Our lab explores how ecological systems change across space and time by working at the intersection of data and the environment. We are innovative in integrating multi-source data to examine questions at scales that cannot be investigated using individual field studies.
​
By integrating and harmonizing large and disparate data sets -- millions of mosquito trap records, public health surveillance, community science, and long-term climate and land cover data -- we aim to to identify patterns and understand processes to generate actionable insights at scales beyond individual field studies. This work centers on three interconnected themes: distributional ecology, spatiotemporal phenology, and prediction and forecasting.​​​​​​
​
​​
Where are species and communities located and why?
Understanding and predicting species distributions is a core question in ecology with contemporary and future relevance across multiple sectors.
​
We investigate the drivers of species distributions by integrating biodiversity, environmental, and surveillance data within species distribution models (SDMs) and joint species distribution models (jSDMs).
By combining digitized museum and monitoring records, community science data, and new field collections, we quantify how climate, landscape structure, and species traits shape current and future distributions across scales.
These approaches enable prediction of both single-species and community-level patterns, advancing fundamental knowledge in population and community ecology, while supporting applications in vector-borne disease prevention and management under ongoing environmental change.​​
jSDM poster presented by Amely Bauer at the Entomological Society of America Annual Meeting in Vancouver 2021 and Bauer et al. 2024: https://doi.org/10.1002/ecs2.4771
When are species and communities located and where?​​​​
Phenology, or the timing of seasonal recurrence of key life history events, is central to life on Earth. In vector-borne disease systems, the seasonal onset, peak abundance, and decline of vectors and hosts determine when and where transmission can occur, yet key gaps remain in understanding how these dynamics align across species and landscapes, and how they shift under environmental change. ​​​​Our lab integrates long-term surveillance, ecological monitoring, and community science data to quantify spatiotemporal patterns in species phenology and abundance to understand and predict responses to environmental conditions.
We are developing tools to extract phenological metrics from complex, multigeneration species, leveraging change-point detection and ML/AI approaches. These tools support ongoing work on multivoltine phenology, vector–host overlap, and more broadly, ecological responses to climate variability across scales
Phenological overlap increases host–vector interactions and transmission potential, while mismatch reduces both.
Amely Bauer is developing R tools to parse overlapping mosquito generations and predict onset and offset using change-point detection and ML/AI.
When will species and communities be located and where?​​​​
Anticipating future states of ecological systems is a rapidly advancing field in ecology, with the potential to transform how we understand and manage ecosystems under environmental change. Ecological forecasting provides a powerful framework by integrating long-term biological observations with environmental data and the spatiotemporal processes that drive ecological outcomes. Recent advances in data availability and computational methods now enable a shift from retrospective analyses toward actionable, near-term predictions.​​
Yasmin Tavares used Gaussian Markov Random Fields to capture weekly spatiotemporal structure of 2018 Florida Department of Health West Nile virus sentinel chicken seroconversion. Red indicates greater spatiotemporal structure; blue lower structure.
Our lab develops and applies forecasting frameworks that integrate multisource ecological and public health data with spatiotemporal models. By digitizing and harmonizing long-term surveillance records and linking them with environmental drivers, we identify lagged climate signals and dynamic processes that shape future species distributions and virus activity. Ongoing work focuses on building automated data pipelines, environmental data assimilation, and decision-support tools to support proactive surveillance, prevention, and management.
Alex Baecher developed spatiotemporal models using long-term Florida Department of Health West Nile virus sentinel chicken surveillance data and hindcasted values over a 20 year time period across the state. Published here: https://doi.org/10.1016/j.scitotenv.2025.180308
Our work has been funded through multiple federal and state programs, including the USDA, Department of Defense, CDC, and Florida Department of Agriculture and Consumer Services, as well and University of Florida Seed Funds
​
We are collaborative and happy to connect!
Further information highlighting this work, as well as the work of all team members and broader collaborations can be found here: ​​​​
​
https://scholar.google.com/citations?hl=en&user=MVSUB9sAAAAJ&view_op=list_works&sortby=pubdate
​​
​
Campbell Lab GitHub:
​
https://github.com/Campbell-Lab-FMEL​​​​​​