Theoretical Ecology lab teas are informal talks to the lab group(s) of Simon Levin, Steve Pacala, Andy Dobson, and other interested folks around Princeton or visiting.  The speakers come from that same set.  The talks are more or less strictly limited to 30 minutes including the usual questions and interruptions.  Of course, lively discussion often continues beyond 30 minutes -- after the talk is concluded.

Talks are scheduled by Kerstin Wiegand and Urmila Malvadkar. The lab tea email list is maintained by Urmila Malvadkar.  Email malvadkr@princeton.edu to have your name added to this list so you too can receive reminders about upcoming lab teas. Click here for Spring 2000 schedule and summaries.
 

All talks take place at 2:00 pm in Eno 209.
 

Tuesday, September 19	Sonia Altizer
Tuesday, September 26	Ben Strauss
Tuesday, October 3	Jerome Chave
Tuesday, October 10	Eduardo Zea
Tuesday, October 17	Stuart Sandin
Tuesday, October 24	Luis Borda de Agua
Tuesday, November 7 - special seminar	Rick Condit
Tuesday, November 14	Kerstin Wiegand
Tuesday, November 21	Andy Dobson
Tuesday, November 28	Fred Guichard
Tuesday, December 5	Eirikur Palsson
Tuesday, December 12	Tim Brown
Tuesday, January 9	AVAILABLE SLOT
Tuesday, January 16	AVAILABLE SLOT
Tuesday, January 23	Jerome Chave

 

Titles and abstracts most recent first (posted approximately one week before the talk):
 

Tim Brown

The New World army ant, E. burchelli, is a diurnal swarm raiding species that forages in huge raids of up to 200,000 individuals.  Although decades of study have revealed much of the behavioral biology and ecology of army ants, little is known of the fundamental behavioral mechanisms underlying the exquisite coordination and organization achieved by these ants in their raiding and nomadic behavior.  For anyone with an interest in self-organization and complex systems, watching 100,000 ants foraging effectively together without centralized control is enough to make one move to Costa Rica permanently (wait until you see the pictures of thousands of army ants consuming a huge cockroach alive).

Although there has been extensive work on E. burchelli at BCI (Panama) and Corcovado (Costa Rica), very little research on these ants has taken place at La Selva (Costa Rica).  I'll present an overview of past field and modeling work on E. burchelli.  Then I'll discuss the results of my field work this summer on E. burchelli at La Selva, its relation to my current
modeling work on harvester ants and future questions that remain to be answered (and we'll eat delicious junk food and look at cool video and images of army ants swarming, all in 30 minutes or less).
[back to schedule]

Eirikur Palsson

[back to schedule]
A mathematical model for cell movement in multicellular systems has been developed that allows us to simulate and visualize, in three dimensions, individual cell movements in a number of multicellular
systems. These include cell movement during, aggregation and slug stage, of {\it Dictyostelium discoideum}, embryogenesis, limb formation and wound healing.  The building blocks of the model are
individual cells, and each cell has certain given properties. The basic properties are that a cell deforms under force (either stretch or compress), while conserving its volume, it adheres to other cells and it can generate an active motive force. The response of a cell depends on its internal parameter state, and on the information it receives from its external environment, which includes neighbor cells, the extracellular matrix and chemical signals.  The movement and deformation of each cell is then determined by summing up all the forces that a cell experiences from its surroundings, and using that force in the equations of motion.
 

Here I introduce this model and and show examples of its applications and compare the results with experimental data. Among the simulations I show, is how different cell types can sort out based solely on
differences in adhesion. The results are compared to cell sorting experiments done by Steinberg {\it et al}~\cite{Steinberg63,Foty96} using values for adhesion within the range of the experimental values.
I also present results from simulations of {\it Dictyostelium} movements.  First I show simulations of the aggregation stage, where cells are aggregating chemotactically, towards a signaling center, in response to cAMP waves. In these simulations one can observe stream formation and how the mound arises due to the inward motion of the cells towards the signaling center. I will also present simulations of 2-D slugs, where where I studied the affect cell adhesion and cell chemotaxis have on the soring of Prespore and Prestalk cells in the slug. These findings will be compared to observations of 2-D slugs done by Bonner~\cite{bonner98}.
 

\end{document}

{\bf References}

[1] M. S. Steinberg, Reconstruction of tissues by dissociated cells,
Science, 141:401-408, 1963.

[2] R. Foty {\it et al} Surface tension of embryonic tissues predict
their mutual envelopment, Development, 122:1611-1620, 1996.

