can live for over 1000 years. Diatoms and trees have the virtue of being easy to count, so it is natural to seek to model the dynamics of changes in numbers. On the other hand, many herbaceous perennials (like clonal herbs or turf-forming grasses) are difficult or impossible to count, and for these plants it is natural to model the dynamics of fluctuation in biomass or proportional space occupancy. The theory of plant population dynamics is linked to the rest of plant biology through a series of fundamental trade-offs, reflecting the fact that individual plants are constrained in what they can do. There are important trade-offs in reproduction because a plant could produce many small seeds or a few large seeds, but it is not an option to produce many large seeds. Other trade-offs involve investment decisions: for instance a plant can invest in growth or defence and this leads to a trade-off between competitive ability and palatability to herbivores. Alternatively, high growth rate in full sun may trade-off against a high death rate in low light (the cost of shade tolerance). An important set of trade-offs involve competing demands for resource capture. Thus a plant could invest in its root system to forage for phosphorus, or in its shoot system to forage for light, but it cannot maximise investment in competitive ability for light and soil nutrients. Finally, there is an important trade-off between competition and colonization because good dispersers tend to be inferior competitors; this is exemplified by the r-K continuum where colonizers (r strategists) have a set of traits like rapid generation time, small seeds, wind dispersal, and high light requirements, whereas late successional species (K strategists) tend to live longer, produce fewer, larger seeds, and to have more shade-tolerant, slowergrowing juveniles. Underpinning the theory of plant population dynamics is the invasion criterion, which states that all persistent populations must exhibit the tendency to increase when rare.