We invite contributions spanning the full intersection of machine learning, data science, and climate research. The Climate Informatics Conference welcomes work that advances understanding of the Earth system, supports climate adaptation and mitigation, or improves the transparency and accessibility of climate data and models.
Topics of interest include, but are not limited to:
- Knowledge-guided/hybrid models for Earth-system modeling
- Prediction, nowcasting, downscaling, and extremes
- Data assimilation, reanalysis, and remote sensing
- Causal discovery, interpretability, calibration and uncertainty quantification
- Climate impacts, adaptation/mitigation decision support and policy
- Reusable datasets & benchmarks, evaluation tools
- Principled positions on ethics, reproducibility, and open science
Submissions should clearly articulate their relevance to climate science, use appropriate baselines, and include rigorous evaluation such as uncertainty, robustness, and limitations. Authors are strongly encouraged to enable reproducibility by sharing code, data, and other relevant artifacts when possible.
Tracks & Formats
- Full papers
- Extended abstracts
Important Dates
All below deadlines are at 11:59pm (23:59) AoE (Anywhere on Earth).
- Conference paper & extended abstracts: Jan 16, 2026
- Supplemental material for full papers: Jan 16, 2026
- Notification of results: Feb 17, 2026
- Camera-ready copy of full papers: Mar 10, 2026
Paper Categories
To ensure that research submissions are reviewed by the most appropriate experts and to align with our publishing partner, the Environmental Data Science (EDS) journal, we have structured the submissions into four distinct paper types. These categories apply to both full conference papers and extended abstracts, with the exception of position papers.
You should view these categories as a way to logistically organize and guide the reviewing process, ensuring your work is evaluated within the correct context and against relevant benchmarks.
Your main task as an author is to carefully choose the single category that best fits your submission’s primary contribution.
- How to Choose Your Category: Review the descriptions below to find the category that most accurately reflects the core novelty and focus of your submission. If your work spans multiple categories, select the one that represents the most significant contribution of your manuscript.
- Mandatory Selection: You will be asked to identify the type of your submission in the submission portal.
Detailed paper types
Aligned with the Environmental Data Science journal, where conference proceedings will be published, we invite submissions that fit into one of the four types:
- Application papers: Research progress, or tackling a real-world problem, in an environmental field, enabled by data science. For example, AI or data science could be used for understanding of climate applications, or improving forecasting tools.
- Methods papers: Novel data science methodology inspired by an environmental problem or application. Typically the methodology should be demonstrated in one or more climate applications.
- Data papers: Describe the development and utility of re-usable environmental data sets to support the advancement of climate informatics research (e.g., benchmarking data sets to support objective evaluation of data sciences methods). The data must reside in publicly accessible repositories. These papers promote data transparency and data reuse.
- Position papers (for full papers only): Examples include but are not limited to:
- providing an authoritative, personal view on the uptake or obstacles to AI and data science approaches for environmental problems.;
- exploring issues related to the use of environmental data, including ethical, legal and policy issues, as well as data standards, protocols, and services.
Call for conference papers
The material submitted should relate directly to Climate Informatics as a topic. This can include algorithms with potential application in geosciences, or datasets and physical problems which pose a specific problem in geosciences. We particularly welcome submissions applying machine learning and advanced statistics to a climate-related issue.
Accepted papers will be published in Environmental Data Science Journal and they will be considered for talks or posters at the conference.
Submission Instructions
Original, unpublished full papers of up to 8 pages (acknowledgement and references not included) are invited. The manuscript should be prepared using the EDS LaTeX/Word templates.
Call for extended abstracts
Extended abstracts should be at least 300 words and at most 2 pages total, including figures and references. The research should be original. Preliminary or ongoing work is welcome.
Each submission must include at least one figure or schematic that visually summarizes the research.
Submissions should contain enough information for reviewers to assess quality and be organized in sections such as:
Introduction & Objectives: motivation, background, main objectives or contributions, and fit to climate informatics.
- Introduction & Objectives: motivation, background, main objectives or contributions, and fit to climate informatics.
- Methods & Data: methods, models, and datasets.
- Preliminary Results (and/or Expected Outcomes): main results if available, or expected outcomes, implications, and limitations.
Submission Instructions
The submission should be prepared using one of the following templates.
Accepted abstracts will be considered for posters.
Review Process and Criteria
The review process for all submissions is double blind: author identities must not appear in the manuscript or supplementary material (directly or indirectly). Any breach may result in desk rejection. Posting related work on arXiv is allowed; if cited, do so in the third person and avoid language that reveals authorship. Reviewers’ identities are hidden from authors.
CMT Acknowledgment
The Microsoft CMT service was used for managing the peer-reviewing process for this conference. This service was provided for free by Microsoft and they bore all expenses, including costs for Azure cloud services as well as for software development and support.