Satellite ECM 2024 Conference: CIEM24

To understand how climate shapes ecosystem dynamics, we need to explore the role and origins of stochasticity in ecological systems. In this talk, I will introduce the three general types of stochasticity: demographic, environmental, and individual heterogeneity. Using the Levins model, a classic ecological framework, I will demonstrate how these different forms of stochasticity influence ecosystem dynamics. By extending the Levins model, I will illustrate its fundamental significance in analyzing species-rich systems, eco-evolutionary dynamics, consumer-resource interactions and functional responses, and host-parasitoid interactions.

Although spatial environmental stochasticity, or spatial heterogeneity in environmental conditions, often underlies ecosystem dynamics, non-random spatial patterns in species distribution can also arise under uniform environmental conditions. These biologically-driven patterns, known as diffusion-induced or Turing patterns, were first described by Alan Turing, who showed that they develop from a diffusion-induced instability.

I will conclude my talk by examining a host-parasitoid interaction in a coffee plantation, presenting preliminary findings on the variety of spatial patterns produced by this system, and summarizing a general methodology for studying spatially-extended stochastic systems in an accessible manner.

In recent years there has been an increasing awareness of the risks of collapse or tipping points in a wide variety of complex systems, ranging from human medical conditions, pandemics, ecosystems to climate, finance and society. They are characterized by variations on multiple spatial and temporal scales, leading to incomplete understanding or uncertainty in modelling of the dynamics. Even in systems where governing equations are known, such as the atmospheric flow, predictability is limited by the chaotic nature of the system and by the limited resolution in observations and computer simulations. In order to progress in analyzing these complex systems, assuming unresolved scales and chaotic dynamics beyond the horizon of prediction as being stochastic has proven itself efficient and successful. When complex systems undergo critical transitions by changing a control parameter through a critical value, a structural change in the dynamics happens, the previously statistically stable state ceases to exist and the system moves to a different statistically stable state. To establish under which conditions an early warning for tipping can be given, we consider a simple stochastic model, which can be considered a generic representative of many complex two state systems. We show how this provides a robust statistical method for predicting the time of tipping. The method is used to give a warning of a forthcoming collapse of the Atlantic meridional overturning circulation.

References: Peter D. Ditlevsen and Susanne Ditlevsen (2023), Warning of a forthcoming collapse of the Atlantic meridional overturning circulation. Nat Commun 14, 4254

Analyses of both models and data in ecology are still focused on equilibrium or long-term dynamics, with some notable exceptions.  Although recent work on tipping points does include approaches based both on underlying changing environments and dynamics on different time scales, the possible situations where dynamics on different time scales are important are much more general.  Using new mathematical ideas one can address questions of dynamics on ecological time scales, rather than longer times, and include other kinds of underlying environmental change.  The importance of this way of analyzing ecological systems is clear in consideration of changing environments due to anthropogenic influences. The analyses demonstrate that there are wide ranges of ecological situations where standard analyses based on assuming asymptotic behavior are misleading.  Additional cases where explicit time dependence is included in dynamics shows further complications.  Different kinds of situations where long transient behavior is expected can be identified.  In particular, adding space, which essentially makes systems very high dimensional, is often likely to lead to long transient dynamics.  This work also, unfortunately, points out challenges in trying to identify systems where future sudden shifts in system state due to transients are going to occur, since transient behavior of a system with long transients will be asymptotic or long-term behavior of a corresponding system without transients. Examples of ecological systems illustrating the conclusions, including coral-algal-grazer systems will be discussed in light of the general theoretical results.

I have spent 35 years of my professional career observing different ecological systems and how they respond to environmental change. In the beginning, it was obvious that stochastic climate played a crucial role in driving primary productivity and how this association pervaded into the entire ecosystem. It was not until several years later that I found out the importance of temporal and spatial scales for a better understanding of climate stochasticity and its effects on ecological processes. This included a necessary look at the geological time and evolutionary life histories involved in the complex responses of organisms to climate variability. I have ended up with the feeling of gathering a limited empirical knowledge of a range of organisms and ecosystems, and I struggle now to find out if there are good questions that remain unanswered. For instance, is there an evolutionary memory selected to tune the response of
organisms to climate stochasticity? How has climate cyclicity over the recent Quaternary affected life histories and the responses of organisms to this new climate regime? How the evolutionary pace of life has shaped the response of different organisms to climate stochasticity? I will summarise here my field experience over the years and how I deal with what remains between the trivial evidence (i.e. climate influences ecological processes) and the challenges of the unanswered (i.e. copying with the responses of complex ecosystems).

