Read an Excerpt
The Phytochemical Landscape
Linking Trophic Interactions and Nutrient Dynamics
By Mark D. Hunter PRINCETON UNIVERSITY PRESS
Copyright © 2016 Princeton University Press
All rights reserved.
ISBN: 978-1-4008-8120-8
CHAPTER 1
Introduction
Imagine a world in which all primary producers have exactly the same chemistry. In this world, all autotrophs are equally palatable to herbivores, decompose at equal rates, and do not respond phenotypically to changes in the biotic and abiotic environment. There is no evolutionary change in plant chemical traits because there is no heritable variation upon which natural selection can act. Succession after disturbance is characterized only by the accumulation of equally palatable plant biomass. Algal blooms in lakes and oceans represent changes in the quantity, but not quality, of primary production. Our current vegetation-based classification of Earth's biomes is redundant.
The full ecological and evolutionary consequences of living in such a world are hard to imagine in part because we take for granted the enormous chemical variation that exists among the primary producers on Earth. Perhaps we are so familiar with the different autotrophs that characterize our forests, grasslands, and rocky intertidal zones, with the varying palatability of the species we can and cannot eat, that we are desensitized to the profound importance of that variation. As ecologists and evolutionary biologists, we encounter two major (and related) problems when we ignore the importance of variation in the chemistry of primary producers. First, our understanding of ecological and evolutionary processes is incomplete at best and misleading at worst. Second, we will be unable to predict the consequences of losing variation in plant chemical traits from our ecological communities. As human activities cause us to lose members of autotroph communities to extinction, we lose so much more than just scientific names. We lose the phenotypic variation that characterizes those scientific names, and their associated ecological and evolutionary interactions. We simplify the phytochemical landscape.
In this book, I describe how variation in the chemical traits of primary producers is of fundamental importance in ecology, and how that variation can be used as an organizing force for synthesis. Specifically, by focusing on variation in autotroph chemistry on the phytochemical landscape, we can better link studies of trophic interactions to those of ecosystem processes. Autotroph chemistry influences, and is influenced by, nutrient dynamics at the ecosystem level. Similarly, autotroph chemistry influences, and is influenced by, trophic interactions among organisms. The result is a series of feedback loops that link trophic interactions with the cycling of matter, mediated through the nexus of variation in autotroph chemistry on the phytochemical landscape (figure 1.1). These feedback loops represent powerful trait-mediated indirect effects (Werner and Peacor 2003) in which the chemical traits of primary producers link trophic interactions with nutrient cycling. In chapter 2, I introduce in more detail the concept of the phytochemical landscape, while in chapter 3 I describe the diversity of phyto-chemistry that exists in our ecological systems. In subsequent chapters (chapters 4–7), I provide evidence to support the importance of each of the feedback loops illustrated in figure 1.1. In chapter 8, I bring the individual feedback loops together into a synthetic whole. Finally, in chapter 9, I discuss priorities for future work and provide some testable predictions that arise from using the concept of the phytochemical landscape to understand the links between trophic interactions and nutrient dynamics.
1.1 A MATTER OF PERSPECTIVE
I doubt that there is a single best way to describe an ecological system. As scientists, we carry our experience, our skill sets, and our biases with us as we try to understand the ecological world. We also target our favorite research questions, which naturally influence the perspective that we take when we try to understand the world around us. Because of these diverse experiences, skill sets, biases, and questions, there are many complementary ways in which we can search for pattern and synthesis in ecology (Sterner and Elser 2002, Brown et al. 2004, Schmitz 2008b, Bardgett and Wardle 2010, Loreau 2010, McCann 2011, Cavender-Bares et al. 2012). I mention this explicitly because the approach that I offer here is one among many ways of seeking synthesis in ecology — it arises from my experience and my biases; it's a matter of perspective. My hope is that there is some strength in this perspective, an opportunity for insight that emerges specifically because of the approach that I describe. Other approaches, other perspectives, will offer their own unique insights into ecological processes. I view these varying perspectives as complementary, and not as alternatives. Moreover, figure 1.1 is meant to illustrate how trophic interactions and nutrient dynamics are linked through the nexus of autotroph chemistry. I do not mean to suggest that these feedback processes are the only factors influencing trophic interactions, the chemistry of primary production, and nutrient cycling. Many edaphic, climatic, and biological forces affect the components of figure 1.1, and can act as external influences on the feedback loops that are the focus of this book. These are discussed in chapter 9.
