Do Trophic Interactions Cause Population Cycles?

My motivation in editing this book has been to present as compelling and credible a story as possible. Although I am personally convinced of the soundness of our argument, that food web architecture plays a key role in the cyclic dynamics of many animal populations, I am not sure that others will be so convinced. In this final chapter, therefore, I exercise my prerogative as editor to have the last word, a final attempt to convince the skeptics and to answer the critics.Perhaps the most compelling case comes from the Mikael Münster-Swendsen monumental study of a needleminer infesting Danish spruce forests (chapter 2). Mikael is the only person I know of who has, almost single-handedly, and with considerable precision, measured all the variables suspected of affecting the dynamics of a particular population over an extended period of time (19 years) and in several different localities (seven isolated spruce stands). Others have longer time series from more places, but none has been so complete in terms of the number of variables measured. This exhaustive study enabled him to build a model of the complete needleminer life system, and use this model to home in on the factors responsible for the cyclical dynamics. However, the story would not have been complete without multivariate time series analysis, which led to the discovery of parasitoids as the cause of the key feedback process, density-related reduction in fecundity. The lesson from Münster-Swendsen's work is clear: If we want to understand population dynamics, we need long time series for all the variables likely to affect the dynamics of the subject population(s). In other words, we need to consistently monitor ecological systems over long periods of time and in many different locations. If there is a weakness in his study, it is the absence of the final definitive experiment. Such an experiment would be relatively easy and cheap to do (relative to those described in other chapters), because isolated spruce stands are common in Denmark and parasitoids emerge from the soil a week or two after the needleminer. Thus, parasitoids could easily be excluded by spraying the ground with an insecticide after needleminer emergence.

Parasitic Worms and Population Cycles of Red Grouse

Many years before Charles Elton collected the detailed data on fur returns to The Hudson’s Bay Trading Company, or described the regular fluctuations in small mammal numbers, scientists and naturalists had observed and were proposing explanations for the cause of periodic crashes in numbers of red grouse known as “grouse disease.” MacDonald (1883) claimed “that it was more than eighty years since the alarm of grouse disease was sounded in this country,” implying that naturalists were starting to examine the phenomenon nearly 200 years ago. In 1873, The House of Commons established a Select Committee to consider the game laws of the United Kingdom and, since this had followed a year of particularly severe population collapse in red grouse numbers, they took exhaustive evidence on a wide range of possible causes of “grouse disease.” An examination of the letters in The Times and The Field shows that the debate over the cause of the population crashes was contentious and as heated as many of the recent debates over the causes of population cycles. Scientific studies were initiated by Cobbold (1873) who examined grouse killed during a population crash, published a pamphlet that described the presence of large numbers of “strongle worms,” and advocated the theory that the cause of grouse disease was wholly due to the presence of nematode worms. In 1905, the Board of Agriculture appointed a Committee of Inquiry on Grouse Disease to investigate the life history of the parasite and the causes of “grouse disease.” The extensive survey and detailed analysis was quite remarkable for the time, and was presented in a two-volume publication (Lovat 1911). The Committee surveyed grouse populations, undertook experiments and, after nearly 2000 dissections, came to the conclusion that “the strongyle worm, and the strongyle worm alone, is the immediate causa causans of adult ‘Grouse Disease.’“ The Principal Field Officer was E. A. Wilson, a gifted artist and scientist who was later appointed as the Scientific Director to Captain Scott’s Antarctic expedition on the Terra Nova. Unfortunately, Wilson never saw the production of the final report as he died with Scott during their return from the South Pole.

Population Cycles of the Autumnal Moth in Fennoscandia

Most species of insect herbivores are restricted to low densities, but some display large-scale density fluctuations, including periodic outbreaks (Faeth 1987, Mason 1987, Hanski 1990, Hunter 1995). The tendency to reach high densities has been related to certain life history traits (Hunter 1991, 1995, Tammaru and Haukioja 1996). However, all populations of a given outbreaking species do not necessarily display high densities. In those cases, outbreaks are frequently more pronounced in populations in physically severe and marginal habitats (Wallner 1987, Myers and Rothman 1995). The autumnal moth, Epirrita autumnata (Borkhausen) (Lepidoptera: Geometridae) is an example of a species with both outbreaking and nonoutbreaking populations. In mountain birch [Betula pubescens ssp. czerepanovii (Orlova) Hämet-Ahti] forests of northern and mountainous Fennoscandia (hereafter northern populations), E. autumnata displays fluctuations with a statistically significant periodicity of 9-10 years (Tenow 1972, Haukioja et al. 1988, Bylund 1995). During outbreaks, forests may be totally defoliated and trees may even die over large areas (Tenow 1972, Lehtonen and Heikkinen 1995). In more southern parts of the species' Holarctic distribution (hereafter southern populations), outbreaks are absent and populations remain at low densities. Cycles of northern E. autumnata populations vary in their amplitude (Tenow 1972). Outbreak densities that produce conspicuous defoliation are typically reached in only some areas, and often in different areas during successive peaks (Tenow and Bylund 1989). Empirical data indicate a fairly regular pattern of fluctuations, that is synchronous on a regional scale, also in populations with moderate or low peak densities (Bylund 1997). Thus, there are two main questions regarding population regulation of northern and mountainous E. autumnata—what causes the cycles, and what causes spatial variations in outbreak severity? In southern populations, the main question is what prevents outbreaks? Larvae of E. autumnata hatch early in spring at the time of birch bud break. Birches (Betula spp.) are the main host plants, although larvae are able to feed on many deciduous trees and shrubs (Seppänen 1970).

