The GST Climate Datathon is a call for open data and tools to support the GST ahead of COP27. Outcomes should be aimed toward making data sets interoperable, providing aggregated insights, support tools, and visualizations. Three winners from each category will have the opportunity to present their findings at the COP27 Presidency’s ‘Science and Information Day’.
There are three thematic areas of the global stocktake: mitigation, adaptation, and means of implementation and support (including finance, and loss and damage). Information is being prepared for the global stocktake between now and June 2023, with technical dialogues also ongoing until June 2023. At COP28 in 2023 the outputs of the global stocktake will be considered and are expected to inform parties in updating their commitments toward the Paris Agreement.
We ask the hackers to embark on an unusual task — help us collectively build the web that expands upon and ties different datasets together. Hackers are to examine publicly available data sets, determine their data taxonomies, make recommendations for improvements, develop tools for interoperability, fill in data gaps with modeling, record metadata, establish clear data provenance for databases, use projection analysis to determine collective action.
- September 21st @ 9:00-9:30 AM ET | Launch Datathon @ NY Climate Week
- October 15th | Submissions closed
- October 15th-20th | Submissions reviewed
- October 21st | Winners identified and announced
- November 10th @ COP27 | Winners will present on Science Day
Submissions will be evaluated based on their quality, clarity, credibility, and potential impact across these dimensions:
- Help data sets talk to one another for aggregated insights to enable insights on collective progress
- Develop new insights and analyses about our collective progress toward the Paris Agreement
- Help to visualize our collective progress toward the Paris Agreements
- Data storytelling and narration
- Other data innovation
All submissions should be made via the GitHub repository. Included with all submissions should be a visual presentation that describes your work. Some examples of the form this visual presentation can take is: a slide deck, video, written description, or website. To submit your project, please
- Fork a copy of the datathon repository
- Navigate to the folder of the prompt you wish to provide a submission for
- Create a subfolder under the Submissions/<Prompt Theme>/<Prompt folder> with the format “TeamName_PromptTitle”
- Include/Embed relevant documentation and/or presentations in the README file
- Include any relevant code snippets and/or datasets
- Submit a pull request to the repository
If you have significant amounts of code (for e.g., if you have built a web tool/app) or if you have created a GitHub page for your submission, you can also make use of GitHub’s Submodule system to include a submodule to your own GitHub repository within your submission.
Join the datathon Discord channel here!
If you are participating in the datathon individually, check out the Discord Team Matching channel to find your dream team!
** Crowd-sourced and compiled datasets** Crowd-sourced datasets via our google form are available on this Google Drive folder
DDL-OEF-CAD 2.0 Data model
The DDL-OEF-CAD 2.0 Data model is an attempt led by DDL and OEF (and supported by the CAD 2.0 community) to create a data model that would allow climate data from disparate sources to be harmonized for use in a Digitally-enabled Independent Global Stocktake.
Participants are encouraged to transform their data into the format of the data model (accessible here when submitting any datasets, and to leave any comments they have on the data model as an issue within the same GitHub Repository linked above.
R
Spatial Data Programming with R
Python
Python for Data Science Cheatsheet
GitHub