I also want to stress that this book is not about the primacy of plants. I am not suggesting that variation in the chemistry of primary producers in some way "drives" all of the ecological and evolutionary processes on Earth. Indeed, it is absolutely central to the arguments I present here that autotroph chemistry varies in direct response to nutrient dynamics at the ecosystem scale, and varies in response to trophic interactions. It is now abundantly clear that processes such as trophic cascades (Carpenter et al. 1985) (figure 1.1, arrow 2) and soil nitrogen flux (Aber et al. 1989) (figure 1.1, arrow 3) generate variation in the chemical traits of primary producers at landscape scales. These traits then feed back to influence species interactions and nutrient cycling (figure 1.1, arrows 1 and 4, respectively). Autotrophs are central only in the sense that variation in their chemical traits provides a nexus through which links between trophic interactions and nutrient cycling can take place. Specifically, in no sense am I suggesting that primary producers are more important than herbivores, predators, mutualists, or decomposers in the organization of ecological communities. Rather, I'm suggesting that to understand fully the role of trophic interactions in natural systems, we need to understand how differential consumption generates, and is generated by, variation in the chemistry of primary production.
1.2 THE NATURE OF FEEDBACK
Similarly, this is not a book about the relative importance of top-down and bottom-up forces in ecology. Rather, it is a book about feedback processes. Feedback loops play a central role in ecology at all levels of organization (Odum 1953, Royama 1992) and make irrelevant many arguments about primacy. Figure 1.2A shows an extreme example of a perfectly coupled system in which Consumer B consumes only Producer A. The dynamics of B are driven entirely from the bottom up, and the dynamics of A are driven entirely from the top down. The interaction as a whole cannot be characterized as bottom-up or top-down in any meaningful way. Figure 1.2B extends this to a system in which a variable plant trait (nitrogen content) varies in response to both herbivore damage and environmental nitrogen availability. The presence of feedback loops makes it impossible to characterize any part of the system as driven from the top down or the bottom up.
Why does this approach have merit? As the chemical traits of autotrophs respond to both nutrient cycling and trophic interactions, they establish pathways by which unexpected indirect effects emerge (Hunter et al. 2012). For example, figure 1.2C illustrates a pathway by which spatial patterns of environmental nitrogen availability could influence the dynamics between an herbivore host and its agent of disease. Likewise, disease dynamics between host and parasite may feed back to influence the spatial and temporal availability of nutrients in the environment. As disease ecologists learn more about the contingent nature of parasite-host interactions (Duffy et al. 2012, Orlofske et al. 2012), interactions between nutrient dynamics and agents of disease are starting to appear in the literature (Civitello et al. 2013, Lehahn et al. 2014). These kinds of indirect effects, which are experimentally tractable, can be predicted using the perspective shown in figure 1.1.
Feedback loops have other important properties. In the presence of the time lags inherent in ecological systems, feedback loops can generate complex temporal dynamics (May 1974). Ecologists still struggle to explain much of the temporal variation in species abundances and ecological processes that we observe in nature. But we should expect temporal variation to emerge in systems such as those illustrated by figure 1.2C. Critically, time lags in the responses of primary producer chemistry to trophic interactions or nutrient dynamics could generate complex temporal dynamics, including cycles and deterministic chaos (Royama 1992). It is a fundamental feature of many phytochemical traits that they are plastic, changing markedly in response to interactions with the biotic and abiotic environment. The timescales of such changes vary from a few seconds to years within individual autotrophs (Liechti and Farmer 2002), and to millennia within whole terrestrial plant communities (Petit et al. 1997). It is a prediction of the perspective described here that temporal variation in trophic interactions and ecosystem processes emerges in part from (a) feedback loops that include variation in autotroph chemistry, and (b) time lags in the responses of autotroph chemistry to abiotic and biotic interactions.
1.3 WHICH AUTOTROPHS AND WHICH TRAITS?
For simplicity, I use some shorthand throughout this book. I use the word "plant" or "autotroph" to include all primary producers, including autotrophic bacteria and archaea, and autotrophic members of the paraphyletic Protista (including diatoms, dinoflagellates, chlorarachniophytes, euglenids, brown algae, and the members of the Plantae that are not characterized as land plants). That being said, the ecological importance of variation in their chemical traits is much better known for some groups (e.g., land plants, cyanobacteria) than for others; the better-known groups will necessarily provide many of the examples in the book. Although the great majority of ecological processes presented here operate in both terrestrial and aquatic environments (chapter 9), there are some specific to each. I discuss relevant differences between terrestrial and aquatic habitats as they arise in each chapter.