Population Cycles of the Larch Budmoth in Switzerland

The population dynamics of the larch budmoth (LBM), Zeiraphera diniana, in the Swiss Alps are perhaps the best example of periodic oscillations in ecology (figure 7.1). These oscillations are characterized by a remarkably regular periodicity, and by an enormous range of densities experienced during a typical cycle (about 100,000-fold difference between peak and trough numbers). Furthermore, nonlinear time series analysis of LBM data (e.g., Turchin 1990, Turchin and Taylor 1992) indicates that LBM oscillations are definitely generated by a second-order dynamical process (in other words, there is a strong delayed density dependence—see also chapter 1). Analysis of time series data on LBM dynamics from five valleys in the Alps suggests that around 90% of variance in Rt is explained by the phenomenological time series model employing lagged LBM densities, R, =f(Ni-1,Ni-2,) (Turchin 2002). As discussed in the influential review by Baltensweiler and Fischlin (1988) about a decade ago, ecological theory suggests a number of candidate mechanisms that can produce the type of dynamics observed in the LBM (see also chapter 1). Baltensweiler and Fischlin concluded that changes in food quality induced by previous budmoth feeding was the most plausible explanation for the population cycles. During the last decade, the issue of larch budmoth oscillations was periodically revisited by various population ecologists looking for general insights about insect population cycles (e.g., Royama 1977, Bowers et al. 1993, Ginzburg and Taneyhill 1994, Den Boer and Reddingius 1996, Hunter and Dwyer 1998, Berryman 1999). These authors generally concurred with the view that budmoth cycles are driven by the interaction with food quality. A recent reanalysis of the rich data set on budmoth population ecology collected by Swiss researchers over a period of several decades, however, suggested that the role of parasitism is underappreciated (Turchin et al. 2002). Before focusing on the roles of food quality and parasitism in LBM dynamics, we briefly review the status of other hypotheses that were discussed in the literature on LBM cycles. First, the natural history of the LBM-larch system is such that food quantity is an unlikely factor to explain LBM oscillations.

Population Cycles Inferences from Experimental, Modeling, and Time Series Approaches

Some of the most interesting debates in population ecology have taken place within the context of population cycles. Their study has been a fertile ground for the development of ideas on how population models should be formulated and confronted with data. It is the setting in which the use of field experiments became established in ecology (e.g., Krebs and DeLong 1965), and also the context of many methodological and conceptual developments in the fields of population demography (Leslie and Ranson 1940), pest management (Berryman 1982), and community dynamics (Sinclair et al. 2000). Yet, as with many other issues in population dynamics, identifying without ambiguity the causes of population cycles in general, and for any organism in particular, continues to prove an extraordinarily difficult task. The major purpose of this book is to review recent research developments on the role of food web architecture, and more specifically on the effects of food, predators, and pathogens in population cycles. Its stated aim is to present evidence that population cycles could be caused by food web architecture in some natural systems. Whereas in chapter 1 Alan Berryman promotes a research program centered on the analysis of time series data for formulating, selecting, and even testing hypotheses on population cycles, the case studies encompass a much broader diversity of research approaches. The authors and coworkers of the seven case studies have combined time series analysis, model building, natural history observation, and experiments in different proportions to reach the conclusion that trophic interactions play an important role in generating cyclic dynamics. This diversity of approaches reflects, in part, a taxonomic divide between vertebrates and invertebrates, experiments being more common with the former, but also profound differences in research traditions. Indeed, the investment required to estimate population size and quantify the causes of mortality of moths and beetles is substantially less than that required for estimating the abundance of voles, hares, and grouse and their predators. From these practical constraints, divergent research traditions have evolved.

