Through the CBIOMES Scholar program, the collaboration’s Steering Committee supports former CBIOMES postdocs as they transition into junior faculty roles and begin building research groups of their own.
Reporting by Helen Hill for CBIOMES
Through the CBIOMES Scholar program, the collaboration’s Steering Committee supports former CBIOMES postdocs as they transition into junior faculty roles and begin building research groups of their own.
As part of this initiative, Bror Jönsson—now a Research Assistant Professor at the University of New Hampshire and continuing his close collaboration with the Sathyendranath Group at Plymouth Marine Laboratory (PML) in the UK—has received funding for a new CBIOMES Scholar project titled “Using Machine Learning and Lagrangian Particle Tracking to Estimate Rates in the Ocean.”
The project aims to bridge the spatial and temporal scales that influence primary production in the ocean by integrating machine learning, ocean color products, and Lagrangian particle tracking. In earlier work, Jönsson developed a framework that seeded virtual particles into existing velocity fields, attached satellite data to the particles, and calculated rates of change along their trajectories.
Building on this foundation, Jönsson will analyze each trajectory as a sparse time series using modern ML-based time series models to generate robust estimates and forecasts of fluxes and rates. These will be compared with estimates from existing CBIOMES model simulations.
To address the challenge of short de-correlation timescales between simulated drifters and the physical water masses they are assumed to follow, he will incorporate in-situ drifter data from the Global Drifter Program. A key component of the project involves exploring primary production on hourly timescales. Previous studies using geostationary satellite data revealed significant diurnal variations in production rates, though they raised more questions than answers.
Jönsson will use hourly ocean color products from the Korean GOCI and GOCI2 geostationary satellites, combined with machine learning models, to investigate how productivity rates change throughout the day. By comparing biomass at the end of one day and the beginning of the next, he also aims to estimate community night respiration.
Bror Jönsson received his B.Sc. and M.Sc. in Biology from Uppsala University and his Ph.D. in Physical Oceanography from Stockholm University, both in Sweden. Before joining the faculty at UNH in 2023, he was a longtime member of Shubha Sathyendranath’s CBIOMES Group at PML.
Jönsson joins current CBIOMES Scholars Chris Follett (University of Liverpool), Sangwon Hyun (University of California, Santa Cruz), Jesse McNichol (St. Francis Xavier University, Canada), Greg Britten (Woods Hole Oceanographic Institution), and Sinikka Lennartz (University of Oldenburg, Germany).
Story image: Bror Jönsson (right) sharing his research during a CBIOMES Annual Meeting poster session – image credit: The Simons Foundation.