About This Project

Mangroves are valuable ecological resource providing habitat for many species. Species assemblages vary in life history strategies and spatial and temporal scales, thus recording real diversity using traditional methods is almost impossible. Here, we aim to predict species interaction networks combining eDNA and machine learning to provide insights in ecosystem function and emergent properties, a fundamental information to more appropriate conservation strategies and restore ecosystem services.

Ask the Scientists

What is the context of this research?

In the face of an accelerated loss of wild flora and fauna worldwide, there is an urgency to conserve and restore ecosystem services. To achieve this, recent studies point to focus on species interaction networks rather than the rarity status of species to more appropriate conservation strategies and to maintain ecosystem integrity(1,2). In this way, mangroves are an invaluable ecological resource as habitat for a wide variety of species, from the canopy of its trees to below its roots. Given the spatial and temporal complexity that underlies the study of the composition of the species diversity(3), in this proposal we combine environmental DNA (eDNA) technology and machine learning to better understand the function of the mangrove ecosystem and emergent properties

References

What is the significance of this project?

The adoption of eDNA has been facilitated by the improved access to sequencing technologies and the constant advances in the methodological framework. This opened the opportunity to explore innovative tools and adjust protocols to overcome challenges in studying eDNA of marine diversity. In this context, studies based on eDNA represent a few opportunities not only to envisage the species diversity that is sheltered by mangroves but also brings the opportunity, using machine learning, to predict species interaction networks in key ecosystems, such as mangroves. This provides insights into the underlying changes in ecosystem function and emergent properties, a fundamental information to implement more appropriate conservation strategies and restore diversity and ecosystem services.

What are the goals of the project?

Given the spatial and temporal complexity that underlies the study of the composition of the species diversity that inhabits mangroves, the present proposal aims to predict species interaction networks through machine learning. To do so, we will first characterize the composition of taxonomic diversity that inhabit the mangrove ecosystem using environmental DNA metabarcoding. This provide insights in the ecosystem function and emergent properties, a fundamental information to implement more appropriate conservation strategies and restore diversity, function, and ecosystem services. This information is imperative not only for prioritizing available economic resources but also for the survival of vulnerable or threatened mangrove-dependent species.