Spring 2009

Titles and abstracts

Wednesday February 4th at 1.30pm

Leading indicators of critical transitions in ecological systems
Ryan Chisholm
Transitions in ecological systems often occur without apparent warning, and may represent shifts between alternative persistent states.  Examples include lake eutrophication, desertification and climate change.  In many cases, ecosystem managers would like to detect such transitions before they occur in order to avert or mitigate a shift to an undesirable state.  Recent theoretical research has sought to develop and test potential leading indicators of such transitions.  I will review some recent papers on this topic and present some ongoing and published work by myself and other people in the lab group.  I will be particularly interested to get feedback from the audience on possible future research directions in this area.

Thursday February 12th at 1.30pm

Pandas, People, and Policies
Jianguo (Jack) Liu
Meeting the needs of humans and wildlife is a major challenge for society. Wolong Nature Reserve, a 200,000-ha rural area for giant panda conservation in China, is an excellent laboratory to study such a challenge. The “flagship” reserve is home to approximately 10% of wild giant pandas and almost 5,000 local residents. Furthermore, it is an open system that has been increasingly influenced by forces such as urbanization and globalization. A long-term interdisciplinary study indicates that interactions among panda habitat, people, and government policies are very complex and dynamic. Many findings from the reserve are also applicable to other coupled human and natural systems around the world.

Thursday February 19th at 2.45pm

Optimization of drug dosing regimens using PK/PD models
Patricia Geli Rolfhamre
Realistic mathematical models of the relationship between drug dose and drug effect, so-called pharmacokinetic/pharmacodynamic (PK/PD) models, have been studied extensively over the years. In this talk, I will present a stochastic model which serves two purposes.  First, it is a stochastic alternative to previously published deterministic PK/PD models used for predicting the growth dynamics of bacterial populations and second, it captures the phenomenon of postantibiotic effect. This phenomenon of continued inhibition of bacterial growth after removal of the antibiotic drug is of high relevance in the context of optimizing dosing regimens and the results will be discussed in relation to a set of experimental data.

Wednesday February 25th at 1.30pm

What limits fire in southern Africa?
Sally Archibald
Testing the relative importance of vegetation, climate, and people in driving fire regimes in southern Africa, and potential for change.

Wednesday March 4th at 1.30pm

Search Research: New insights on Lévy walks as stochastic search strategies
Fred Bartumeus
Lévy walks are a class of random walks with power-law distributed move steps that can generate beneficial statistical properties in uninformed search processes. Many scientists have started to look at these properties in animal movement trajectories without fully understanding the essential conditions and the mechanisms by which Lévy walks may enhance random encounter success. As a result of this and the difficulty of characterizing power-law distributions from empirical data, there is controversy about both the empirical evidence and the potential role of Lévy walk strategies in animal movement. Based on simulations and random walk theory I will show when and why Lévy walks are beneficial as search strategies, and I will discuss the generality of the results found in comparison with other random walk strategies. In order to progress in showing (or rejecting) any evidence of stochastic searching in a biological and evolutionary context, it is first essential that biologists and ecologists clearly understand what are the basic assumptions, the mechanistic principles, and the degree of generality of the Lévy walk hypothesis.

Wednesday March 11th at 1.30pm

Lew Ziska

Wednesday March 18th at 1.30pm

Microbial Biogeography: asking classic questions about understudied communities
Allison Shaw
Since the 18th century, scientists have studied plant and animal biogeography (patterns of biodiversity over space and time).  Of particular interest has been to what extent community composition is determined by historical events versus local environmental conditions.  Although it is now accepted that both factors are important for determining macroorganism distributions, it is still an open question for microorganisms.  Only more recently has data become available to allow us to ask similar questions about microbial communities.  The first microbial biogeography studies have focused almost exclusively on spatial patterns in terrestrial microbial communities.  I'll be presenting two projects that I've been working on with Claire Horner-Devine (at the University of Washington) for the past few years.  The first examines spatial patterns in marine microbial communities, while the second looks at temporal patterns.

Wednesday March 25th at 1.30pm

Non-structural carbon storage in determining tree species’ growth and survivorship: A modeling analysis
Anping Chen
Empirical work has highlighted the critical role of nonstructural carbon storage in determining forest tree species’ growth and survivorship. Here we evaluate the functional roles of carbon storage with a simple mathematical model. We focus our analysis on how carbon storage affect saplings’ (1) growth rate; (2) shade tolerance; and (3) survivorship under temporally heterogeneous light environment before they reach canopy.
    Model analysis shows that sapling diameter growth rate is negatively affected by carbon storage when the size of a tree is small; but this negative effect almost diminishes when a tree’s diameter is big enough. The tolerant time for a tree under constant negative net carbon gain is carbon storage (α) dependent, and the maximum length of the period of A<0 is (1+α) times for a storer than a non-storer. However, the time for a tree or tree cohort to reach canopy is independent of carbon storage. Mathematical analysis also shows that carbon storage also enhances the survival fraction under negative net carbon gain. Together, our simple mathematical work here strongly supports the fundamental contribution of carbon storage in mechanistically determining the trade-off between growth in light and survivorship in shade.

