The BioMassters

Help conservationists make use of satellite imagery to estimate the aboveground biomass of Finnish forests. Competition hosted by MathWorks. #climate

$10,000 awarded
jan 2023
976 joined

An image of a forest in Finland

“I believe that we are just at the beginning of the Earth Observation big data revolution. Joint effort and open science are the fundamental bricks to build new models and services for addressing societal challenges.”

— Dr. Andrea Nascetti, Faculty of Sciences, University of Liège


Forests are adding and removing carbon dioxide from the air all the time. How do we know how much? The answer lies in aboveground biomass (AGBM), a widespread measure in the study of carbon release and sequestration by forests. Current approaches to measuring biomass, range from destructive sampling, which involves cutting down a representative sample of trees and measuring attributes such as the height and width of their crowns and trunks, to remote sensing methods. Remote sensing methods offer a much faster, less destructive, and more geographically expansive biomass estimate. In turn, such timely and detailed information allows landowners and policy makers to make better decisions for the conservation of forests.

The Solution

The goal of the BioMassters challenge was to estimate the yearly biomass of 2,560 meter by 2,560 meter patches of land in Finland's forests using Sentinel-1 and Sentinel-2 imagery of the same areas on a monthly basis. The ground truth for this competition came from LiDAR (Light Detection And Ranging) and in-situ measurements of the same forests collected by the Finnish Forest Centre on an annual basis.

The Results

Challenge participants made over 1,000 submissions and achieved impressive results, with the top-scoring finalist achieving an Average RMSE (lower is better) of 27.63, the second-place winner scoring 27.68, and the third-place winner scoring 28.04. These scores represent not only a substantial improvement over the benchmark model (which had a score of 101.98), but great promise for exceeding some of the most widely available open-source biomass measurements available.

To see the winners' models and write-ups, see the publicly accessible winners' repository for this competition.