How can we use mathematics and machine learning to detect carbon dioxide emissions that deviate from normal (anomalies) at a global scale and near to real time?

Laia is a mathematician that works at the Finnish Meteorological Institute since 2021. She studied mathematics at the University of Barcelona and completed her PhD (2016) in number theory in a joint collaboration between the University of Barcelona and the University of Luxembourg. In 2018, she moved to Finland, where she worked as a postdoctoral researcher at Aalto University in different fields, including wireless communications, post-quantum cryptography, and applications of machine learning.

Laia’s research focuses on studying greenhouse gas emissions by using satellite data and techniques ranging from mathematics to machine learning and data analysis.

Research objectives

• Develop data-driven methods to identify spatial and temporal anomalies in satellite-based CO₂ observations.
• Implement unsupervised machine learning techniques to cluster anomalies, differentiating sources such as anthropogenic from biogenic origins.
• Offer a scalable and adaptable model that can be further enhanced as newer satellite missions, most importantly Copernicus Anthropogenic CO₂ Monitoring Mission (CO2M), join the effort.