Environmental Insights Explorer Analytics Prediction Competition

Environmental Insights Explorer Analytics Prediction Competition

Develop a methodology to calculate an average historical emissions factor of electricity generated for a sub-national region, using remote sensing data and techniques.

The Environmental Insights Explorer team at Google is keen to gather insights on ways to improve calculations of global emissions factors for sub-national regions. The ultimate goal of this challenge is to test if calculations of emissions factors using remote sensing techniques are possible and on par with calculations of emissions factors from current methodologies.


Current emissions factors methodologies are based on time-consuming data collection and may include errors derived from a lack of access to granular datasets, inability to refresh data on a frequent basis, overly general modeling assumptions, and inaccurate reporting of emissions sources like fuel consumption. This begs the question: What if there was a different way to calculate or measure emissions factors? We’re challenging the Kaggle community to see if it’s possible to use remote sensing techniques to better model emissions factors. You will develop a methodology to calculate an average historical emissions factor for electricity generation in a sub-national region.

We’ve provided an initial list of datasets covering the geographic boundary of Puerto Rico to serve as the foundation for this analysis. As an island, there are fewer confounding factors from nearby areas. Puerto Rico also offers a unique fuel mix and distinctive energy system layout that should make it easier to isolate pollution attributable to power generation in the remote sensing data.

Participants will be tasked with developing a methodology to calculate an average annual historical emissions factor for the sub-national region. Participants will also be asked to provide an explanation of the conditions that would result in a higher/lower emissions factor, as well as a recommendation for how the methodology could be applied to calculate the emissions factor of electricity for another geospatial area using similar techniques. Bonus points will be awarded for smaller time slices of the average historical emissions factors, such as one per month for the 12-month period, and additional bonus points will be awarded for participants that develop methodologies for calculating marginal emissions factors for the sub-national region.


  • First Prize: US $10,000
  • Second Prize: US $7,500
  • Third Prize: US $5,000
  • Fourth Prize: US $2,500

Deadline:- 24-03-2020

Take this challenge