Wednesday February 13th at 12:30 pm ***IN GUYOT 100***

From Black-Scholes to Bacterial Black-Boxes: Using human-associated bacterial communities and null models of population dynamics and to explore the links between ecology and economics
Alex Washburne
If you choose a species this year at random from a list of species in an area (say, a fish species in a fishery), which probabilistic model will give you the best estimate of the species' abundance next year and the correct uncertainty of that estimate? I will motivate this problem with an empirical anomaly that arose while analyzing a time-series of bacterial relative-abundances on the human body. I will illustrate a model for stochastic population dynamics of independent species and show that one can unfold the relative-abundance data - which we have a lot of, and in which we observe many species over long periods of time - to get tantalizing, indirect glimpses of a minimally-informed (no functional ecology, no taxonomic groupings, etc.) picture of how species abundances change with time that arises directly from real data and is derived from an underlying, mechanistic model. Currently (though with a very high degree of uncertainty) the population dynamics over just one timestep appear to follow the Black-Scholes model (specifically, a geometric brownian motion, GBM, with log-normally distributed fold-changes between timesteps), but some body sites exhibit occasional, extreme events that occur more frequently than predicted under GBM, and the dynamics over several timesteps exhibit strong, negative auto-correlation uncharacteristic of GBMs that could be caused by (1) mean reversion, (2) artifacts of the data, or (3) something else entirely that I haven't thought of. As indicated by the inclusion of (3), this work is far from complete and your informed feedback will be greatly appreciated. Bring your crazy-speculative-hats for full enjoyment of this talk!

back to schedule

Wednesday February 20th at 12:30 pm ***IN GUYOT 100***

Network structure and information transfer in fish schools
Brin Rosenthal
Obligate schooling fish provide a convenient example of a highly dynamic communication network, one that results in high levels of global structure from leaderless local interactions. Improved technology allows for precision automated tracking of trajectories in fish schools, as well as estimation of the sensory information available to each individual in the group. With these tools we build up a data-driven sensory interaction network used for communication of alarm information. We use this network to investigate how specific network structures lead to observed group-scale phenomena. More specifically, we study how group organization is related to rapid information transfer, and filtering of false positive alarms.

back to schedule

Wednesday February 27th at 12:30 pm ***IN GUYOT 100***

The metabolic ecology of social insect colonies
James Waters

The consistent and nonlinear relationship between animal metabolic rates and their body size is among the most widespread patterns in biology and the mechanistic basis for this relationship, if one exists, is highly controversial. One of the challenges in empirically testing models in metabolic ecology is that the necessary manipulations are often destructive. Social insect colonies are dynamic model systems to investigate the scaling of resource transport network structure and function. Transport networks provide the pathway for the acquisition and assimilation of nutrients from an organism's environment but an organism's development and behavior are central to setting these rates of these processes. Social insect colonies integrate the balance of supply and demand across levels of biological organization such that the individual components are simultaneously serving as the supply transport network and also the source of energetic demand. Whole colonies of seed harvester ants were found to exhibit a robust hypometric allometry in which mass-specific metabolic rates decreased with increasing colony size similar to the pattern for individual physically integrated organisms. The scaling of worker activity, interaction patterns, and nest geometry with colony size provides insight into the mechanisms by which metabolic regulation emerges in functionally integrated collective animal groups.

back to schedule

Wednesday March 6th at 12:30 pm

How cod shapes its world
Anieke van Leeuwen

Cod populations in the Northwest Atlantic and Baltic Sea have collapsed in the early 1990s. The decline of this dominant predator led to ecosystem wide effects and a trophic cascade detectable down to phytoplankton biomass. The subsequent lack of cod population recovery (up to date) is classically explained by hypotheses which invoke environmental factors, such as climate change. In this paper we explore the potential for mechanisms intrinsic to the community as drivers of stabilization of an ecosystem state with low top-predator biomass. By the analysis of a stage-structured biomass model we show that the size-dependent predator-prey interactions in this system can account for the occurrence of alternative stable states in which cod is either present or absent. In the coexistence state with its prey, cod exerts a strong predation pressure on the prey population, which affects the prey size-distribution and results in larger densities of the preferred prey sizes for cod than available in absence of predation. Cod thus shapes its food environment to its own benefit. Furthermore, cod biomass and yield tend to increase in response to increased exploitation, unless a stock collapse is imminent. These results are consequences of the indirect effects of predation and harvesting, whereby increased mortality relaxes competition among surviving individuals, leading to an increase in food intake and hence increased somatic growth and reproduction.