[3] J. Bonner, A Way of following individual cells in the migrating slugs
of {\it Dictyostelium discoideum}, PNAS, Vol. 95, pp. 9355-9359, 1998
[back to schedule]

Frederic Guichard

While local interactions can have a strong influence on the large scale porperties of ecosystems, ecological models often include physical disturbance as an imposed source of variability that is not allowed to interact with biotic processes at the local scale. In intertidal communities dominated by mussels, wave disturbances create gaps in the mussel bed that recover through a succession from empty space to oportunistic species (barnacles and macroalgae), and mussel bed. I present a grid-based simulation of mussel disturbance dynamics as a generic forest fire model (FFM). The model allows local interactions among discrete successional stages defined by aggregating individuals and species into community elements. I first describe the case where each cell of the lattice can be empty, occupied by a mussel bed element, or ``burning'' which corresponds to a newly disturbed cell having unstable edges. The FFM is generalized by allowing global and local (density-dependent) transition rules. I compare the large scale properties of the simulation with field data from 6 sites along the Oregon coast.  Although the dynamics of both natural data and the simulation can show very different behaviors, I look for constrained behaviors in field data that are also observed in the simulation and explained by the geometry of local interactions and by percolation thresholds. The simulation including intermediate stages and a mean-field approximation are also presented.
[back to schedule]

Andy Dobson

Species area curves have been widely used to estimate the eventual rate of extinction as natural habitats are converted to alternate uses.  In this talk I'll discuss an inherent flaw in this assumption and use a modified species area relationship to illustrate why a simple species-area extrapolation may underestimate the true ultimate rate of species extinction.  I'll then use data from the flora of California to support the theoretical arguments.

        California falls into the sea
        That'll be the day I go back to Annandale
                My Old School - Becker & Fagan '72 (whoa!)

I'll conclude by discussing why previous attempts to falsify the species-area extrapolation may have misled us.
[back to schedule]

Kerstin Wiegand

Field data suggest that savannas are patch-dynamic systems composed of many patches (a few hectares in size) in different states of transition between grassy and woody dominance. Following this hypothesis, locally and temporally favorable conditions can lead to the conversion of patches of open savanna into bush encroached thickets. However, over many decades, growth, inter-bush competition, and mortality may transform these thickets back into open savanna. In arid savannas, the conversion of patches of open savanna into bush encroached thickets might simply be driven the spatiotemporal rainfall distribution.

I’ll present my approach in modeling a patch-dynamic savanna, which is motivated by the very different spatial scales of inter-tree competition and rainfall distribution.
[back to schedule]

Rick Condit

[back to schedule]

Tuesday, October 24

Luis Borda de Agua

[back to schedule]
Although fractals have for long been applied in ecology, multifractals have, in contrast, received little attention. In this work we explore the application of the multifractal formalism to the power law form of the species-area relationship. The generalization of the power law species-area relationship by multifractal methods is potentially far reaching: it ties species-area theory to a number of indices used widely to quantify diversity, generates a number of new hypotheses about patterns of biodiversity for further investigation, and it develops new tools with utility in addressing practical issues, such as, conservation.  While fractal sets are described in terms of a single number (the fractal dimension), multifractals require an infinite number of “dimensions”, usually called a spectrum. The characterization of multifractals can be obtained by two different approaches: the method of moments and the method of histograms. The former deals with the statistical moments of the species abundance distribution, but can only be applied when the moments exhibit power law scaling with area. The latter method is of more general application and uses histograms of the species abundance distribution obtained at different areas. The method of moments shows that the power law form of the species area relationship, and the Shannon, Simpson, and Berger-Parker diversity indices belong to a family of equations relating the species number, species relative abundance and area through the moments of the species abundance distribution. Explicit formulas for these diversity indices, as a function of area, are derived. In addition, the method of moments implies relationships between a species range and its relative abundance. The method of histograms highlights the dependence of the shape of the species relative abundance distribution on area. The application of these methods is illustrated with data on tree and shrub species collected in a 50 ha plot in the Barro Colorado Island, Panama. Results from the method of moments show that some moments of the species abundance distribution have power law scaling with the area. Results from the method of histograms show that after appropriate transformation the species abundance distributions obtained at different areas converge to a single curve: the multifractal spectrum.
[back to schedule]

