Scott D. Cooper

WHAT DETERMINES THE STRENGTH OF A TROPHIC CASCADE

Trophic cascades have been documented in a diversity of ecological systems and can be important in determining biomass distribution within a community. To date, the literature on trophic cascades has focused on whether and in which systems cascades occur. Many biological (e.g., productivity : biomass ratios) and methodological (e.g., experiment size or duration) factors vary with the ecosystem in which data were collected, but ecosystem type, per se, does not provide mechanistic insights into factors controlling cascade strength. Here, we tested various hypotheses about why trophic cascades occur and what determines their magnitude using data from 114 studies that measured the indirect trophic effects of predators on plant community biomass in seven aquatic and terrestrial ecosystems. Using meta-analysis, we examined the relationship between the indirect effect of predator ma- nipulation on plants and 18 biological and methodological factors quantified from these studies. We found, in contrast to predictions, that high system productivity and low species diversity do not consistently generate larger trophic cascades. A combination of herbivore and predator metabolic factors and predator taxonomy (vertebrate vs. invertebrate) explained 31% of the variation in cascade strength among all 114 studies. Within systems, 18% of the variation in cascade strength was explained with similar predator and herbivore char- acteristics. Within and across all systems, the strongest cascades occurred in association with invertebrate herbivores and endothermic vertebrate predators. These associations may result from a combination of true biological differences among species with different phys- iological requirements and bias among organisms studied in different systems. Thus, al- though cascade strength can be described by biological characteristics of predators and herbivores, future research on indirect trophic effects must further examine biological and methodological differences among studies and systems.

Resolving ecological questions through meta-analysis: Goals, metrics, and models

We evaluate the goals of meta-analysis, critique its recent application in ecology, and highlight an approach that more explicitly links meta-analysis and ecological theory. One goal of meta-analysis is testing null hypotheses of no response to experimental manipulations. Many ecologists, however, are more interested in quantitatively measuring processes and examining their systematic variation across systems and conditions. This latter goal requires a suite of diverse, ecologically based metrics of effect size, with each appropriately matched to an ecological question of interest. By specifying ecological mod- els, we can develop metrics of effect size that quantify the underlying process or response of interest and are insensitive to extraneous factors irrelevant to the focal question. A model will also help to delineate the set of studies that fit the question addressed by the meta- analysis. We discuss factors that can give rise to heterogeneity in effect sizes (e.g., due to differences in experimental protocol, parameter values, or the structure of the models that describe system dynamics) and illustrate this variation using some simple models of plant competition. Variation in time scale will be one of the most common factors affecting a meta-analysis, by introducing heterogeneity in effect sizes. Different metrics will apply to different time scales, and time-series data will be vital in evaluating the appropriateness of different metrics to different collections of studies. We then illustrate the application of ecological models, and associated metrics of effect size, in meta-analysis by discussing and/or synthesizing data on species interactions, mutual interference between consumers, and individual physiology. We also examine the use of metrics when no single, specific model applies to the synthesized studies. These examples illustrate that the diversity of ecological questions demands a diversity of ecologically meaningful metrics of effect size. The successful application of meta-analysis in ecology will benefit by clear and explicit linkages among ecological theory, the questions being addressed, and the metrics used to summarize the available information.

EFFECTS OF POPULATION DENSITY ON INDIVIDUAL GROWTH OF BROWN TROUT IN STREAMS

Some studies suggest that lotic populations of brown trout (Salmo trutta) are regulated through density-dependent mortality and emigration to the extent that mean growth rates of resident survivors are unrelated to trout densities. To test this, we studied the relationship between density and growth, mortality, and emigration of brown trout in two alpine streams and a set of stream channels in eastern California. We sampled trout at the scale of “segments” (5–31 m long riffles, runs, and pools) and “sections” (340–500 m in length) of Convict Creek over a 3-yr period. Trout were also sampled during 6 yr in seven 90-m sections of Mammoth Creek. For 2 yr, we manipulated trout densities in Convict Creek by removing trout from two sections and adding trout to two other sections. We also manipulated densities in seven 50-m stream channels, using a natural size distribution of trout in one year and underyearlings only in a second year. In both streams, average size (body length or mass) of underyearlings in fall...

Prey Exchange Rates and the Impact of Predators on Prey Populations in Streams

We present four lines of evidence that the magnitude of prey exchange (=immigration/emigration) among substrate patches has an overwhelming influence on the perceived effects of predators on prey populations. (1) An extensive review of the literature on predation effects in benthic and littoral freshwater revealed a significant relationship between prey exchange rate and observed predator impact. In streams, studies showing significant predator effects used cages with smaller mesh sizes than studies showing nonsignificant effects. Similarly, there was a highly significant correlation between cage mesh size and the magnitude of predator impact on common prey. Large—scale stream studies indicated that prey drift and colonization rate were inversely related to predator impact on benthic prey. (2) These patterns were confirmed by field experiments and observations where mesh size was directly manipulated or where exchange rates varied among taxa. In Colorado streams we saw greater predator impacts on Baetis prey when immigration/emigration was restricted vs. when the mesh size of the cage was relatively large. Similarly, the effects of trout in California stream pools were greater when prey turnover rates were low. (3) A re—analysis of Peckarskys (1985) data shows an inverse relationship between predator impact and prey mobility within a field experiment. (4) Finally, a model that incorporates both predation and exchange of prey indicates that we ought to expect a lower magnitude of predator effects when exchange rates are high. These results suggest that some discrepancies in past studies may be explained by differences in the exchange rates of prey, and that differences in predator effects across different systems or habitats may be related to variation in the rates of prey dispersal and colonization.

