Periodic Reporting for period 1 - ECONET (Rewiring of ecological networks under global change)
Reporting period: 2023-12-01 to 2025-11-30
One of the main consequences of the biodiversity crisis is the reorganization of ecological networks. This reorganization, usually termed ‘interaction rewiring’, can drastically alter nature’s functions and services to humanity. Therefore, predicting interaction rewiring is paramount to predicting the future structure, functioning and stability of the ecosystems. Until now this has not been done because quantifying and predicting rewiring into the future requires extensive data on species traits and interactions and computationally efficient and ecologically relevant analytical tools. The main objectives of ECONET are to quantify the potential of rewiring in mutualistic networks globally and to predict how mutualistic networks rewire due to global change (climate change, land cover change, human-driven extinctions).
For achieving the aims of the project, I have used empirical data of pollination and seed dispersal networks to predict probabilities of all pairwise interactions in metanetworks (network covering all possible pairwise interactions) with machine learning. I have quantified the species' rewiring potential as their interaction niche breadths in the metanetworks both in hummingbird-plant pollination networks in the Americas and bird/mammal-plant seed dispersal networks globally. I have also assessed the links between the potential of rewiring and the anthropogenic threats that the interacting species face using bird/mammal-plant seed dispersal networks globally. The first key paper with the conceptual and methodological framework for understanding the potential of rewiring has been published in Ecology Letters. This publication sets the scene for all the following works of this project but also beyond when studying the functional roles of interacting species.
ECONET has advanced biodiversity science by developing a predictive framework for ecological resilience that moves beyond descriptive network approaches. By integrating species traits, interactions, and occurrences, I created new metrics—rewiring capacity and rewiring potential—that quantify how ecological networks reorganize under environmental change. This provides, for the first time, a globally applicable and data-efficient method to anticipate ecosystem responses even in data-poor regions. Our case study on plant–hummingbird pollination networks in Americas shows that species exhibit high functional flexibility, indicating that many ecosystems may retain pollination services despite environmental change if habitat diversity is maintained.
Complementing this, my global analysis of seed disperser communities introduces a three-dimensional vulnerability framework, combining exposure, sensitivity, and adaptability to assess how human habitat modification affects ecosystem function. The framework identifies priority areas for targeted interventions—from low-intensity management in adaptable systems to rewilding and restoration in highly vulnerable regions (e.g. Southeast Asia, Madagascar).
Potential impacts:
Early-warning indicators of ecosystem function loss.
Data-driven prioritization of conservation and rewilding.
Integration of rewiring and vulnerability metrics into conservation.
To ensure further use of the work, future steps include expanding trait and interaction databases and applying the concepts and methods across additional ecosystems and interaction types. Overall, the project sets a new benchmark for predictive ecology, equipping policymakers and practitioners with actionable tools to safeguard biodiversity and ecosystem functions under global change.
Complementing this, my global analysis of seed disperser communities introduces a three-dimensional vulnerability framework, combining exposure, sensitivity, and adaptability to assess how human habitat modification affects ecosystem function. The framework identifies priority areas for targeted interventions—from low-intensity management in adaptable systems to rewilding and restoration in highly vulnerable regions (e.g. Southeast Asia, Madagascar).
Potential impacts:
Early-warning indicators of ecosystem function loss.
Data-driven prioritization of conservation and rewilding.
Integration of rewiring and vulnerability metrics into conservation.
To ensure further use of the work, future steps include expanding trait and interaction databases and applying the concepts and methods across additional ecosystems and interaction types. Overall, the project sets a new benchmark for predictive ecology, equipping policymakers and practitioners with actionable tools to safeguard biodiversity and ecosystem functions under global change.