Data Science in Science

Data Science in Ecology

Dear Colleagues,
Biodiversity loss, climate change, and accelerating human pressures on natural systems have made the need to understand and predict ecological dynamics more urgent than ever. Ecology has simultaneously undergone a data revolution, shifting from small-scale, investigator-collected field observations to continent- and global-scale datasets generated by remote-sensing platforms, automated sensors, genomic assays, and coordinated monitoring networks. The questions ecologists can now pose, and the spatial and temporal scales at which they can pose them, have expanded dramatically.
The statistical and computational tools available to ecologists have not kept pace. Modern ecological datasets are heterogeneous, massive, and subject to observation error and missingness. Extracting reliable inference requires methods that can fuse multiple data streams, incorporate mechanistic understanding, propagate uncertainty, and support the conservation and management decisions that depend on credible ecological science.
This Special Issue of Data Science in Science invites original research and review papers at the frontier of statistical methodology and computation in ecology. Three environmental and ecology themed events are partnering on this Special Issue, and we especially welcome submissions building on work presented at these participating events:

• 2026 ENVR Workshop: the biannual meeting of the Statistics and the Environment Section
  of the American Statistical Association,
• LACSC-TIES-EnviBayes-EnvrASA 2026: International Conference on Statistics, Data Science,
  and Computing for the Environment and Climate Change, and
• ISEC: International Statistical Ecology Conference.

Participation in one of these conferences is not required; submissions are open to all.
Submissions may introduce new methods, benchmark or extend existing ones, present novel data resources, or develop integrative frameworks applied to real ecological problems. We evaluate submissions on the basis of scientific rigor, technical depth, and ethical standards, regardless of perceived novelty, and all articles will conform to the regular standards and scope of the journal.
Topics of interest to this call include but are not limited to the following:

1. Data integration and multi-source modeling Frameworks for jointly analyzing heterogeneous ecological data, including integrated population models, occupancy models with multiple observation types, and hierarchical structures that reconcile data collected at different scales or under different sampling protocols.
2. Citizen science and opportunistic data Statistical methods for working with large-scale presence only or volunteer-collected data, addressing spatially biased sampling, heterogeneous detection,
   and the tradeoffs between data volume and data quality.
3. Machine learning and AI for ecological systems Development and application of modern ML and AI approaches to ecological prediction, species identification, abundance estimation, and habitat modeling, with attention to interpretability, generalization, and integration with biological knowledge.
4. Bayesian modeling and uncertainty quantification New contributions to Bayesian hierarchical modeling, scalable posterior computation, prior specification, and uncertainty propagation in complex ecological models, including methods that support transparent, decision-relevant inference.
5. Mechanistic and agent-based modeling Statistical frameworks that embed ecological process models, including individual-based and agent-based formulations, enabling inference on mechanisms from observational or experimental data.
6. Causal inference in ecology Methods for estimating causal effects from non-experimental ecological data, including approaches for handling confounding, leveraging natural variation, and assessing the consequences of environmental interventions.
7. Spatiotemporal modeling of ecological dynamics Approaches for characterizing how species, populations, and communities change across space and time, including models for non-stationarity, range shifts, and abrupt transitions.

Sincerely yours,
Toryn L. J. Schafer, Mevin B. Hooten, Erin M. Schliep, Benjamin W. Hoose, Elise F. Zipkin, John Fieberg, and David S. Matteson
Special Issue Editors
Data Science in Science