I am honored by the Editors’ decision to put together this Virtual Issue of some of the papers I have published in the Journal of Ecology and grateful for this opportunity to reflect on them—it has been many years since I have read most of these papers. I have been fortunate to work with an extremely talented group of students, postdoctoral fellows, and collaborators over the years; much of this work is more due to their efforts than my own. The ten papers provide a record of the major themes we have pursued and, without too much distortion, can be put into three groups.
1. Defining competitive ability and discerning associated traits
Goldberg, D.E. and L. Fleetwood. 1987. Competitive effect and response in four annual plants. Journal of Ecology 75:1131-1143.
Goldberg, D.E. and K. Landa. 1991. Competitive effect and response – hierarchies and correlated traits in the early stages of competition. Journal of Ecology 79:1013-1030.
Baraloto, C., P.M. Forget, and D.E. Goldberg. 2005. Seed mass, seedling size and neotropical tree seedling establishment. Journal of Ecology 93:1156-1166.
My earliest papers in Journal of Ecology were associated with my attempt to clarify the definition of competitive ability by distinguishing between competitive effect and competitive response. As part of my dissertation work at the University of Arizona investigating the mechanisms underlying vegetation patterns in the Sierra Madre of northern Mexico, I had tested the hypothesis that evergreen species were poorer competitors than deciduous species by comparing response of seedlings of both to the presence/absence of surrounding vegetation in field experiments. While this made obvious sense in the context of the problem I was addressing, it didn’t match with many experiments in the literature where competitive ability was measured as the ability to suppress other plants. Musing about this led to the obvious-in-retrospect distinction between competitive effect and response. Although sometimes portrayed as distinct measurements, they are better defined in terms of how the same measurement is compared. Competitive response compares the degree to which different “target” taxa are suppressed by neighbours, while competitive effect compares the degree to which different “neighbor” taxa suppress targets. I first published this distinction in 1983 in the American Journal of Botany, where I also argued that competitive effects and responses should be measured and compared on a per-unit size basis using an additive design and that different traits would likely be correlated with competitive effect vs response. The 1987 Journal of Ecology paper was my first attempt to assess the latter and was based on an experiment with four species conducted with Linda Fleetwood, an undergraduate student at the time. The 1990 paper was a more comprehensive experiment, looking at all combinations of 7 species as both target and neighbor, as well as measurements of 6 different traits that had been proposed to relate to effect or response. Again, this experiment involved undergraduates; this time an entire class of students in the introductory ecology course I taught. The students designed small experiments involving subsets of species and Keith Landa, the graduate student course coordinator, realized that with a few additional suggestions to students, we could also get a complete competitive matrix to compare effect and response hierarchies and associated traits. As expected, these studies showed plant size was a primary determinant of per-individual competitive effect and RGRmax was a primary determinant of per-gram competitive effect. In contrast, competitive response showed the reverse trends and was uncorrelated with competitive effect. This work laid the foundation for the next step in this research theme, digging deeper into the mechanisms of resource competition, how effect and response related to resource use and then how this distinction between individual-level effect and response could help clarify the “Grime-Tilman” debate about the community consequences of competition in plants—the latter explored in a book chapter around that time (Goldberg 1990).
Also related to this set of papers is the work of my PhD student Chris Baraloto, who eschewed the old fields and wetlands of the North America Midwest and the deserts of Israel where I have done most of my field work in favor of the tropical forests of French Guiana. Chris was interested in the processes maintaining diversity and focused on various traits related to seedling establishment in different microhabitats—essentially the traits related to competitive response. Chris’s 2005 paper in Journal of Ecology in 2005 shows a clear advantage to large seeds for seedling survival and size across 8 species regardless of microhabitat, contrary to some models of coexistence based on seed size variation.
2. Community level consequences of interactions
Goldberg, D.E. and G.F. Estabrook. 1998. Separating the effects of number of individuals sampled and competition on species diversity: an experimental and analytic approach. Journal of Ecology 86:983-988.
Zamfir, M. and D.E. Goldberg. 2000. The effect of initial density on interactions between bryophytes at individual and community levels. Journal of Ecology 88:243-255.
Rajaniemi, T.K., V.J. Allison, and D.E. Goldberg. 2003. Root competition can cause a decline in diversity with increased productivity. Journal of Ecology 91:407-416.
Farrer, E.C. and D.E. Goldberg. 2011. Patterns and mechanisms of conspecific and heterospecific interactions in a dry perennial grassland. Journal of Ecology 99:265-276.
Herben, T. and D.E. Goldberg. 2014. Community assembly by limiting similarity vs. competitive hierarchies: testing the consequences of dispersion of individual traits. Journal of Ecology 102:156-166.
