Research

Species clusters as a signature of competition and niche differentiation come as a revision of the traditional thinking that coexisting species should be more different than expected by chance. However, theoretical predictions typically assume complete knowledge of the map between competition and measured traits. First, most models assume that niche strategies along a single niche axis can fully predict competitive interactions, while in reality niche space is often multidimensional. Second, while measurable traits such as body size correlate with niche strategies such as diet preferences, the correlation is hardly perfect, and often noisy. Yet it is commonly assumed that measurable traits are actual niche axes, so patterns driven by competition should be visible as patterns in the distribution of these traits. These assumptions limit the plausible application of these patterns for inferring competitive interactions in nature.

Upon relaxing these restrictions in Lotka-Volterra competition models, I found that the clustering pattern is robust to contributions of unknown or unobserved niche axes. However, it may not be visible unless measured traits are close proxies for niche strategies. We conclude that patterns along single niche axes may reveal properties of interspecific competition in nature, but detecting these patterns requires natural history expertise firmly tying traits to niches. Indeed, in my study of trait-based clusters among tree species on Barro Colorado Island in Panama, only those traits previously shown to correlate most strongly with species performance turned out to show pattern.

Can we do better in our efforts to connect patterns of biodiversity with theoretical predictions by integrating traits into high-dimensional spaces?

References

R. D'Andrea, J. Guittar, J. O'Dwyer, H. Figueroa, J. Wright, R. Condit, A. Ostling (2020). “Counting niches: Abundance-by-trait patterns reveal niche partitioning in a Neotropical forest”. Ecology

R. D’Andrea, A. Ostling, J. O’Dwyer (2018). “Translucent windows: how uncertainty in competitive interactions impacts detection of community pattern”. Ecology Letters, arXiv.

The neutral theory of biodiversity assumes that phenotypic differences between species have no bearing on their interactions with each other and the environment. In a neutral world, ecological dynamics results solely from random births and deaths, as well as dispersal and speciation. While these assumptions are demonstrably false in many systems, the predictions of neutral theory for macroecological patterns of species abundance and diversity often match observations (Figure 2). This highlights the problem that macroecological patterns are consistent with different modes of community assembly -- and in fact, neutral patterns can emerge from non-neutral dynamics. Neutral theory thus poses a challenge to community ecologists interested in understanding the role that ecological processes like competition and niche differentiation play in shaping and maintaining biodiversity: it is not enough to show that the pattern is consistent with niche differentiation; one must show that it is consistent with niche differentiation to the exclusion of more parsimonious alternatives.

Trait clusters are a counterintuitive consequence of competition being mediated by species similarity, which is a critical component of niche differentiation mechanisms. Another counterintuitive effect is that extinction rates may be higher in communities undergoing niche differentiation than in completely neutral species assemblages. In communities subject to immigration from a regional pool, niche differentiation may protect some species from extinction, but promote exclusion of many more at a faster rate than predicted by neutrality (Figure 3). Indeed, a niche-differentiated community is likely to experience higher extinction rates, higher turnover, and lower local persistence than a neutral community, unless each species occupies its own ecological niche. This insight is particularly relevant to hyperdiverse communities where a combination of niche differentiation and niche sharing among species may be a more appropriate description.

Critics of neutral theory typically focus on its failure to distinguish between species. However, neutrality in nature is also broken by differences between conspecific individuals at different life stages, which in many communities may vastly exceed interspecific differences between individuals at similar stages. Mature trees may produce orders of magnitude more seeds than young trees, and have much lower mortality. Does demographic stage structure affect macroecological patterns in neutral theory? Using a two-stage neutral model where fecundity and mortality change between juveniles and adults, I found that the abundance distributions are broadly similar to unstructured neutral models, with a curious exception: significant departures from unstructured predictions occur if adults have sufficiently low fecundity and mortality. In nature, this scenario has analogues in eusocial insects, where one cast of individuals never reproduces, and in humans, where adults live long past their reproductive stage. Indeed, our preliminary tests suggest that orders with eusocial insects show signs of departure from log-series abundance distributions predicted by unstructured models and commonly observed in other insect orders. Similarly, human given names follow a power law distribution, as predicted by the structured neutral model. This finding partially rehabilitates species abundance distributions from past criticisms of their inability to distinguish between biological mechanisms.

References

R. D'Andrea, Theo Gibbs, James P. O'Dwyer (2020) Emergent neutrality in consumer-resource dynamics. PLoS Computational Biology

R. D’Andrea, A. Ostling (2017). “Biodiversity maintenance may be lower under partial niche differentiation than under neutrality” Ecology 98 (12): 3211-3218. doi: 10.1002/ecy.2020 (F1000Prime Recommended)

R. D’Andrea, J. O’Dwyer (2017). “The impact of species-neutral stage structure on macroecological patterns” Theoretical Ecology 10: 433-442. doi: 10.1007/s12080-017-0340-5

In nature, environmental conditions often vary across the landscape (Figure 4). This has significant implications for coexistence and patterns of biodiversity. Plants may be able to coexist in a forest by specializing on different local environments predominant at different sites. Dispersal limitation may further promote coexistence by minimizing competitive encounters between different species. Research in our lab corroborates these ideas. We showed that a tradeoff between high seed output and high seed survival in plants can support many more species in a patchy landscape than in a well-mixed landscape. This is because a patchy landscape promotes the formation of areas with high conspecific density, thus reducing competition across species relative to competition between members of the same species.

We are currently working on methods using the spatial distribution of trees in a tropical forest to infer niche differentiation by adaptation to different soil conditions -- the idea being that different soil types will filter for species in different niches. We are also looking at how plant-soil feedbacks may alter the correspondence between plant strategies and the soil conditions on which they will be found.

The explicit inclusion of space in ecological models raises questions about scale: How must scales of environmental variation compare with scales of species dispersal in order for this variation to promote coexistence? What kinds of trait patterns do we expect at different spatial scales? What kind of locally-acting processes between individuals will scale to the community level?

References

R. D'Andrea, J. O'Dwyer (2021). “Competition for space in a structured landscape: the effect of seed limitation on coexistence under a tolerance-fecundity tradeoff ”. Journal of Ecology