Learning about AWS sustainability datasets at “Code Green” workshop and hackathon

Participants and volunteers at the re:Invent 2019 Code Green workshop and hackathon.

At the 2019 re:Invent conference in Las Vegas, Amazon Sustainability and the Amazon Sustainability Data Initiative (ASDI) hosted a four-hour workshop and hackathon to showcase ASDI’s collection of sustainability-related datasets and new ways to put those datasets to use. Called “Code Green,” the event also introduced conference attendees and participants to geospatial weather and climate data on Amazon Web Services (AWS).

Participants walked through the creation of an app designed to select the most sustainable location for an event. In this workshop, the hypothetical example was an AWS DeepRacer autonomous car race. To find the most sustainable location for the event, participants queried the NOAA GHCN-D dataset for predicted temperature in the possible locations to select the location with the lowest heating and cooling costs.

Screenshot of the functional app created at the Code Green hackathon.

The workshop created an air temperature search API built using Amazon API Gateway with an endpoint backed by AWS Lambda. The AWS Lambda queries GHCN-D using Amazon Athena, then formats and returns the results. An Amazon Simple Storage Service (Amazon S3)-hosted website is then used to call the air temperature search API.

The workshop and related code is available on GitHub.

The hackathon happened concurrently with the workshop, which allowed participants to work in teams of one to five people, with the only requirement being that entries use ASDI data for their solutions.

Participants were challenged to find ways to use environmental data to improve the lives of those around them. Their solutions could solve anything from predicting when the fish would be biting to finding the largest collections of plastic waste in our waters for removal. Entries were judged on applicability to sustainability, novel use of data, technical excellence, and presentation quality.

Eight entries were submitted. The winning team created an app visualizing the intensity of PM2.5 air pollution using OpenAQ data. You can view their solution on GitHub. The app and the OpenAQ data are valuable tools for cities to track how the air pollution is changing in their area, helping determine if the efforts to improve air quality are working. The first-place prize team won free tickets to AWS re:Invent 2020 for the entire team.

A number of other entries were awarded AWS Promotional Credits, including projects visualizing deforestation in Brazil, forecasting cloud cover and wind speed, and charting how large events affect air quality.

If you have ideas on future Code Green events, please comment or email us to let us know. The material for both the hackathon and workshop are available on the ASDI GitHub repo.

If you’re interested in the attending the workshop, the event will be held again at the AWS Loft in San Francisco on January 28th and at the AWS Loft in New York City on February 14th.