Effect size in ecological experiments: The application of biological models in meta-analysis

Some of the most interesting and important questions in ecology require examination of the strength of different processes across environmental gradients and among organisms with different traits (Quinn and Dunham 1983; Tilman 1989; Cooper et al. 1990; Sarnelle 1992; Osenberg and Mittelbach 1996). Metaanalysis (see, e.g., Gurevitch et al. 1992; Gurevitch and Hedges 1993; Arnqvist and Wooster 1995; Curtis 1996) combines results from independent studies to examine patterns of effect across taxa or environments and, thus, may represent a powerful tool to test ecological theory. A meta-analysis requires that a common metric of effect size be extracted from each of the studies. Here, we focus on choosing a metric that best facilitates ecological inferences. We begin with a brief description of the standard definition of effect size in meta-analysis, as used in recent papers. We then discuss potential problems with this approach and suggest an alternative that is more explicitly tied to the dynamics of ecological systems. Using two examples drawn from predator-prey experiments, we then illustrate the limitations of the standard metric and the conceptual advantages of one ecologically based alternative. We conclude by discussing the link between metrics of effect size and ecological models.

The effect of physical disturbance on the relative abundances of two filter-feeding insects in a small stream

SummaryWe examined the importance of disturbance in determining the relative abundances of two lotic filter-feeders, Simulium virgatum and Hydropsyche oslari, in a small, coastal stream in southern California, USA.In most years, winter spates effectively scour substrata in fast-flowing areas, thereby drastically reducing stream insect populations. Newly-opened space in these areas is quickly colonized by simuliids. The abundance of simuliids, however, gradually declines as hydropsychid abundance increases in early summer. To determine if these changes in insect abundance represent seasonal changes or successional changes following disturbance, we performed a field experiment where hard substrates were disturbed at 2 wk, 4 wk, or 8 wk intervals, or were left undisturbed. We found that the numbers of simuliids increased and the numbers of hydropsychids decreased as the frequency of disturbance increased. Although seasonal recruitment patterns and longitudinal position in the strem had important effects on the colonization rates of these insects, time since last disturbance was a prime determinant of the relative abundances of Simulium and Hydropsyche. These results and additional observations suggest that Simulium virgatum is an opportunistic species that quickly colonizes new space, but that it is displaced by the slower-colonizing but competitively superior Hydropsyche oslari. Disturbance promotes the coexistence of these two species by preventing the attainment of a climax state where Hydropsyche monopolizes available space.

THE IMPORTANCE OF DATA-SELECTION CRITERIA: META-ANALYSES OF STREAM PREDATION EXPERIMENTS

The value of meta-analysis in ecology hinges on the reproducibility of patterns generated by quantitative synthesis. Meta-analysts will vary in the criteria they use to screen studies and select data within studies, even when addressing exactly the same question. We summarize some of the many decisions that an ecologist must make in deciding what data to include in a synthesis. We then show, using multiple meta-analyses taken from the same literature on stream predation experiments, that meta-analytic conclusions can be colored by selection criteria that are not specifically a function of the relevance of the data. As a consequence, we recommend that meta-analysts perform several meta-analyses using different selection criteria to examine the robustness of reported findings. We also advise ecological meta-analysts to minimize use of selection criteria that are based on judgments of study quality when extracting data from the literature, because of the potential for unconscious bias. The influence of quali...

Effects of Macroalgae on a Stream Invertebrate Community

The effects of macroalgae on stream invertebrates were studied in riffle zones in a coastal southern California stream. Dense growths of macroalgae (Cladophora glomerata, Nostoc sp.) were experimentally removed at different times of the year and in different years, and the insect communities which developed were compared with those in unmanipulated controls. Marl precipitated by algae was also removed in one experiment. The presence of macroalgae was associated with greater total densities and taxon richness of invertebrates, and nearly all taxa responded significantly to algal removal on at least some dates. Insects formed most of the community and were classified according to three categories of macroalgal effects on benthic densities: 1. Negatively affected by macroalgae (and marl) due to competition for space--e.g., Blepharicera (strong response to both algal taxa); large Simulium (strong with Cladophora, weak with Nostoc). 2. Positively affected due to structural habitats created by algae--e.g., Micrasema, Rhyacophila and Hydropsyche (all strong); Tinodes (weak-Nostoc); Rheotanytarsus (strong-Nostoc). 3. Positively affected by both macroalgal structure and associated food resources (macroalgae or epiphyton)--e.g., Baetis and Chironomidae (strong-Cladophora, weak-Nostoc); Hydroptila, Ochrotrichia, and Euparyphus (strong-Cladophora); and endosymbiotic Cricotopus (strong-Nostoc). Natural disturbances will indirectly affect invertebrate distributions and abundances by affecting the distributions and abundances of macroalgae.