The Role of Insect Parasitoids in Population Cycles of the Spruce Needleminer in Denmark

The spruce needleminer, Epinotia tedella (Cl.) (Lepidoptera: Tortricidae), is a small and abundant moth associated with Norway spruce (Picea abies Karst.). Larvae mine spruce needles, usually those more than 1 year old, and each requires about 35 needles to meet its food demands. In central Europe, the spruce needleminer is regarded as a temporary, serious pest when densities reach several thousand per square meter. However, it seldom causes significant damage in Scandinavian countries. An exception was the heavy infestation in southern Denmark in 1960-61. The spruce needleminer has one generation per year. Adults emerge in June and deposit eggs singly on spruce needles. Larvae mine the needles from July through October and then descend on silken threads in November to hibernate in the forest litter as prepupal larvae in cocoons. Pupation occurs in early May and lasts 3-4 weeks. Like many other forest defoliators, spruce needleminers are associated with a diverse fauna of parasitic Hymenoptera (parasitoids) (Münster-Swendsen 1979). Eggs are attacked by a minute wasp (Trichogramma sp.) that kills the embryo and emerges as an adult a few weeks later. Because spruce needleminer eggs have all hatched by this time, the parasitoids must oviposit in the eggs of other insect species. In other words, this parasitoid is not host-specific and therefore not expected to show a numerical response to spruce needleminer population changes. Newly hatched moth larvae immediately bore into needles and, because of this, are fairly well protected against weather and predators. However, specialized parasitic wasps (parasitoids) are able to deposit their eggs inside a larva by penetrating the needle with their ovipositor. Two species, Apanteles tedellae (Nix.) and Pimplopterus dubius (Hgn.), dominate the parasitoid guild and sometimes attack a large percentage of the larvae (Münster -Swendsen 1985). Parasitized larvae continue to feed and, in November, descend to the forest floor to overwinter with unparasitized individuals. In late April, however, the parasitoids take over and kill their hosts. Besides mortality from endoparasitoids, up to 2% of the larvae die within the mine due to an ectoparasitoid and a predatory cecidomyid larva.

Population Cycles Causes and Analysis

Ever since Elton’s classic book Voles, Mice and Lemmings (Elton 1942), understanding and explaining the causes of regular multiannual cycles in animal populations has been a central issue in ecology. Many hypotheses have been erected and incessantly argued about, but no clear picture has emerged. Below I briefly sketch the major hypotheses without any attempt to be complete or to comment on their relative merits or demerits. Detailed reviews and discussion can be found in Keith (1963), Krebs and Myers (1974), Finerty (1980), Myers (1988), Royama (1992), and Stenseth (1999). (H1) Physical effects (e.g., Elton 1924, Bodenheimer 1938). Perhaps the most obvious hypothesis is that cycles in animal populations reflect the response of birth and death rates to an external physical factor that is itself cyclic. Two of the more specific physical hypotheses involve periodic climatic factors and sunspot activity. (H2) Predator effects. Lotka (1924) and Volterra (1926) demonstrated that cyclic dynamics are inherent in simple predator-prey models, leading to the hypothesis that regular cycles can result from interactions between predator and prey populations. (H3) Pathogen effects. Anderson and May (1980) showed that, under certain conditions, simple models of infectious disease transmission can generate cycles in host and pathogen populations. This is similar to H2 with the pathogen as a predator. (H4) Plant effects. Several hypotheses have been proposed for the possible role of plants in generating population cycles of herbivores. One is a generalization of H2 in which the plant is considered the prey and the herbivore the predator (Elton 1924, Pitelka 1957). Another involves nutrient cycling: In this hypothesis, nutrient deficiencies are assumed to reduce the resistance of plants, resulting in larger herbivore populations, but nutrients released in feces and decaying animal and plant matter cycle back to the plants, increasing their vigor and resistance, and resulting in reduced herbivory (e.g., White 1974). Another hypothesis argues that herbivore feeding induces sustained chemical and/or physical changes in the plant (delayed induced resistance), which then reduce the reproduction and/or survival of future herbivore generations (Benz 1974, Haukioja and Hakala 1975).

Evidence for Predator-Prey Cycles in a Bark Beetle

The southern pine beetle, Dendroctonus frontalis Zimmermann (Coleoptera: Scolytidae), is an economically important pest of pine forests in the southern United States (Price et al. 1992). This native bark beetle is able to attack and kill living trees, typically loblolly (Pinus taeda L.) or shortleaf (Pinus echinata Mill.) pine, through a process of mass attack coordinated by pheromones emitted by the beetle (Payne 1980). During the attack process, thousands of beetles bore through the outer bark of the tree and begin constructing galleries in the phloem layer. Trees can respond to beetle attack by exuding resin from a network of ducts, but the large number of simultaneous attacks usually overcomes this defense, literally draining the resin from the tree. Oviposition and brood development then occur in the girdled (and ultimately dead) tree. Once a tree is fully colonized the attack process shifts to adjacent trees, often resulting in a cluster of freshly attacked trees, trees containing developing brood, and dead and vacated trees (Coulson 1980). These infestations can range in size from a single tree to tens of thousands, although the latter only occur in areas where no control methods are applied. Approximately six generations can be completed in a year in the southern United States (Ungerer et al. 1999). Like many other forest insect pests, D. frontalis populations are characterized by a considerable degree of fluctuation. The longest time series available are Texas Forest Service records of infestations in southeast Texas since 1958 (figure 5.la). These data suggest that the fluctuations have at least some periodic component, with major outbreaks occurring at intervals of 7-9 years (1968, 1976, 1985, and 1992). A variety of different analyses, including standard time series analysis and response surface methodology (Turchin 1990, Turchin and Taylor 1992), suggest that D.frontalis dynamics are indeed cyclic and appear governed by some kind of delayed negative feedback acting on population growth (see chapter 1). This effect can be seen by plotting the realized per-capita rate of growth (R-values) over a year against population density in the previous year (figure 5.1b).

