In the context of a rapid decline in biodiversity, effective monitoring is crucial to evaluate the presence, absence, and abundance of species in their natural environments. Among all ecosystems, river biodiversity is declining at an alarmingly higher rate. However, monitoring these watercourses is often challenging due to the intrinsic difficulties of accessing such environments.

Environmental DNA (eDNA) is emerging as a powerful and non-invasive method for biodiversity monitoring, often outperforming traditional invasive practices such as electrofishing, as demonstrated in various studies. Nevertheless, the temporal and spatial resolution of this method, particularly in lotic (flowing water) environments, still requires thorough evaluation.

This presentation will showcase the results of a four-year study comparing traditional monitoring techniques for amphibians and fishes with eDNA metabarcoding analyses to assess both vertebrate and invertebrate biodiversity. Special emphasis has been placed on eDNA capture methods, accompanied by an analytical comparison of their performance. 

Global warming undoubtedly contributes to the erosion of biodiversity. However, the mechanisms by which ecosystems respond to rising temperatures remain unclear, preventing the formalisation of a predictive framework capable of a priori identifying more sensitive communities. Theoretical approaches using food web models, while able to infer fundamental processes governing community response to warming, may be limited by the amount of biological information they consider. In this talk I will question the genericity of classical results by comparing them with what is obtained when considering (i) empirical food web structures, (ii) effect of ecosystem type, and (iii) the influence of temperature on the foraging behaviour of consumers.  

Among ecological networks, metawebs represent the general pool of all living entities and their potential interactions within a given area, the construction of which generally requires a high effort for the selection and analysis of data available in the scientific literature. We evaluated the feasibility to exploit already existing ecological data to increase speed and easing the construction of a regional metaweb. The metaweb obtained completely in-silico, was then analyzed to verify its adherence to the literature and the likelihood of the interactions coded within it.

The results of our analyses highlight the adherence of the metaweb built in-silico to general knowledge of the role of some species considered important, without resorting to field studies or ad-hoc selection of sources. 

The result of this experimentation therefore suggests the feasibility of our project pipeline as a tool for building regional metawebs, which may be useful in the future as a starting point for carrying out preliminary analyzes and comparisons between ecological networks, especially urban ones.  This last point turns out to be important since we still need to increase the amount of open data present in repositories about this specific habitat.

Ecological communities, like many other complex systems, continue to intrigue researchers to find a way to explain how the biodiversity observed in nature is maintained. This raises a fundamental question: what sustains the stability and the coexistence of species in these ecosystems? Ecological models have largely rested on the premise that species primarily engage in pairwise interactions. Yet, in ecological systems, interactions can often occur in groups of three or more individuals. That is why, in this work, we will study a model for competitive community, exploring how higher-order interactions together with different network structures, affect species coexistence and the stability of ecological dynamics. Relying on numerical simulations and an analytical treatment of the system’s dynamics, our findings reveal that, network topology together with the presence of higher-order interactions, play pivotal roles in the emergence of coexistence and stability of multi-species competitive communities. For example, we find that for well-mixed populations, when species present the same physiological rates (e.g. birth and death rates), even a small fraction of higher-order interactions are able to stabilize the dynamics. Instead, when physiological rates are different between species, their relative variance dictates the critical fraction of higher-order interactions needed to achieve stable coexistence. These discoveries represent a step forward in our understanding of ecological dynamics and open up promising avenues for future research.

Ecological theory suggests that modular networks can limit the spread of disturbances, thereby enhancing ecosystem stability, yet empirical support from natural environments is limited. We conducted two complementary experiments in coastal marine systems to assess whether modularity limits the expansion of algal turfs following disturbances to habitat-forming species. In the first study, we combined a field experiment with a metacommunity model to assess the role of modularity in buffering the spatial spread of algal turfs within three replicated, canopy-dominated macroalgal networks. Results showed that algal turfs largely remained within the initially perturbed module, supporting the hypothesis that modularity can effectively constrain the spread of spatial disturbances. The model reinforced the empirical findings, suggesting that modular networks exhibit greater resistance to perturbations than random macroalgal networks. Building on this, we compared turf diffusion in networks with different topologies, by establishing modular and random networks in Posidonia oceanica meadows. This second study further showed that modular networks are more effective than random ones at containing disturbance spread. These experiments underscore the key role of modularity in mitigating disturbances in natural systems, with important implications for conservation planning and habitat management in fragmented landscapes.

The distribution of biomass across trophic levels (e.g. herbivores and carnivores) is a key measure in ecology, linking trophic structure and community dynamics.

However, biomass distributions have typically been investigated from a food chain perspective, ignoring channels of energy transfer (e.g. omnivory) that occur in complex food webs. In this talk I will present recent work which addresses this shortcoming, characterising the biomass structure of freshwater, marine and terrestrial food webs, spanning a broad gradient in community biomass. We find a consistent, sub-linear scaling pattern whereby predator biomass scales with the total biomass of their prey with a near ¾-power exponent within food webs - i.e. more prey biomass supports proportionally less predator biomass. Across food webs, a similar sub-linear scaling pattern emerges between total predator biomass and the combined biomass of all prey within a food web. These general patterns in trophic structure are compatible with a systematic form of density dependence that holds among complex feeding interactions across levels of organization, irrespective of ecosystem type.