Predictability and Constraints on the Structure of Ecological Communities in the Context of Climate Change Public Deposited


Attribute NameValues
Abstract or Summary
  • Ecologists must increasingly balance the need for accurate predictions about how ecosystems will be affected by climate change, against the fact that making such predictions at the ecosystem-level may be infeasible. Although information about responses of individual species to a changing environment is increasing, scaling such information to the community level is challenging. To date, predicting responses of ecological communities to climate change is constrained by limited theoretical and empirical knowledge about the response of communities and ecosystems to change. My dissertation addresses several knowledge gaps in our understanding of community structure under climate change. This research draws from a rich experimental tradition in the species-diverse model ecosystem of the US Pacific Northwest rocky intertidal to test ecological theory. In Chapter 2, I assessed whether the response of multiple species of coralline algae to global change could be predicted from basic first principles of chemistry, physiology, and ecology. Given the rate of global change, and the time-consuming process of experimentally determining species responses to climate change, I hypothesized that species can be grouped using existing theory, either by their evolutionary relatedness or by their ecological traits, such that climate responses are similar within a group. Such a scheme would greatly reduce the number of experiments needed to characterize species climate vulnerability, requiring the characterization of the response of groups of species to climate change, rather than individual species. Using a suite of five co-occurring species of intertidal articulated coralline algae (Corallina vancouveriensis, Corallina officinalis, Bossiella plumosa, Bossiella orbiginiana, and Calliarthron tuberculosum), I applied this framework to generate ten mutually exclusive hypotheses that could explain organismal response to ocean acidification, a consequence of global climate change that threatens marine calcifying species. I found that all species had similar responses to ocean acidification, and that responses were generally predicted by the body size of the individual. Despite the power that such a framework provides in understanding group-level response to climate change, predicting community-level response requires knowledge of how organisms affect one another. In Chapter 3, I quantified species interactions in a series of removal experiments to estimate the reciprocal effects between a canopy-forming intertidal kelp (Saccharina sessilis) and a suite of understory species that persist beneath the kelp canopy. This experiment was replicated in different oceanographic conditions across a large latitudinal gradient, as a step towards understanding how interactions might change with climate change. However, the experiment demonstrated that interactions between the canopy and understory were consistent among different environmental conditions. Furthermore, the strongest effect was that of understory species, particularly articulated coralline turf algae, on the canopy species. The coralline turf algae both facilitated the recruitment of the canopy species and buffered the canopy from abiotic stress during its adult life stage. Combining experimental results and observational surveys, a hypothesized interaction network for these species was constructed, highlighting the importance of direct and indirect species interactions in promoting species coexistence. A long-standing controversy in ecology is whether or not species interactions can be inferred from observational data, as opposed to from experimental tests. Although the rocky intertidal ecosystem is unique for its ease of experimental manipulation, quantifying species interactions experimentally is often difficult or impossible. As an alternative, many have turned to statistical methods to estimate species interactions from observational data, namely, from patterns in species pairwise co-occurrences. In Chapter 4, I examined these co-occurrence methods and their potential relationship to experimentally measured species interactions. I first used a suite of different co-occurrence methods to generate a set of predicted species interactions of macrophytes and invertebrates from observational surveys conducted in the rocky intertidal zone of Oregon. I then compared the predicted species interactions to the same pairwise species interactions determined experimentally and assembled from the literature. Overall, of the seven methods tested, each generated a different set of predicted species interactions from the same data, and all methods predicted interactions that did not match those in the experimental database. Thus, predicting species interactions from patterns in occurrence remains elusive. Importantly, much work remains to be done to understand the link between species co-occurrences and their actual interactions with one another on the landscape. A key limiting frontier in climate change ecology is determining the influence of species interactions on species distributions across the landscape, and the sensitivity of such interactions to changes in climate. Finally, in Chapter 5, I used theory from the published literature and knowledge from my previous chapters to make predictions the recovery of low rocky intertidal communities after a disturbance. The process of community development after disturbance has been studied in many ways, from the successional studies of the early 1900s, to modern community assembly theory. In recent years, a focus on the unpredictability of community assembly has emerged, paying particular attention to the role of historical contingency, or priority effects, in determining the recovery trajectory of a community. Priority effects occur when the arrival of a species after a disturbance inalterably changes the composition of the developing community, driving the assembly of widely different communities at a small spatial scale. I conducted a community assembly experiment in three different low intertidal zone community "types", each characterized by different dominant macrophyte species (Saccharina sessilis, Phyllospadix spp., and algal "turfs"). Replicating this experiment at six sites along the Oregon coast, I found that both regional and local dynamics constrain the recovery of communities after disturbance. Half of the time, the community returned to the state of the nearby community type. The remaining communities were influenced by priority effects that could be predicted based on 1) regional dynamics favoring some species over others, or 2) the timing of arrival of important facilitating species. Overall, understanding the dynamic relationship between the persistence of diverse communities and a changing environment remains one of the challenges of our time. My dissertation highlights some of the challenges in predicting the future composition of communities under climate change, but also provides some ways forward. Integration of experimental, theoretical, and observational studies builds the scaffolding of prediction, whereby understanding the constraints on species physiology, the interactions among species, and community assembly can help frame the context in which predictions are made.
Resource Type
Date Available
Date Copyright
Date Issued
Degree Level
Degree Name
Degree Field
Degree Grantor
Commencement Year
Committee Member
Academic Affiliation
Non-Academic Affiliation
Rights Statement
Peer Reviewed


In Administrative Set:
Last modified: 10/28/2017

Downloadable Content

Download PDF

EndNote | Zotero | Mendeley