Skoltech is an international graduate research-focused university that was founded by the group of world-renowned scientists in 2011. Skoltech's curriculum focuses on technology and innovation, offering Master's programs in 11 technological disciplines. Students receive rigorous theoretical and practical training, design their own research projects, participate in internships and gain entrepreneurial skills in English. The faculty is comprised of current researchers with international accreditation and achievements.

Sber and Skoltech to create carbon balance estimation software

As part of their collaboration, Sber and the Skolkovo Institute of Science and Technology (Skoltech) will develop software for carbon balance estimation based on Earth remote sensing data, enabling integrated analysis for effective decarbonization decisions and carbon footprint reduction strategies. Its artificial intelligence elements will make it possible to obtain more accurate results and expedite analytics over large areas.

Sber is an industrial partner of the Skoltech-based Center for Artificial Intelligence Technology established at the end of 2021 using a federal grant. The center aims to develop both fundamental AI technologies and AI applications for ESG risk monitoring, prediction, and optimization, and addressing decarbonization issues.

In 2021, Skoltech and its partners, the Space Research Institute of RAS, Shirshov Institute of Oceanology of RAS, and Yu.A. Izrael Institute of Global Climate and Ecology, collected a set of the most recent data on the sources and absorbers of carbon-containing greenhouse gases and developed a software concept as the first step in developing and adapting AI methods to carbon budget tasks. Also, the team made a list of regions for pilot-testing the software.

Recognizing that high quality of training datasets is essential for AI algorithms to be accurate and reproducible, the team collected remote sensing data on forests covering a total of over 6.5 million hectares and statistics from 2010 to 2020 for the pilot regions.

The data helped assess carbon reserves and flows in forests using full-carbon-cycle biochemical process models. The researchers developed hybrid approaches based on carbon cycle modeling to evaluate the carbon balance in forests, and used deep neural networks to analyze remote sensing data.

The team expects to complete software development and testing by the end of 2024.

The joint project was coordinated by Sber’s Research and Innovations Department. 

Alexander Vedyakhin, first deputy chairman of the executive board at Sber:

“Achieving carbon neutrality is a priority for the entire world and not only Russia and Sber. We should hold back climate change before it causes irreversible damage. To monitor the progress, we need appropriate software, which we are developing jointly with Skoltech. I am sure that AI will help make our tool as accurate and task-relevant as possible. We all know that you can’t effectively manage a process unless you measure it accurately.”

Alexander Kuleshov, the president of Skoltech:

“With the global economy heading toward the green agenda, our country could end up in a deep crisis if we do not learn to at least calculate our carbon balances using methods recognized by the international scientific community. We have all we need in Russia for this and we should now prove by theory and experiment that these methods work. I am sure that we will cope with this challenge together with Sber.”

Contact information:
Skoltech Communications
+7 (495) 280 14 81

Share on VK