September 3, 2021
The Spotlight Effect
Methodology report: Using satellite imagery to estimate the utilisation rate of steel facilities.
We need more transparency in steel
Considerable efforts have been made to improve data transparency in the steel sector. Despite this, the industry is currently unwilling to provide facility level production and emissions data on a consistent basis. If information precedes action, then this data needs to be made public as soon as possible. To improve data transparency, we developed a methodology to monitor steel production and emissions from blast furnace (BF) and basic oxygen furnace (BOF) routes.
Whitepaper goals
- Detail our methodology to estimate steel production at the facility level in near real time;
- Present the provisional results of this methodology at country and facility level; and
- Illustrate how this data can be used to align the steel sector with a zero carbon world.
Based on a first iteration of the countries modelled, our satellite-derived estimates currently cover 65% of operating BF/BOF capacity, which corresponds to 35% of installed global steel capacity. Our next iteration in Q4 2021, will expand coverage to the remaining BF/BOF facilities. At the time of writing, when compared to monthly national statistics from the World Steel Association, our models achieved a mean average error (MAE) of 18%, with the MAE ranging from 9% to 28% depending on the country.
To showcase our methodology, we provide August 2021 crude steel production estimates and find China will not meet its policy objective of reducing 2021 production below 2020 levels without a top-down crackdown in Central and Western provinces. While the use cases in this whitepaper are strictly illustrative in nature, our production estimates will form the basis of our first use case: facility level production cost curves for economic and financial scenario analysis.