ICTP-SAIFR Activities

A large body of existing ecological theory relies on very strong and unrealistic assumptions about the way individuals move and get to interact with each other and with the environment. Specifically, several models assume that individuals behave like the molecules of an ideal gas: following completely random trajectories through the entire area occupied by the population and only interacting with each other when their trajectories intersect. In this presentation, I will first discuss how traditional population dynamics models emerge from ideal gas assumptions. Then, I will present our ongoing research to refine those models using more elaborated tools from random walk theory, spatially-extended nonlinear dynamical systems, and stochastic calculus. I will discuss examples covering both the development of new theory and its application ecological data.

Seventy years ago, two radio engineers emerged from the frenzy of World War II and entered the new field of radio astronomy.Robert Hanbury Brown and Richard Twiss developed an entirely new instrument and technique, based on "correlated noise,"to measure theangular radius of previously unresolvable stars. Initially greeted with skepticism, their work led directly to the birth of quantum optics.  At nearly the same time, Goldhaber et al discovered a tiny unexpected correlation in the first true particle physics experiments; until recently, the "GGLP" effect played a minor role in particle physics.  It would take another 15 years until the connection between these apparently disparate phenomena was realized by Shuryak and others around 1976, just as the new field of heavy ion physics was emerging.  Thus did Hanbury Brown's discovery give birth to femtoscopy, the most direct method to probe space and time at the scales of 1e-15 meter and 1e-22 second.  I will discussthe structures and insights that femtoscopy has revealed in ultra-relativistic ion collisions at RHIC and the LHC and their rolein establishing the hydrodynamic nature of the quark-gluon plasma.  I will briefly discuss current progress to bring a high-energy physics approach to telescope arrays and relaunch stellar intensity interferometry with fast digital electronics and massive computing.

Type Ia Supernovae (SN Ia) are a central pillar of cosmology that is vital to the adoption of the LambdaCDM “Standard Model of cosmology”, but they also uniquely elucidate persistent gaps in the theory. I will detail recent monumental leaps in progress of SNIa analyses that have culminated in the Pantheon+ cosmological analysis which is an improvement of a factor of 2 over its predecessor and that is the current best measurement of the dominant energy density components of the universe and the current best constraint on extensions to LambdaCDM including evolving dark energy. Additionally I will discuss the recent Pantheon+SH0ES constraints of the local value of the Hubble Constant that are now even more robust to systematics and that have now crossed the significant 5 sigma threshold of discrepancy when compared to that of Planck+LambdaCDM. Lastly, I will summarize the lessons learned over the last 5 years and their impact on future SN Ia surveys.  

Theoretical approaches to describe population dynamics tend to focus on a particular scale, balancing the trade-offs of including each type of interaction. For small scales, individual-based models (IBM) can account for an arbitrary level of detail. However, due to its high computational cost, it is limited to small population sizes. For large scales, density-field descriptions (DFD) have been used, but they neglect the dynamics below a cut-off scale. To reconcile both approaches one can perform a coarse-graining procedure that patches together IBMs and DFDs. Since, in general, the emergent properties of the IBMs are out-of-equilibrium and regulated by a complicated set of forces, one should look for computational ways of linking both descriptions. In this talk, I will go through the multiscale framework that collaborators and I have developed to address the large-scale patterns produced by plankton communities.

A fundamental problem in ecology is how individual-level behavior produces macroecological outcomes. The challenges involved in addressing this issue are not only related to the complexity that nature exhibits across scales but also to the fact that theoretical approaches are inevitably scale-constrained, focusing either on the small or large scales. In this seminar, I will present how we applied the multiscale framework that collaborators and I have developed to address the kilometer-scale patterns produced by plankton communities. To further our understanding regarding their bottom-up regulation, we performed a coarse-graining procedure of an agent-based model for a phytoplankton-zooplankton community. The coarse-graining procedure provides a scalable density-field description that maps large-scale features of spatial patterns to individual-level behavior. This powerful connection can guide remote-sensing applications: revealing species' traits and behavior and helping to anticipate regime shifts in the ocean.