Sizing the Stack: Predicting Stack Heights Using Satellite Imagery

Duke MIDS logo
: WattTime
: Energy, NGO
: 2022

Our aim with this project was to estimate the heights of power plant stacks from publicly available satellite imagery data. Our client, WattTime, is working to build the first automated system for monitoring the emission from power plants globally, utilizing the height of flue-gas stacks to better understand the dispersal of pollutants as well as the contribution of power plants to climate change. We estimated the length of the shadow cast by the flue stack and use physical information to obtain the estimations of stack heights. We have applied different methods to detect the region of shadow pixels, among which change point detection is the chosen method with the best performance. We’ve developed the pipeline of change point detection to predict stack heights and applied the pipeline to create a database with over 16,000 global stack heights.

Additional team members:
Yijia Zhang in theĀ Department of Statistical Science