Scale effects and extrapolation in ecological experiments

Abstract Most ecological experiments are performed on spatial and temporal scales that are much smaller and shorter than the systems and time frames of interest. Available data, however, suggest that experimental results often change with the size of the experimental arena and the duration of the experiment (i.e. are scale-dependent). As a consequence, the interpretation of experimental results often requires extrapolation from the limited spatial and temporal scales of experimental systems to the much larger and longer scales of natural systems. In this paper we discuss the implications of scale-dependence, particularly spatial scale-dependence, for the design and interpretation of experiments. We suggest that the problem of extrapolation across scales should be avoided when possible, either by matching the physical size of experimental units with the size of the system of interest or by designing small-scale experimental systems so that the processes of interest are given a realistic representation. When this is not possible it becomes necessary to translate experimental results to other scales, which requires that the mechanisms that generate scale-dependence are understood and that they can be incorporated into models that make predictions for other scales. We review and classify sources of scale-dependence in ecological responses to perturbations and describe attempts to incorporate these mechanisms into scaling models. Among the mechanisms we describe are exchange processes, nonlinear averaging in heterogeneous systems, and arena artifacts. At present, we do not know if available scaling models can make accurate quantitative extrapolations from experimental to natural scales. Thus, the primary, current value of scale models is the identification of scale ranges with particularly weak or strong scale-dependence. We also note that well-known statistical methods for design, parameter estimation and inference can be used as a framework for extrapolation in field experiments, in the sense that observations from a small number of experimental units can be used to draw conclusions about whole systems. We discuss the value of different statistical designs as tools for extrapolation and note that the choice of scale of an experiment is a critical design decision. The scale of a design is determined by grain (size of experimental units or blocks) and the extent or range covered by the design. The scale range, delimited by grain and extent, determines the scale of the background heterogeneity that can influence the strength of treatment effects. Moreover, both again and extent are related to the variance among experimental units, which means that the choice of scale influences the statistical power of a design as well as the magnitude of the aggregation error, a bias that can arise when the mean value of a set of measurements made in small experimental units are taken to represent a larger, more heterogeneous system.

How to Avoid Train Wrecks When Using Science in Environmental Problem Solving

I collaborations are increasingly common in many areas of science, but particularly in fields involved with environmental problems. This is because problems related to human interactions with the environment typically contain numerous parameters, reflect extensive human alterations of ecosystems, require understanding of physical–biological interactions at multiple spatial and temporal scales, and involve economic and social capital. Distilling useful scientific information in collaborative interactions is a challenge, as is the transfer of this information to others, including scientists, stakeholders, resource managers, policymakers, and the public. While this problem has been recognized by historians and philosophers of science, it has rarely been recognized and openly discussed by scientists themselves (but see NAS 1986). The participation of individuals from a diverse set of scientific disciplines has the potential to enhance the success of problem solving (USGS/ESA 1998). However, obstacles often arise in collaborative efforts for several well-known reasons. First, it is often difficult to find a common language because of disciplinary specialization (Wear 1999, Sarewitz et al. 2000). Second, existing scientific knowledge (theories, models, etc.) may reflect a historical scientific and sociopolitical context that may make it ill suited to address current environmental problems and questions (see, for example, Ford 2000, NSB 2000). Third, collaborations involving multiple disciplines may create difficulties owing to mismatches in space and time scales, in forms of knowledge (e.g., qualitative versus quantitative), and in levels of precision and accuracy (see, for example, Herrick 2000). Fourth, scientists are partly conditioned by nonscientific values. A social fabric may dictate scientists’ worldviews, lead them to favor certain assumptions over others, and underlie the way they study ecosystems (Boyd et al. 1991). In this article, we argue that the success of interdisciplinary collaborations among scientists can be increased by adopting a formal methodology that considers the structure of knowledge in cooperating disciplines. For our purposes, the structure of knowledge comprises five categories of information: (1) disciplinary history and attendant forms of available scientific knowledge; (2) spatial and temporal scales at which that knowledge applies; (3) precision (i.e., qualitative versus quantitative nature of understanding across different scales); (4) accuracy of predictions; and (5) availability of data to construct, calibrate, and test predictive models. By definition, therefore, evaluating a structure of knowledge reveals limitations in scientific understanding, such as what knowledge is lacking or what temporal or spatial scale mismatches exist among disciplines. The epistemological exercise of defining knowledge structures at the onset of a collaborative exercise can be used to construct solvable problems: that is, questions that can be an-