back to schedule

Wednesday March 13th at 12:30 pm

"Incorporating microbial dynamics into the LM3 soil carbon model"
Ben Sulman
[Abstract TBA]

back to schedule

Wednesday March 27th at 12:30 pm

Solving timescale problems in biodiversity models
Ryan Chisholm

The neutral theory of biodiversity accurately predicts many static patterns of biodiversity but fails spectacularly in its predictions of dynamic patterns. In tropical forest tree communities, two of these dynamic patterns are decadal-scale abundance fluctuations and geological-timescale species ages. I will present two models that aim to solve these problems while preserving the accurate static predictions of neutral theory.

back to schedule

***FRIDAY APRIL 5TH*** at 12:30 pm

Do regeneration strategies control the emergence and maintenance of symbiotic nitrogen fixation in different types of forests?
Efrat Shefer

Symbiotic N2-fixation by plants is the main source of nitrogen input to most natural ecosystems. Yet, there is no good explanation to why N2-fixers are favored in ecosystems where they ultimately act to remove soil N constraints on plant growth (mainly tropical forests), whereas, in other ecosystems N2-fixers are either rare and/or quickly excluded by competition with others even though limitation persists (mainly temperate or boreal forests). My study examines whether variation in life-history traits determines the emergence and maintenance of symbiotic N2-fixers in different types of woody plant communities globally, and how the life-histories of fixers are adapted to different disturbance regimes. I developed a model of a community in which a symbiotic N2-fixing and a non-fixing plant species interact and look for the factors that control community organization and the emergence of individual strategies. I will present a spatially implicit version of this model and under which conditions it predicts that N2-fixers will dominate, coexist with non-fixers or be outcompeted by them. Finally, I will suggest a spatially implicit version of the same model and questions I would like to study with it.

back to schedule

***SPECIAL MEETING IN GUYOT 10*** Monday April 8th at 2:00 pm

Mutation rules and the evolution of sparseness and modularity in biological systems
Tamar Friedlander

Biological systems show two structural features on many levels of organization: sparseness, in which only a small fraction of possible interactions between components actually occur; and modularity, the near decomposability of the system into modules with distinct functionality. Recent work suggests that modularity can evolve in a variety of circumstances, including goals that vary in time such that they share the same subgoals (modularly varying goals). Here, we studied the origin of modularity and sparseness focusing on the nature of the mutation process, rather than variations in the goal. We use simulations of evolution with different mutation rules. We find that commonly used sum-rule mutations, in which interactions are mutated by adding random numbers, do not lead to modularity or sparseness except for special situations. In contrast, product-rule mutations in which interactions are mutated by multiplying by random numbers, a better model for the effects of biological mutations, lead to sparseness naturally. When the goals of evolution are modular, in the sense that specific groups of inputs affect specific groups of outputs, product-rule mutations lead to modular structure; sum-rule mutations do not. Product-rule mutations generate sparseness and modularity because they keep small interaction terms small.

back to schedule

Wednesday April 17th at 12:30 pm

Adaptations of fish and fisheries to rapid climate velocities
Malin Pinsky

Climate change presents a profound challenge to the sustainability of coastal systems, but most research has ignored the important coupling between human responses to climate and the cumulative impacts of these responses on ecosystems. Fisheries are a prime example of this feedback: climate drives shifts in species distributions and fisheries adapt to these shifts. However, changes in the location and intensity of fishing are their own major ecosystem impact. This talk will present some initial results, but will focus primarily on new and still developing ideas about how we could model the coupled natural & human dynamics of this system and begin to understand the conditions that would allow fishing communities and the ecosystems on which they rely to be sustained in the face of climate change. Audience feedback and critical thinking is highly encouraged.