Wednesday April 1st at 1.30pm

A hierarchical Bayesian model of the evolution of the winner-loser effect in animal fighting
Adrian de Froment
I will present work in progress on a framework for understanding winner-loser effects in animal fighting using a fusion of evolutionary game theory and hierarchical Bayesian estimation. When two animals fight over a resource, the outcome is affected to by their prior fighting experience, in addition to their relative ability and motivation. Prior winning experiences increase an individual’s chance of winning subsequent contests, and prior losing experiences decrease them, all other things being equal. These “winner and loser effects” are so widespread in nature as to seem almost a universal of social life, and are important because they can account for the emergence and stabilization of social structures such as dominance hierarchies. Existing models of self-organizing social structure assume winner-loser effects into existence and build from there. No evolutionary model exists linking the parameters describing the social and resource environment, through the equilibrium strategies employed by individuals, to emergent social structure. With such a model, we could begin to ask interesting questions relevant to many species (including our own) about which conditions are most likely to lead to robust or fragile social structures. I will outline a very simple, intuitive model of how and why animals evolve strategies that integrate information gained during past fights and use that information to guide their behavior in subsequent ones.

Wednesday April 8th at 1.30pm

Evolution of collective behavior in mass migrating system
Vishwesha Guttal
A fundamental question in biology is to understand how evolution can favor collective behaviors which benefit the group as a whole (e.g., mass migrations) when the natural selection is acting at the level of individuals who act in their own interest. This question is particularly important when the individuals concerned are unrelated to one another. I will present preliminary results from our group on the evolution of collective behavior in the context of mass migration that occurs in a variety of organisms. The results are based on simulating massive number of individuals who interact in a large spatially explicit and noisy environment using a personal supercomputing system. The work is still in its initial stages and any feedback would be great.

Wednesday April 15th at 1.30pm

Why are bacterial biofilms slimy?
Carey Nadell
In both biotic and abiotic environments, bacteria often live in surface-bound, densely populated groups. One of the most widespread and visually conspicuous microbial behaviors associated with group life is the secretion of a polymer matrix (EPS) consisting mostly of polysaccharides, with traces of protein and DNA. The functions of this matrix are not very well known, but it is often assumed to be a public good that benefits all cells in the vicinity of an individual secreting it. Some theory and circumstantial evidence, on the other hand, suggests the opposite--namely, that EPS secretion is a competitive behavior. I am now performing experiments to test these ideas and will report on my early results.

Wednesday April 22nd at 1.30pm

Modeling superinfection and malaria
Eili Klein
Malaria poses a unique modeling challenge because individuals can be infected by more than one genetically distinct parasite at a time, called superinfection. While the regulation mechanisms that drive this are unclear, the result is that the within-host dynamics of the parasite significantly influence the population dynamics. However, models of superinfection in malaria have generally not acknowledged the impact that clinical infections play on the within-host dynamics, which are important when considering the impact of drug resistance. In this talk, I will be outlining the basis for a new superinfection model of antimalarial drug resistance and giving some preliminary results.

