Award: OCE-2147809

The annual transfer of biologically produced carbon from the ocean surface into the interior is often referred to as "the biological pump" or "biological carbon export", which has a magnitude comparable to the modern fossil fuel emissions rate (~10 petagrams of carbon per year), but with large uncertainties. In the North Atlantic Ocean, biological carbon production is evidenced by a massive spring phytoplankton bloom, one of the largest and most rapid greening events on Earth that is observed by satellites. However, some studies have found that this ostentatious chlorophyll bloom is not well correlated to the annual biological pump strength. In fact, many chemistry-based approaches identify similar magnitudes of biological carbon export in the subpolar and subtropical regions of the North Atlantic Ocean, though the latter exhibits almost no seasonal surface "greening". Determining what drives these conflicting results by accurately quantifying where, when, and how much biological productivity and subsequent carbon export is occurring will be required to improve our understanding of marine ecosystems, global carbon cycle processes, species distributions, and fishery management. 

 

In this project, we refined and tested commonly applied methods for estimating net primary production (NPP) and net community production (NCP), key metrics of biological pump strength and efficiency, using the growing array of profiling floats in the North Atlantic Ocean that carry biological and chemical sensors. For the NPP analysis, we first developed a novel method for translating float fluorescence observations into chlorophyll estimates by matching float and satellite observations in space and time to create profile-specific chlorophyll correction factors. Applying the correction factors to float fluorescence observations resulted in chlorophyll estimates that were consistent with satellite chlorophyll values. We then used the float and satellite chlorophyll information to estimate NPP from a widely-used model and evaluated how sensitive the NPP estimates were to changes in the data that were input into this model.  We found that NPP estimates were most sensitive to the fluorescence correction and the use of depth-resolved float data, indicating that the combined use of satellite data (to adjust float fluorescence values) and float data (to capture subsurface variability in chlorophyll) may be a valuable way to improve upper ocean productivity estimates (Buzby et al., in review). The results were particularly striking in the subtropical ocean where differences between NPP estimates derived using the existing, static float fluorescence correction versus our new profile-specific correction changed sign throughout the year. We also found that satellite-based NPP estimates seem to miss a significant fraction of deep productivity identified by floats in waters below the view of satellites. 

 

Net community production, which is NPP plus the respiration of non-primary producers, determines the amount of biological carbon available for export from the upper ocean. Scientists often apply a budget framework to explain changes in a chemical property over time, with the residual, unexplained changes interpreted as the net biological signal - NCP. Due to observational limitations, this approach is usually one-dimensional and, thus, does not account for horizontal processes that may influence the chemical property. This is a known methodological weakness, particularly for profiling floats that can drift across large-scale horizontal gradients in ocean properties - changes that would be interpreted as part of the NCP signal. To address this issue, we developed a method that leverages gap-filled, global ocean data products to quantify chemical property changes caused by a float's movement over time (Cornec and Fassbender, 2025). Applying the correction method improved NCP estimate precision by 50-80% and accuracy by 10-100%, depending on location in the global ocean, with the largest improvements found in the Southern Ocean. After applying this correction to NCP estimates in the North Atlantic Ocean, we found that NCP in the deeper waters of the nutrient-poor subtropics rivals the magnitude of NCP in the upper waters of the nutrient-rich subpolar region. Elevated subtropical NCP occurs despite NPP that is much lower than what is found in the subpolar region, indicating a significantly more efficient transfer of energy between trophic levels in the subtropics (Cornec et al., in prep). These findings emphasize the importance of persistent, subsurface observations for uncovering signals that are not visible from space, and add context to long-held debates about the trophic status of the subtropical oceans.

 

This project supported one graduate student and one postdoctoral scholar who contributed to/co-lead development/translation of the OneArgo-MAT (https://zenodo.org/records/15264329) and OneArgo-R (https://zenodo.org/records/6604735) software toolboxes, respectively. These tools have significantly reduced barriers to Argo float data access. The methods developed as part of this project are broadly applicable and will help to advance understanding of the biological pump and the global carbon cycle.

 

Last Modified: 12/05/2025
Modified by: Andrea J Fassbender