Another major theme in my research has been around the linkage between individual-level interactions, especially competition, and the community-level consequences of those interactions. I was motivated by two realizations about the large experimental literature on plant. First, almost all experiments on plants measured the consequences of competition for components of individual fitness (Goldberg and Barton 1992), yet the theory of competition was largely focused on coexistence and maintenance of diversity at equilibrium and therefore represented the longer-term population dynamic outcome of interactions. Second, the typical field experiment on competition quantified competitive response to diffuse competition from all neighbor species and so could not compare intra- and interspecific competition, which is fundamental for considering co-occurrence in communities (Goldberg and Barton but see Farrer’s experiments below). Thus, we had (and still have to some degree) a strong discrepancy between the questions asked by experiments on species interactions and by the theory of species interactions. In one strand of my response to this problem, I proposed two different experimental approaches to quantifying the community level effects of competition (Goldberg 1994, Goldberg et al. 1995). Three papers in Journal of Ecology use or expand those approaches. Manuela Zamfir used both approaches as part of her dissertation on competition in bryophytes at Uppsala University, and found that the community density series produced more consistent results than the combined monoculture approach. George Estabrook, a colleague at the University of Michigan, helped me deal with the problem that increasing total community density to increase competition also inevitably can increase diversity by sampling effects. And my students Tara Rajaniemi and Victoria Allison used the combined monocultures approach to separate the community-level effects of root vs shoot competition along a productivity gradient. Rajaniemi et al. (2003) showed that root competition alone could be responsible for decreasing diversity under fertilized conditions—a quite surprising result for many of us who expected that only size-asymmetric light competition could cause this result.
A fourth paper in Journal of Ecology took a different approach to understanding community level consequences. Emily Farrer, then a Ph.D. student at the University of Michigan, directly tested whether intra-specific competition was consistently greater than inter-specific competition in a series of elegantly-designed field experiment in a dry sand prairie. She found that this necessary criterion for coexistence was generally met for interactions among adults, although not for effects of adults on seedlings.
Finally, a fifth paper in this set, written with Tomas Herben of Charles University and the Czech Academy of Sciences, took an entirely different approach to understanding the community consequences of interactions that also incorporates the focus on traits related to competitive ability from the first set of papers described above. Tomas and I were both frustrated with the many papers that inferred mechanisms of community assembly from patterns of over- or under-dispersion of traits but did not often even discuss evidence for the assumed function of those traits. We used an approach I am increasingly convinced is an essential tool for understanding the community and ecosystem consequences of individual-level traits: conducting in silico experiments with highly-parameterized models of real plant communities. In this case, we used validated models of montane grassland and Michigan fens, manipulated the degree of dispersion of one trait at a time, and examined the long-term effect on coexistence and diversity. This kind of experiment is simply not possible with real plants, yet can reveal much about what traits actually do in real communities.
3. Competition along productivity gradients
Goldberg, D.E. and A. Novoplansky. 1997. On the relative importance of competition in unproductive environments. Journal of Ecology 85:409-418.
Suding, K.N. and D.E. Goldberg. 1999. Variation in the effects of vegetation and litter on recruitment across productivity gradients. Journal of Ecology 87:436-449.
Because some of the most consistent patterns in vegetation occur along productivity gradients, understanding how the role of species interactions changes along such gradients is a fundamental challenge in plant ecology. Two papers we published in the Journal of Ecology bear on this problem. Designing experimental treatments to mimic rainfall regimes along a productivity gradient from a desert to a Mediterranean climate brought home to me the importance of considering both frequency and amounts of resource supply. I had also begun to realize that survival and growth responded quite differently to neighbors, with survival more often facilitated and growth more often inhibited. (As an aside, this contrast is yet another reason why individual-level experiments are difficult to scale up to population dynamics.) Together with a postdoctoral fellow from Israel, Ariel Novoplansky, we wove these ideas together into the two-phase resource dynamics hypothesis that predicts quite different patterns for competition along productivity gradients, as well as different patterns for gradients mediated by nutrients and by water.
As a first-year Ph.D. student in my lab, Katie Suding, addressed quite a different aspect of productivity gradients that she argued had been largely overlooked in considering experimental results on competition intensity: whether or not litter was also removed along with living vegetation. She manipulated litter and/or live plants along productivity gradients in several different community types and found that either one alone had largely facilitative effects along the gradient, while the combination of the two resulted in largely negative net effects across most of the gradient.
This 1999 paper was also the beginning of a theme that isn’t well represented in any Journal of Ecology papers, but has become an important part of my thinking about plant communities: the importance of non-trophic interactions (i.e., not involving either consuming resources or being consumed as a resource). It is now well-recognized that facilitation is common in plants, especially in more stressful habitats. But is perhaps less well-appreciated that the nontrophic mechanisms that cause net positive effects are probably much more widespread—what changes among environments is not whether facilitation OR competition occurs, but the relative magnitude of positive (always non-trophic) and negative (often but not always resource competition) mechanisms and so a change in the net balance of those mechanisms. Further, the prevalence of nontrophic mechanisms of interactions involving plant-soil and plant-microclimate feedbacks also implies functional linkages between community and ecosystem ecology. Developing theory that can accommodate the diverse nontrophic mechanisms of interactions yet allows the development of generalizations about plant community and ecosystem dynamics is one of the great challenges for the next generation of ecologists.
Deborah E. Goldberg
Goldberg, D.E. and P.A. Werner. 1983. Equivalence of competitors in plant communities: a null hypothesis and an experimental approach. American Journal of Botany 70:1098-1104.
Goldberg, D.E. 1990. Components of resource competition in plant communities. Pages 27-49 in: J. Grace and D. Tilman (eds.). Perspectives in Plant Competition. Academic Press.
Goldberg, D. E. 1994. On testing the importance of competition for community structure. Ecology 75:1503-1506.
Goldberg, D.E., R. Turkington, and L. Olsvig-Whittaker. 1995. Quantifying the community-level effects of competition. Folia Geobotanica and Phytotaxonomica 30:231-242.