Screen Shot 2019-08-26 at 2.07.15 PM

Fourth Annual Traits Workshop

Reporting by Helen Hill for the MIT Darwin Project

The fourth Workshop on Trait-Based Approaches to Ocean Life, held August 18-21, 2019 at Chicheley Hall in Buckinghamshire in the UK was a wonderful opportunity for Darwin Group members to catch up with former colleagues while sharing current directions in marine ecology viewed through a traits lens.

This year’s meeting was structured around four themes: Traits, environments, ecology and evolution; Food-webs and Trophic Interactions; Multifarious lifestyles of marine microbes;  and Traits, Networks and Ecosystem Function.

Current Darwin Group members Stephanie Dutkiewicz, Chris Follet, Chris Hill, and Deepa Rao, as well as group alumni B. B. Cael (UH), Fanny Monteiro (Univ. of Bristol, UK), David Talmy (Univ. of Tennessee, USA), Ben Ward (Univ. of Southampton, UK), and Emily Zakem (USC) were among the approximately 80 scientists who shared their work at this year’s meeting.

Darwin Group Abstracts

B. B. Cael* 
Co-Authors: Nick Hawco, Siobhan Braybrook

A tale of two cobalts // Quantifying leaf cell shape (Poster)

Cobalt (Co) is an important micronutrient in the ocean whose geochemistry is relatively poorly characterized. In particular, the global turnover of Co is not well-constrained as different processes have been argued to control this turnover that have also been argued to operate on a wide range of timescales. Using corresponding radiocarbon measurements, we assign ages to a database of Co and labile Co (LCo) concentration measurements. We then use a regularized inverse Laplace transform to tease apart the multiple timescales of Co turnover. Surprisingly, Co turnover is robustly best characterized by a relatively sharp, bimodal distribution, with ~1/3 of Co turning over on a ~250a timescale and ~2/3 turning over on a ~1500a timescale. LCo, which is independently measured and constitutes ~1/3 of total Co, is robustly characterized by a relatively sharp, unimodel diststribution turning over on a similar multi-centennial timescale. LCo is chemically similar to the Co2+ ion and is susceptible to scavenging by Mn-oxidizing bacteria, whereas other Co is strongly bound by organic molecules and inert. Thus, somewhat surprisingly, it appears that the deep ocean total Co inventory is well-described as comprising two distinct pools, turning over on two distinct timescales, potentially by distinct mechanisms (e.g. bacterial scavenging & ventilation). // Leaves’ epidermal cells display a diversity of shapes, from regular and rectangular to complex and jigsaw-like. Understanding of the function(s) as well as the phylogenetic or environmental patterns of cell shape arguably remains elusive because of the unsatisfying shape metrics used to date. Drawing inspiration from potamology (river science), we argue that the natural shape metric is mean total absolute curvature. This metric is surprisingly weakly correlated with the more commonly used solidity metric for cells from hundreds of different plant species. We describe additional insight available from a curvature-based approach and discuss next steps.

Stephanie Dutkiewicz
Co-authors: Phil Boyd and Ulf Riebesell

Biogeochemical and ecological redundancy in phytoplankton communities (Talk)

Global change-driven stressors such as warming, ocean acidification and alterations to resource supply are likely to have differing effects on the diverse set of phytoplankton species. Differing responses are likely to lead to the extinction of some species. Here we explore how losses of species, and their specific traits, might impact the marine biogeochemistry and ecology. We use a global three-dimensional model that includes functional and size traits to capture complex regional diverse planktonic communities. The communities encompass size classes from .6 to over 2um, and several functional groupings (e.g. diatoms, diazotrophs) and trophic strategies. We run a series of experiments where we remove a single size class, functional group or combination of phytoplankton types to explore redundancy in phytoplankton communities. We find that removing nitrogen-fixing phytoplankton has the largest impact on biogeochemical properties such as global primary production. However, interestingly, we find that removing size classes and functional groups (other than diazotrophs) has little impact on globally integrated primary production, but many of the experiments do show large regional shifts in biogeochemistry. We also find non-intuitive ecological changes, with shifts in the rest of the communities and large impacts on higher trophic levels in almost all experiments. Our results suggest a lack of ecological redundancy, though potentially more biogeochemical redundancy, but only when placed in a global context.

Christopher Follett 
Co-authors: Matthew J. Church, Stephanie Dutkiewicz, and Michael J. Follows

Geometric Niche Partitioning of Nitrogen-Fixers in the Sea (Talk)

Nitrogen-fixing organisms play a key role in alleviating nitrogen limitation and stimulating new production in oligotrophic regions of the global ocean. Shifts in the elemental ratio of resource supply help explain the biogeography of nitrogen fixation, but the factors which control the functional class of nitrogen fixer (Diatom-Diazotroph Associations, Trichodesmium, UCYN-A, Crocosphaera) remain poorly understood. Focusing on the generation, diffusion, and reduction of oxygen, we use the trait of cell size to derive a cap on the growth efficiency of multi-cellular, nitrogen-fixing, consortia. We find that under certain conditions, this maximum efficiency depends entirely on the relative sizes and locations of fixing and non-fixing cells. This geometric theory correctly predicts the temporal and spatial niche partitioning between UCYN-A and Crocosphaera as demonstrated by 1 year of monthly nifH gene abundance data at Station ALOHA, an oligotrophic time-series site in the oligotrophic Pacific Ocean. We further explain how this geometric efficiency factor could be straightforwardly encoded in computational, trait-based, models. When combined with previous work, this theory provides a blueprint for understanding the unique size structuring of nitrogen-fixing consortia and their spatial-temporal structuring in the oligotrophic sea.