Compartmental models play an important role to describe the dynamics of systems that involve mass movements between different types of pools. We develop a theory to analyse the average ages of mass in different pools in a linear compartmental system with time-dependent (i.e. non-autonomous) transfer rates, which involves transit times that characterise the average time a particle has spent in a particular pool. We apply our theoretical results to investigate a nine-dimensional compartmental system with time-dependent fluxes between pools modelling the terrestrial carbon cycle.

Joint work with Alan Hastings, George Chappelle, Matthew Smith, Yiqi Luo, Folashade Agusto, Benito Chen-Charpentier, Forrest Hoffman, Jiang Jiang, Katherine Todd-Brown, Ying Wang, Ying-Ping Wang.

Most marine fishes have biphasic life cycle during which larvae disperse into the open ocean before coming back to a coastal habitat and renew adult population. This incredible journey represents a massive challenge for such tiny organisms that will have to not only orientate themselves, but also find food, escape predators, grow, and face massive external and internal changes to transform into juveniles that look like miniature adult. This transformation, is called metamorphosis and is under the control of thyroid hormones. It must occur during the transition between the oceanic and costal phase. Thyroid hormones are thus not only responsible for triggering metamorphosis, they in fact coordinate multiple processes to ensure that the outcoming juvenile will be equipped to adapt to its new habitat at the right time. Recent evidences are thus suggesting that thyroid hormones have an ecological function that is to ensure that developmental transitions are aligned with environmental transition. However, during such critical period, marine fish can be particularly vulnerable to external factors such as habitat quality, pollution, increase temperature, ocean acidification etc. We will present the experimental evidences supporting this model and discuss its implications.

A core challenge of our time is to predict how organisms will respond to global environmental change. Yet most of our current predictive tools are correlative (statistical) methods that largely overlook the mechanisms underlying organism-environment interactions and thus have a limited capacity to extrapolate responses to unprecedented environmental scenarios. An emerging area of research is committed to developing mechanistic models to capture these interactions – a multidisciplinary endeavor integrating disparate fields such as thermodynamics, physiology, and evolutionary biology. A starting point towards achieving this integration has been to develop biophysical models describing the balances of heat, water and other aspects of energy and mass exchange between organisms and their environment and translate these into metrics of physiological performance and fitness. I will show recent advances in biophysical modeling and examples ranging from simple calculations of body temperature at specific times and locations, to broad-scale patterns including species distributions across both terrestrial and aquatic realms. I will highlight that biophysical models are opening new grounds of research and have the potential to transform ecology into a predictive science.

Forecasting the future impact of climate on ecosystems needs long time series data, which is often scarce for ecological communities. However, it is possible to use biological proxies to characterize past ecosystems. These proxies have several advantages. On the one hand, they spam over time periods including major climate changes. On the other, provide information about community composition, which allow inference about system level properties. Thus, it is possible to identify shifts in ecological communities and disentangle the drivers behind them. In this presentation I will show the preliminary results of our work on reconstructing ecological networks during the Holocene using the palynologycal records in mountain lakes. These networks show us the response of ecological communities to the conditions in their surroundings, including climate and human activities, providing valuable information about how ecosystems respond to global change drivers over time. Our results could help to anticipate how will ecosystems respond to climate in the near future, predict the arrival of major shifts, and highlight the value of ancient biological proxies for climate change research.

Understanding and predicting abrupt ecological shifts in spatially extended ecosystems are vital for environmental management. Ecosystems like those in the North Atlantic Ocean and semiarid regions can undergo significant transitions due to rapid changes in key indicators. Previous research highlights the utility of spatial correlation in forecasting these tipping points, but early identification remains challenging due to factors like data resolution and noise.

It has been shown that the spatial permutation entropy (PE) can detect early warning signals in spatiotemporal data. Here, by analyzing the temporal variability of spatial indicators, specifically spatial correlation and PE, we aim to identify patterns signaling ecological shifts. We apply our approach to North Atlantic chlorophyll levels and semiarid vegetation data, along with model simulations.