The project aims to identify and spot CO2 emitting fixed ground structures such as factories/power plants/ oil & gas refineries. Using publicly available global power plants data, the aim was to train the Machine Learning algorithm to identify every CO2 emitting structure and then using modelling estimate the related CO2 emissions.
After forming a global team spread out in 4 countries and three continents, Little Place team worked on the problem to (1) understand the availability and limitations of the geospatial data (2) Understand the complexity and relationship of environmental factors (3) Work on an MVP to prove the viability of data (4) Draw a theoretical roadmap on the progress that can be made in this field to address the problem.
Little Place Team primarily used the satellite data available from OCO2 Orbitor and TROPOMI and attempted to draw a heat map of CO2 across the earth.
https://docs.google.com/presentation/d/1J6Inxna5avHM6jChxIi5p-f9ee2lOebiyHZx6uxRanQ/