Chris Hill 
Co-authors: Oliver Jahn, Steph Dutkiewicz, Zhen Wu, Greg Britten, Ali Ramadhan

Sub-mesoscale induced variability in trait-based ecosystem models (Poster)

This talk will look at a suite of emerging activities aimed at examining how state of the art trait-based models based on the MIT Darwin ecosystem code and algorithms respond to physical ocean model variability at spatial scales down to below 1 kilometer and at temporary scales that capture the diel cycle. The talk will describe some initial results from a family of models and will look at ways we are trying to synthesize the rather complex systems into big-picture insights. It will discuss current dilemmas on how to appropriately interpret classical ecosystem statistical methods such as Bray-Curtis diversity and Shannon index in the context of a moving, Lagrangian frame of reference with potentially vigorous physical mixing present.

Fanny Monteiro* 
Co-authors: Jamie Wilson, Joost de Vries, Glen Wheeler

Investigating the effect of size and life cycle on coccolithophore ecology and ocean carbon cycling in a trait-based global ocean model (Poster)

Coccolithophores are a key group of marine phytoplankton contributing to the global oceanic production of calcium carbonate via the production of coccoliths. While most studies rely on the small calcifying Emiliania huxleyi, coccolithophores have a large diversity in size, shape and ecology with about 2 species present in the modern ocean and a bi-stage life cycle. For instance, within the dominant placolith-bearing species (e.g., E. huxleyi, Calcidiscus leptoporus, Coccolithus pelagicus), cell diameter ranges from 4 to 2 µm and PIC:POC ratio from .2 to 3. Coccolithophores have also 2 main life stages, haploid and diploid phases, in which they change cell size and coccolith properties. This highlights the need to include different traits and trade-offs in coccolithophore studies. Here we use a trait-based model of the global ocean to explore the effect of size and life cycle on the distribution, diversity and both primary and CaCO3 production of the placolith-bearing coccolithophores. The model combines the trait-based Darwin ecosystem model which accounts for a diverse and adapted population of plankton, with the Earth System model of intermediate complexity cGENIE, which is capable of running large ensembles. We present here on-going results on how to model size diversity and life cycle of coccolithophores, as well as show sensitivity experiments on the effect of these key traits on coccolithophore ecology and ocean carbon cycling.

Deepa Rao 
Co-authors: Stephanie Dutkiewicz, Michael J. Follows

Modeling morphotypes: A trait-based approach to modeling transitions between single cells and colonies of the Phaeocystis genus (Poster)

Deepa Rao presenting her poster

Deepa Rao presenting her poster

The Phaeocystis genus is a cosmopolitan phytoplankton species that has massive, recurrent seasonal blooms in high-latitude coastal environments. Although Phaeocystis plays a fundamental role in carbon and sulfur biogeochemistry, it is often not well represented in marine microbial ecosystem models, partially due to limited observational data. Phaeocystis is a haptophyte with a polymorphic life cycle, that alternates between free-living, flagellated unicell stage (3-8 µm diameter) and a colonial stage consisting of hundreds to thousands of cells living in a mucous matrix (1s µm – mm in diameter). In its colonial stage, Phaeocystis forms near monospecific blooms, indicating that it has a distinct fitness advantage. In essence, Phaeocystis can be considered as a population with a single genotype with multiple phenotypes. In this study, we explore the trait-trade-offs of multiple morphotypes. First, we use proteomics data to develop mechanistic insights into the differences between unicell vs colony stage traits expressed in culture treatments based on the KEGG orthology database of molecular functions. Second, we incorporate this Phaeocystis model in a 1D water column model to examine competition dynamics and the seasonal transitions between single vs colonial cells. We focus on examining colony-specific traits that can infer a fitness advantage and observation-based size scaling for the number of cells per colony. These advantages include a reduction in top-down control, changes in how iron is acquired and increased (size-dependent) internal quota of larger colonial cells. Initial results show that having multiple morphotypes can expand the theoretical and realized niche of model Phaeocystis, potentially enabling it to persist until conditions are favorable.