Stuart Sandin

Coral reef fish populations vary greatly through space and time. Data collected through long-term censuses and from biological responses to experimental manipulations reveal that the variability of reef fish communities has definite pattern and structure. Therefore, to elucidate ecological patterns structuring reef fish populations, it is imperative for the researcher to use variance as a source of information and not simply as a source of error. I have developed a series of models describing the propagation of recruitment variability into adult fish populations. Variance propagation is estimated by linearizing a set of coupled population equations and decomposing the spectrum of the output variance. Patterns of mortality, predation, and foraging each affect the characteristics of the resultant demographic variability in qualitatively different ways. I compared predictions derived from the model with data from our field studies and from the literature. This comparison suggests a prevalence of predator-mediated regulation among reef fish populations. I find predators to be most important in regulating population numbers while density dependent growth and other intraspecific interactions are dominant in regulating population biomass. Such decoupling of variability in population numbers and biomass highlights the importance of selecting the appropriate response variable when searching for evidence of specific ecological processes.
[back to schedule]

Eduardo Zea

A strong case has been made for the importance of spatial structure in enhancing species diversity. This effect ultimately results from the
segregation of species in space, which effectively reduces the cost of competition. Vertical spatial structure in every ecosystem is dominated by the ever-present influence of gravity and the fact that solar energy comes from above. This predictability of vertical spatial structure has allowed organisms to adapt to exploit particular regions of vertical space, e.g. canopy vs. under story trees, deep vs. shallow rooted plants, etc. I am developing a model to explore the significance of plant strategies for mining soil water and allocating resources between the below and aboveground environments. I defined 4 plant functional types, based on allocation and root strategies: Above-Shallow, Above-Deep, Below-Shallow, and Below-Deep. I ran contests between these types under wet, mesic, and dry climates, to try to understand the mechanisms by which each type can succeed or fail in each environment. I am also trying to write the equations for the model. Since I am dealing with two distinct time scales (annual and daily), integro-difference equations seemed appropriate: the annual-scale variables follow difference equations that have components that integrate over the daily time scale. The hard part is that the daily time scale also needs to integrate over the vertical spatial structure. I will present the equations, and hope for enlightening comments and suggestions from the audience.
[back to schedule]

Jerome Chave

We have studied the demography, dispersal properties and growth of an understory palm (Astrocaryum sciophilum, Arecaceae).  This species is known to have a very heterogeenous distribution in French Guiana, and we tried to assess whether these patterns were related to habitat heterogeneity, climatic gradients or dispersal limitation. Our results suggest that the latter is a likely hypothesis, for this species was found to have a very slow dispersal rate. A regional-scale map of the distribution of this palm might enable us to identify Holocene refugia in the Guiana shield. The use of this bioindicator would provide better space resolution than floristic methods, that often have a significant sample bias and than pollen cores, very rare in this region. Genetic techniques are being developed on this species to confirm our hypothesis.
[back to schedule]

Ben Strauss

By modelling demography and the evolution of a quantitative trait in continuous space, Kirkpatrick and Barton (1997) recently suggested that gene flow across an environmental gradient could inhibit adaptation away from the center of a species’ distribution, and lead to the formation of a range boundary despite the species’ genetic potential to adapt to more extreme conditions.  My own work yields closely similar results while eliminating assumptions about the distribution of phenotypes in populations, and the pattern of phenotypic or genotypic variance in space.  Furthermore, my model seems to predict decreased variance within marginal populations.  The cost of this added power is the need to track phenotypic classes individually, instead of population attributes.  This in turn requires new assumptions about patterns of phenotypic inheritance, but there is promise for relaxing these in the longer run.
[back to schedule]

Sonia Altizer

The brown tree snake, Boiga irregularis, is an exotic invasive predator that has caused major ecological and economic damage on the Pacific island of Guam.  In the absence of natural enemies and with a wide and vulnerable prey base, brown tree snakes reached phenomenal densities and have caused the extinction of most native forest birds on Guam.  Current snake control measures such as barrier fences, trapping and chemical controls are expensive and largely ineffective. Biological control using parasites or pathogens is an alternate approach to regulate B. irregularis that can be used alone or in combination with other measures.  Our objectives were to develop a simple mathematical model that describes the population biology of brown tree snakes on Guam, and to assess the potential of different types of pathogens to regulate snake abundance. We also surveyed known reptile parasites to identify potential biocontrol agents and characterize their transmission, virulence, and host specificity.  We then modified our basic model to incorporate the features of three broad classes of pathogens, and evaluated the consequences of epidemiological parameters in light of their effects on snake population regulation.  Finally, our study addresses the potential risks of releasing pathogens on the island and highlights critical knowledge gaps that must be examined before the implementation of a biological control program.
[back to schedule]
 

(name)

(summary)
[back to schedule]