How do we know when ecological predictions remain reliable under global change?
I am a computational ecologist at the University of Konstanz, interested in understanding when ecological predictions remain reliable as environments, ecological relationships and the data we use change. Whether forecasting species distributions, assessing invasion risk or predicting biodiversity responses to global change, we increasingly rely on predictive models to guide scientific research and conservation decision-making. My research combines ecological theory, biodiversity data and machine learning to investigate how changing environmental conditions, imperfect data and model assumptions influence the reliability of ecological predictions.
Research Themes
Reliable ecological prediction
Ecological models are frequently used to forecast biodiversity change, but their reliability depends on whether the conditions under which they were developed remain valid. I investigate how distribution shifts, changing ecological relationships, and data limitations influence predictive uncertainty and the reliability of ecological forecasts.
- Reproducible WiSDM: a workflow for reproducible invasive alien species risk maps under climate change scenarios using standardized open data (2024)
- Mechanistic models project bird invasions with accuracy (2023)
- The global geography of plant invasion risk under future climate and land-use changes (2026)
Human-mediated biodiversity change
Human activities are reshaping biodiversity by altering species movements, distributions, and opportunities for establishment. My research investigates how processes such as urbanization, recreation, and land-use change influence biodiversity patterns, with a particular focus on biological invasions.
- Accounting for residential propagule pressure improves prediction of plant invasion (2016)
- Recreational freshwater fishing drives non-native aquatic species richness patterns at a continental scale (2017)
- Differences in characteristics between naturalized threatened plants and other threatened plants (2025)
Biodiversity data and reproducible science
I develop open biodiversity resources and reproducible workflows that improve transparency and enable more robust ecological analyses.
- The updated Global Naturalized Alien Flora (GloNAF 2.0) database (2025)
- A checklist recipe: making species data open and FAIR (2020)
- Occurrence cubes: a new paradigm for aggregating species occurrence data (2020)
Contact
Please get in touch via email: amy.davis (at) uni-konstanz (dot) de