Understanding the Snowshoe Hare Cycle through Large-scale Field Experiments

The 10-year cycles of the snowshoe hare and lynx seen in Hudson’s Bay fur returns represent a classic example of cyclic population dynamics. Hare cycles have been the subject of time series analysis (Stenseth et al. 1998), population modeling (Royama 1992), and field experimentation (Keith and Windberg 1978, Krebs et al. 1986, Murray et al. 1997). However, only two studies have monitored hare populations in detail over at least one full cycle. The first of these was conducted in central Alberta, Canada, by Lloyd Keith and coworkers, and provided a detailed description of the demographic machinery driving changes in hare numbers (Keith et al. 1977, Gary and Keith 1979, Keith et al. 1984). From this came the “Keith hypothesis” that hare cycles are driven by a sequential two-stage trophic interaction with hare declines initiated by winter food shortages and exacerbated by predator numerical responses that lag hare numbers by 1-2 years (Keith 1983, 1990). Predators force hares to low numbers and recovery does not occur until predator densities reach their lowest levels. The second long-term study of hare dynamics took place at Kluane Lake in the southwestern Yukon, Canada. The Kluane project began as an attempt to test the Keith hypothesis through single-factor manipulations of food supply and predation (Krebs et al. 1986, Sinclair et al. 1988, Smith et al. 1988). The first attempt failed to manipulate predators effectively, and plots containing food supplements were quickly overwhelmed by predators moving into the area. Consequently, the experiments failed to alter hare dynamics. Building on this experience, the second phase expanded the scale of experimental manipulations and developed an effective means of excluding predators from selected areas. The study also added an interaction treatment in which predators were excluded and food supplemented. These experiments were designed to test the roles of food supply, predation, and their potential interaction in the dynamics of snowshoe hares (Krebs et al. 1995). In this chapter we provide a synopsis of the key results obtained from these experiments and discuss how the results alter the current understanding of snowshoe hare dynamics.

Population Cycles of Small Rodents in Fennoscandia

The earliest records of small rodents in Fennoscandia date back to the sixteenth century. Ziegler (1532) and Magnus (1555) reported mass occurrences of the Norwegian lemming (Lemmus lemmus), which supposedly descended from the sky, a hypothesis that prevailed for the next 300 years (Henttonen and Kaikusalo 1993)! The first scientific papers on lemmings (Fellman 1848, Ehrstöm 1852) clearly recognized periodicity of lemming dynamics in Finnish Lapland (for a review see Henttonen and Kaikusalo 1993). Collett (1878, 1895, 1911-12) compiled extensive data on lemmings in Norway more than 100 years ago, providing critical material for Elton (1924) to describe the population cycle of small rodents. As these early records suggest, the Norwegian lemming is the most conspicuous member of the small rodent community in northern Fennoscandia, both in appearance and abundance, but apart from mountainous regions, the Fennoscandian small rodent cycle actually refers to Microtus and Clethrionomys voles rather than to lemmings. At present, the small rodent cycle in Fennoscandia is one of the best documented examples of cyclic population dynamics. Several recent papers review the state of knowledge on small rodent population dynamics in Fennoscandia and elsewhere (Norrdahl 1995, Krebs 1996, Boonstra et al. 1998, Stenseth 1999, Henttonen and Hanski 2000, Turchin and Hanski 2001). One might think that the “puzzle” of rodent cycles has been solved a long time ago, and that the Fennoscandian small rodent dynamics might serve as a useful reference for the study of cyclic populations in general. Unfortunately, this is not so, although substantial progress has been made over the past 15 years, so that we now have a well-supported hypothesis to explain the small rodent dynamics in Fennoscandia. There are several reasons why progress has been slow in unraveling the secrets of the small rodent cycle. First, small rodents occur in great abundance throughout the world and there was a tendency to assume that the rodent cycle, especially in northern latitudes, was a universal phenomenon, calling for a universal explanation (Krebs and Myers-1974). However, this is not so.