Curated, tractable and impactful data projects for researchers, students and data-enthusiasts
Emissions from above
Leverage satellite technology together with machine learning to estimate NOx emissions from large polluters worldwide, where no other emission data exists or the data is unreliable.
Anomaly detection
Help us detect calibration issues and malicious tampering in air quality data. Ensuring the integrity of air quality data is paramount for enforcing regulations and protecting public health.
CREA data challenges are data challenges open to public participation to help us extend our impact in the energy transition and in revealing the trends, causes, and health impacts, as well as the solutions, to air pollution.
We have curated a list of data problems that we think are both tractable and whose solutions have the potential to make a significant impact. For each of these challenges, we have prepared datasets to help you dive in immediately.
Who can participate?
Whether you are an academic researcher, a data scientist, a student, or simply someone passionate about data, we invite you to join us in these data challenges. Your skills, creativity, and dedication can help us with our work influencing policy change capable of saving thousands of lives each year and creating a more sustainable future.
Why participate?
- Seek impact: These challenges represent a unique opportunity to do good. We are curating projects that can have a tremendous impact on public health and/or climate change. Whether you’re looking for a research publication idea, a thesis topic or simply an opportunity to hone new data skills, these challenges offer a chance to make a tangible impact.
- Skip the tedious part: We also provide you with a structured dataset, so you can skip the tedious part and focus on the most interesting bit straight away.
- Open new opportunities: Unfortunately, we cannot compensate every participant. However, we regularly hire data scientists or fund external researchers. Contributing meaningfully to one of these challenges would automatically put you on the shortlist when we hire data scientists or researchers.
How does it work?
For each project, we define the main objective and, when available, references that have partially tackled the problem. We also provide you with a supporting dataset, so that you can immediately dive into the problem.
Our only ask: keep it open
We are only asking that you keep your project open source and keep us informed about your progress.
You can work by yourself, with friends and colleagues, set up an open source project, or work directly with us, at your own pace. To get started, simply select one of the projects below.