David Talmy* 
Co-author: Selina Vage

Trade-offs modify ecosystem biomass structure along trophic gradients (Talk)

In a broad range of terrestrial and aquatic environments, predator biomass scales with prey biomass following power laws with exponents less than one. Prey production and prey biomass has also been shown to scale sub-linearly in diverse systems. We present a general model to interpret widespread patterns of predator-prey and production-biomass scaling. The model allows competition along trophic gradients among prey with traits governed by a trade-off between resource acquisition, and defense against predation. We show that low resource supply selects prey that persist at low biomass density but take up resources rapidly and are subject to intense top-down control by predators. High resource supply selects prey that persist at high biomass density, but with low affinity for resource acquisition, and high ability to resist predators. In the plankton, cell size may be associated with a trade-off between resource acquisition and defense against predators, and there are clear increases in average cell size along trophic gradients. In fish and terrestrial ecosystems, defensive behaviors may be less tightly linked with organism size, and instead may be driven by allocations of time, energy and resources toward predator deterrence and avoidance. Our theory explains density-dependent variation in behavioral, physiological, and morphological traits observed in diverse systems.

Ioannis Tsakalakis (MBL, CBIOMES postdoc)
Co-authors: Stephanie Dutkiewicz, Mick Follows, Joseph Vallino

Patterns of phytoplankton diversity driven by resource fluctuations in a global ocean model (Poster)

Resource fluctuations is a main driver of temporal niches among phytoplankton taxonomic or functional groups, affecting species composition and diversity. In this study, we focus on the two major periodic processes of resource fluctuations in the ocean, the seasonal and the diel cycle, and study their importance for understanding large-scale patterns of phytoplankton biogeography and diversity. We consider three phytoplankton functional groups of differing size that represent distinct adaptations to resource fluctuations: small-size gleaners (high affinity for nutrient uptake), medium-size opportunists (high maximal growth rate) and large-size hoarders (high capacity in nutrient storage). We will present preliminary results on the biogeography of the three functional groups in the global ocean, with a special focus on the latitudinal and productivity gradients. Our analysis includes idealized modeling on the effects of resource fluctuations and analysis using the MIT General Circulation Model. In addition, we use a thermodynamic-based analysis to estimate energy flow through the phytoplankton community and the abiotic environment, focusing on the role of trait adaptations to fluctuating resources.

Ben Ward* 
Co-authors: S. Collins, C.R. Young, B. Sauterey

Functional and taxonomic diversity in a global ocean metacommunity model (Poster)

Recent global surveys have unveiled enormous levels of diversity in marine microbial communities, with molecular analysis suggesting the existence of up 15, genera of marine eukaryotes in the photic layer of the ocean alone. Analysis of this data suggests a division between taxonomic diversity (identified by neutral genetic markers) and functional diversity (identified by association with genes of known function). We present results from am new ‘matrix metacommunity model’ that allows the emergence of both functional and neutral diversity. Environmental dispersal, ecological selection and the adaptive generation of new phenotypes are manipulated within the model, allowing assessment of their influence on community assembly, in terms of both functional and taxonomic diversity.

Emily Zakem* 
Co-authors: Martin Polz, Mick Follows

Incorporating metabolic diversity into trait-based modeling frameworks with metabolic functional types (Talk)

Microbial activity mediates the global flow of carbon, oxygen, nitrogen, and other elements, including climatically significant gases. However, non-photosynthetic microbial activity is typically not resolved dynamically or mechanistically in global models of the marine and terrestrial biospheres, inhibiting predictive capability. Understanding the global-scale impact of complex microbial community activity requires a consistent framework with which to constrain the parameterizations of diverse metabolisms. Here, we describe how the key redox chemistry underlying specific metabolisms can be exploited to parameterize diverse metabolic strategies. By quantitatively relating metabolic yields to chemical gradients, the growth and respiration of microbial biomass is systematically related to stoichiometries of substrate consumption, oxidation, and reduction that constitute biogeochemical fluxes. Linked with parameterizations of resource acquisition rates, whole organism metabolism can be integrated into trait-based modeling frameworks as metabolic functional types. Benefits of this approach include prognostic metabolic biogeography and ‘gene-fluent’ predictions of community metabolism. The theoretically grounded, electron-balanced framework progresses the description of microbial ecosystems towards conservation of energy as well as mass.

Andrew Barton* (Scripps Institution of Oceanography), Stephanie Dutkiewicz, Fanny Monteiro*, and Ben Ward* also served on the meeting steering committee.

The meeting was generously sponsored by Euromarine, Ocean Carbon and Biogeochemistry,  and the Gordon and Betty Moore and Simons Foundations.


* Denotes former Darwin Group Member.


Meeting Website
PDF of Meeting Program

Story image:
(Back row, left to right): Chris Hill, B.B. Cael, Ben Ward, Ioannis Tsakalakis, Andrew Barton, Øyvind Fiksen (Univ. of Bergen – Øyvind was a 2011 visiter to the Darwin Group), Friederike Prowe (GEOMAR, Helmholtz Centre for Ocean Research Kiel and former visiting postoc), (Middle row, left to right) David Talmy, Deepa Rao, Emily Zakem, (Front row, left to right) Fanny Monteiro, Chris Follett, Stephanie Dutkiewicz – image courtesy: D. Rao.