Wednesday April 29th at 1.30pm

Multi-scale analysis of collective decision-making in 2D self-propelled particle models of swarms:  An advection-diffusion with memory approach
Michael Raghib
Self-propelled particle models (SPPs) are a class of agent-based simulations that have been successfully used to explore questions related to various flavors of collective motion, including  flocking, swarming, and milling. These models typically consist of particle configurations, where each particle moves with constant speed, but changes its orientation in response to  local averages of the positions and orientations of its neighbors found within some interaction region.   These local averages are based on 'social interactions', which include avoidance of collisions, attraction, and polarization, that are designed  to generate configurations  that move as a single object. Errors made by the individuals in the estimates of the state of the local configuration are modeled as a random rotation of the updated orientation resulting from the social rules. More recently, SPPs  have been introduced in the context of collective decision-making, where the main innovation consists of dividing the population into naïve and `informed' individuals. Whereas naïve individuals follow the classical collective motion rules, members of the informed sub-population update their orientations according to  a weighted average of the social rules and a fixed 'preferred' direction, shared by all the informed individuals. Collective decision-making is then understood in terms of the ability of the informed sub-population to steer the whole group along the preferred direction. Summary statistics of collective decision-making are defined in terms of the  stochastic properties of the random walk followed by the centroid of the configuration as the particles move about, in particular the scaling behavior of the mean squared displacement (msd).  For the region of parameters where the group remains coherent , we note that there are two characteristic time scales,  first  there is  an anomalous transient shared by both purely naïve and informed  configurations, i.e. the scaling exponent lies between 1 and 2. The long-time behavior of the msd of the centroid walk scales linearly with time for naïve groups (diffusion), but shows a sharp transition to quadratic scaling (advection) for informed ones. These observations suggest that the mesoscopic variables of interest are the magnitude of the drift, the diffusion coefficient and the time-scales at which the anomalous and the asymptotic behavior respectively dominate transport, the latter being linked to the time scale at which the group reaches a decision.  In order to estimate these summary statistics from the msd, we assumed that the configuration centroid follows an uncoupled Continuous Time Random Walk (CTRW) with smooth jump and waiting time pdf's.  The  mesoscopic transport equation for this type of random walk corresponds to an Advection-Diffusion Equation with Memory (ADEM).   The introduction of the memory, and thus non-Markovian effects, is necessary in order to correctly account for the two characteristic  time scales present.  Although we were not able to calculate the memory directly from the individual-level rules, we show that it can be estimated  from a single, relatively short, simulation run using a Gamma density and  a Mittag-Leffler function as templates.  With these functions it is possible to predict accurately the behavior of the msd, as well as the full pdf for the position of the centroid. The resulting ADEM is self-consistent in the sense that transport parameters estimated from the memory via a Kubo relationship coincide with those estimated from the moments of the jump size pdf of the associated CTRW for a large number of group sizes, proportions of informed individuals, and degrees of bias along the preferred direction.  We also discuss the phase diagrams for the transport coefficients estimated from this method. We also note that the time scale to collective decision is invariant  with respect to group size, and depends only on the proportion of informed individuals and the strength of the coupling along the informed direction.

Wednesday May 6th at 1.30pm

Movement and stochastic foraging strategies of C. elegans
Liliana Salvador
A common model to describe foraging movements of organisms is the random walk, in which the duration and direction of the forward movement of the organism is chosen randomly. In the absence of any knowledge of the environment, an important question is: What statistical strategy will be the most efficient to find food? Or more specifically: Which is the distribution from where the organism should choose the duration of its forward movements? Recent studies show that the notion of intermittent locomotion is important to link animal behavior to large statistical properties of movement. In this talk, I will be talking about the intermittent behavior of C. elegans giving some preliminary results on the distribution of its movements.

Wednesday May 13th at 1.30pm

Chasing the cockroach:  How reflexes affect running
Joshua Proctor
In the hopes of understanding running and walking we look to a wonderful and beautiful species: the cockroach!  The insect may not have the aesthetics of a cheetah or Olympic sprinter, but cockroaches are incredibly adept and efficient runners that have the skill to elude boots or insect friendly traps.  In my lab tea, I would like to briefly describe how we, very generally, view cockroach locomotion. This broad summary will attempt to describe how very diverse systems, such as groups of neurons and the exoskeleton, can interact to generate movement.  In the context of this global story, I will focus in greater detail on the effect of reflexes in running. Reflexes attempt to feed back information from one system, the body-limb mechanics, to the neuronal system in order to modulate the commands sent back to the muscles and limbs.  Lastly, I will present the progress we have made in understanding how simple mathematical models of neurons are affected by feedback and conclude with results from a more sophisticated neuro-mechanical model where feedback increases the stability during running.

Wednesday May 20th at 1.30pm

Fairness and cooperation in the commons
Maja Schlueter
Sustainable management of the commons is contingent upon cooperation among its users restraining themselves from individual short-sighted resource overexploitation. Maintaining such cooperation against the myopic self interest of individual users and despite growing environmental pressure is a challenging task. However, numerous experiments and field results have shown that shared social norms can promote cooperation under certain conditions. We want to investigate the role of other regarding preferences, such as a fairness norm, on the establishment and maintenance of cooperation in a common pool resource subject to social and ecological change. The study is informed by recent institutional developments in water management in a semi-arid river basin. We interface an evolutionary game theoretic model with an agent-based approach to identify conditions under which concerns for equity leading to ostracism towards norm violators can promote the evolution of cooperation. Initial results show that coexistence between co-operators and defectors arises when the income difference between co-operators and defectors becomes large. On the other hand, cooperation is more difficult to establish when interactions are such that defectors can occasionally get away with free riding. We are still in the midst of developing the models and welcome feedback on the conceptual approach and potential simulation experiments.
 

Links to previous schedules

    Fall 2002     Spring 2003
    Fall 2003     Spring 2004
    Fall 2004     Spring 2005
    Fall 2005     Spring 2006     Summer 2006
    Fall 2006     Spring 2007
    Fall 2007     Spring 2008
    Fall 2008