back to schedule

***MONDAY APRIL 22ND*** at 2:30 pm

Role of institutions in climate change agreements
Jorge Pacheco

The effects of climate change may be framed as a public goods dilemma, in which the overall perception of risk of future losses plays an important role. Up to now, attempts to reach global cooperation have been unsuccessful. One of the reasons pointed out is the present lack of sanctioning mechanisms to be imposed on those who do not contribute to the welfare of the planet and to the inexistence of a sanctioning institution that supervises abidance to agreements. I will discuss the role of different types of sanctioning in deterring non-cooperative behavior in climate agreements, employing a dynamical evolutionary game theoretical framework . When the perception of risk is small, a polycentric approach in which parties create local institutions that punish free-riders leads to a significant increase in global cooperation. On the contrary, a global institution provides, at best, marginal improvements regarding overall cooperation.

back to schedule

Wednesday May 1st at 12:30 pm

Tropical forest size structure: a dynamic equilibrium
Caroline Farrior

Tropical forests are beautiful places with amazing amounts of diversity. Despite this complexity, these forests are remarkably consistent in the size distribution of trees. Across the globe, tropical forests follow a tight power-law distribution with very similar exponents. Such a consistent and clear pattern warrants an explanation, and that explanation is likely to provide an understanding of basic governing processes of these forests. I will investigate the generation and stability of this pattern with 30 years of data from 50 hectares of a tropical forest in Panama. With insights from the data, we build a stochastic model capable of regenerating the power law. Through analysis we find that this power law is generated through a self-thinning process and depends on only the allometric scaling of tree crown area with diameter, one likely physically constrained parameter. This finding explains the consistency of the relationship across the globe and has the potential to change the way we think about tropical forest dynamics.

back to schedule

***SPECIAL MEETING*** Monday May 13th at 12:30 pm

From gene interactions to the network of overlapping complexes and novel functional modules with CFinder
Illes Farkas
(Hungarian Academy of Sciences and Eotvos University, Budapest, Hungary)

Motivation: Cellular tasks are usually performed by groups of functionally associated proteins, referred to as modules (or complexes). In a network of gene/protein interactions modules appear as densely internally connected groups of nodes, also called communities or clusters. These clusters frequently overlap with each other and form a network on their own in which nodes (links) represent the modules (overlaps). Note also that actual protein (gene) interactions often clearly not binary. Instead, densely internally linked groups of proteins are often more biologically meaningful.

Results: CFinder locates and visualizes overlapping, internally densely linked groups of nodes in undirected, directed and weighted networks and it allows the user to easily navigate between the original network and the network of clusters. In gene (protein) association networks CFinder can be used to predict the function(s) of a single protein and to discover novel functional modules. As an example, we suggest in the yeast PPI (protein-protein association) network two groups of proteins that are likely to perform more specific functions within the broader function of "the establishment of cell polarity".

Website: http://CFinder.org (Network data sets, Free software for Win/Mac/Linux, Documentation, FAQ).

References:
1. Palla G, Derenyi I, Farkas I, Vicsek T. Uncovering the overlapping community structure of complex networks in nature and society. Nature 435, 814 (2005).
2. Adamcsek B, Palla G, Farkas I J, Derenyi I, Vicsek T. CFinder: Locating cliques and overlapping modules in biological networks. Bioinformatics 22, 1021 (2006).
3. Farkas I J, Ábel D, Palla G, Vicsek T. Weighted network modules. New J. Phys. 9, 180 (2007).

back to schedule

***THURSDAY*** May 23rd ***IN GUYOT 100*** at 12:30 pm

The mystery of the Carolina Parakeet: Reconstructing an extinction in retrospect
Colin Carlson

Species distribution modeling has the ability to show changes in a species' distribution during a range contraction, yet it has been widely ignored as a means of examining the decline of a species during an extinction. Here, we use distribution models to detect the temporally and spatially independent extinctions of the two subspecies of the Carolina Parakeet (Conuropsis carolinensis), whose ecologically distinct identity is confirmed by our models; we subsequently revise the estimated extinction dates for these subspecies, with implications for the credibility of the latest sightings of the species in the early 20th Century. Re-analyzing the history of species declines may reveal new conservation-relevant information; in our case, we highlight the value of refugia in the survival of C. c. carolinensis in the face of habitat loss, which demonstrates the value of refugia as a means of stopping the decline of other endangered or threatened species.

back to schedule