Sarah Iams and Karna Gowda highlight Spatio-Temporal Dynamics in Ecology
LEIDEN, NL - Gathering from Japan, Brazil, Israel, Canada, USA, UK, Germany, and Spain, ecologists and mathematicians from around the world came to the Netherlands to discuss animal diffusion patterns, biodiversity, and ecosystem pattern formation at the Lorentz Center. The workshop, "Spatio-Temporal Dynamics in Ecology," was organized by Arjen Doelman (MCRN), Jef Huisman, Johan van de Koppel, and Antonios Zagaris (MCRN), each of the Netherlands. Participants gathered to learn about current challenges and to discuss potential new directions in the mathematical study of ecology. The workshop featured long and short talks, and time was built in for discussion via long lunch breaks, frequent coffee breaks, and small discussion periods attached to each talk.
Speaking on ecological complexity, Jef Huisman (University of Amsterdam, Netherlands) used cyanobacteria as an example to illustrate the importance and difficulty of predicting the temporal evolution of populations. These bacteria sometimes create toxic blooms that poison drinking water, and advance prediction remains a challenge. Addressing a related question, Nigel Goldenfeld (University of Illinois - UC, USA) emphasized that the randomness inherent to discrete individuals in biological populations can lead to population oscillations that do not appear in continuum models, and this should not be overlooked in models of population dynamics. This led to a discussion of the importance of spatial distributions of predators and prey. John Fryxell (University of Guelph, Canada) presented experimental results of spatial and temporal population fluctuations from phytoplankton and zooplankton living in repurposed beer vats. Workshop participants discussed efforts to model the vertical distribution of phytoplankton off the coast of Hawaii, and applications of ocean phytoplankton measurements to infer mixing in the ocean.
Chris Jones (MCRN, UNC - Chapel Hill, USA) showed how data and modeling can be combined via data assimilation to yield insights into the functioning of terrestrial and aquatic ecosystems. In one example, this approach showed how rising anoxic water can lead to massive "fish kills" in a lake ecosystem. This approach is akin to the data assimilation (www.mathclimate.org/research/jdas) work done by members of the MCRN and others. In a method focused on making use of much less data, Maarten Eppinga (Utrecht University, Netherlands) showed that a single snapshot of a system can yield insight into dynamic processes. His work examines regions with sharp transitions between two tree species, using details of the distribution of individuals of each species in the transition zone to extract information about the transition process. Alan Hastings (UC Davis, USA) presented the idea of subsidies, where exchange between different parts of the environment help maintain biodiversity. He gave the example of a bear discarding part of a salmon into a forest as such an exchange.
Many talks at the meeting focused specifically on animal movement. Ghandi Viswanathan (NATAL--RN, Brazil) introduced the audience to the idea of anomolous diffusion in this context. In animal foraging, motion data often has larger jumps than would be predicted by a random walk or standard diffusion model. Jonathan Potts (University of Sheffield, UK), David Sims (Marine Biological Association, UK), and Nick Humphries (Marine Biological Association, UK), discussed the challenges of fitting distributions or models to animal motion data. One frequent observation was that truncated Levy distributions often fit data well. Johan van de Koppel (NIOZ-Yerseke, Netherlands) presented results of motion models for mussel motion data. When mussels moved freely, they followed a Levy law, but when they were in a higher density region with many collisions, their motion was better described by a Brownian motion model.
Pattern formation in ecosystems was another theme of the meeting. The formation of mussel beds presented by van de Koppel is one example. Rob Pringle (Princeton University, USA) described termite mounds, which form a regular pattern across the landscape in regions of Kenya, as thermal refuges. The vegetated mounds are cooler during the day and are hotspots of activity in their ecosystem. The type of regular patterning exhibited by these mounds may hold clues to critical transitions in nature. Simon Levin (Princeton University, USA) cited the history of the study of early warning indicators of transitions. Noting many papers from the 1970s on catastrophe theory, he cautioned the scientific community not to abandon these ideas again. Mary Silber (MCRN, Northwestern University, USA) discussed a suspected ecological transition sequence, where change over time in vegetation patterns is hypothesized to be a transition indicator. She noted that the suspected transition sequence is not the only possible sequence predicted by the models. Ehud Meron (Ben Gurion University of the Negev, Israel) took these ideas in a restoration direction, suggesting that imposing certain patterning on a landscape can aid in the restoration of vegetation in semi-arid regions.
Thank you, Sarah and Karna!
* For the complete schedule and titles of Short Talks given by other MCRN participants, please see here.