I occasionally use the shorthand autotroph quality when focusing on the effects of biochemical traits on consumers or microbes. Autotroph quality is a tricky term because quality is a function of the user, not of the plant trait itself. For example, my daily consumption of caffeine and theobromine (in coffee and chocolate, respectively) would be lethal to most organisms, including many mammals (Gans et al. 1980). These particular alkaloid secondary metabolites have no inherent quality beyond their effects (positive or negative) on particular consumers (Scholey et al. 2010, Messerli 2012). However, there are a suite of chemical traits expressed by primary producers that appear to have significant and generalizable effects on nutrient cycling, trophic interactions, or both (Classen et al. 2007a). These traits include mineral nutrient concentrations and ratios (elemental stoichiometry), secondary metabolites (alkaloids, phenolics, terpenoids, etc.), and structural compounds such as lignin, cellulose, and suberin. These are the chemical traits on which I focus here. Some of them are particularly important in mediating interactions between consumers and nutrient cycling because they have broad biological activity. For example, depending on local chemical conditions, hydrolysable tannins can form complexes with proteins, form complexes with iron, and participate in redox reactions (Hartzfeld et al. 2002). Such chemical interactions are important across a broad spectrum of biological processes from nutrient absorption in the gut of mammalian herbivores to the decomposition of terrestrial plant litter.
The chemistry of primary producers varies across a broad range of temporal and spatial scales. In space, autotroph chemistry varies among tissues within single individuals, among individuals (including phytoplankton cells), among species within communities, among landscapes, and among biomes (Denno and McClure 1983, Hunter et al. 1992). This spatial variation generates a complex phytochemical landscape on which trophic interactions and nutrient dynamics take place. The phytochemical landscape might be imagined as a contour map of primary producer chemistry that varies from high to low mineral nutrient content, high to low toxicity to herbivores, and high to low recalcitrance to heterotrophic microbes that mineralize organic chemical compounds into inorganic forms. The phytochemical landscape has both chemical richness and chemical diversity and, as in traditional landscapes, patches of autotroph chemistry are differentially connected to one another through the flow of materials (chapters 2 and 9).
Moreover, the phytochemical landscape is not static over time. Autotroph chemistry can change within individuals over just a few seconds following herbivory (Zagrobelny et al. 2004). Additionally, the phytochemical landscape changes over weeks and years as the result of selective foraging (Pastor and Naiman 1992) and over centuries during the course of ecological succession and recovery from disturbance (Bishop 2002). Finally, the differential fitness of individuals that vary in chemical traits results in evolutionary change in the chemistry of primary producers on the phytochemical landscape (Ehrlich and Raven 1964), with consequences for subsequent trophic interactions and nutrient dynamics (Classen et al. 2007a). Variation in autotroph chemistry at all of these spatial and temporal scales are simultaneously causes and consequences of ecological interactions linked by feedback processes.
1.4 TRAIT VARIATION AND TRAIT DIVERSITY
There are two related yet separate ways by which variation in the chemical traits of primary producers can influence trophic interactions and nutrient cycling. First, key chemical traits in autotrophs can influence (and respond to) trophic interactions or nutrient dynamics. For example, damage by caterpillars to the leaves of many broad-leaved trees results in the induction of polyphenolic compounds in leaves and decreases in foliar nitrogen (N) levels (Rossiter et al. 1988, Tao and Hunter 2011). In turn, high concentrations of polyphenols and high lignin:N ratios in leaf litter can reduce rates of litter decomposition and nutrient cycling in soils (Madritch and Hunter 2002). In this example, variation in autotroph chemical traits (N, polyphenols) provides a pathway of interaction between insect herbivores and nutrient cycling (Hunter et al. 2012) (figure 1.3A). This is an example of how chemical trait variation can link trophic interactions and ecosystem processes.
Additionally, plant trait diversity within communities can influence both trophic interactions and nutrient cycling (figure 1.3B). Here, trait diversity refers to the richness of different chemical traits that are expressed simultaneously by communities of autotrophs on the phytochemical landscape. The expanding literature that links species diversity and ecosystem function (Cardinale et al. 2006) is in part a reflection of this. In this body of literature, "species diversity" is an easily measurable variable that serves as a proxy for the complex latent variables that actually determine functional phenotypic differences among taxa (Edwards et al. 2013). The quantification of Latin binomials (species diversity or richness) is convenient shorthand for the variation in phenotype, including variation in phytochemistry, which actually mediates ecological processes (Whitham et al. 2003). Whether described as species diversity, functional diversity, or phenotypic diversity, it is apparent that autotroph trait diversity can influence both trophic interactions (Haddad et al. 2009) and nutrient cycling (Kominoski et al. 2007). Similarly, Latin binomials represent phylogenetic relationships among species that share common ancestors and common ancestral traits. Consequently, a trait-based approach may better serve to explain relationships among the phylogenetic diversity of primary producers, species interactions, and ecosystem processes wherein the assembly of autotroph communities is viewed as the assembly of chemical phenotypes (Asner et al. 2014).
(Continues...)
Excerpted from The Phytochemical Landscape by Mark D. Hunter. Copyright © 2016 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
