How I built a free carbon credit risk model using NASA satellite data, dbt, and Streamlit
Likitha Sree Yarabarla spent three weeks building an open-source carbon credit risk model that combines NASA FIRMS satellite fire alerts, World Bank governance indicators, and carbon project data from Verra and Berkeley. The pipeline uses dbt for data transformation and Streamlit for a live dashboard. Key technical challenges included NASA FIRMS requiring a MAP_KEY instead of a bearer token, a changed endpoint for World Bank WGI indicators, and merging Verra numeric IDs with Berkeley's VCS1234 format via string parsing. The live dashboard is hosted at https://likitha-blue-carbon.streamlit.app/ and the full code is available on GitHub at https://github.com/likitha-sree-data/dbt-carbon-mrv. This project provides a free, transparent alternative to proprietary carbon credit risk models, enabling developers and analysts to assess project risks without costly subscriptions.
Enables free, transparent carbon credit risk assessment using open data and tools.