Story
Fish – the missing ingredient in climate models?
26 February 2026
New research led by PML’s Dr Helen Powley reveals that when fish are included in marine ecosystem models, it fundamentally changes predictions of how much carbon the ocean can absorb: highlighting a potential overestimation of the ocean’s climate-buffering capacity.
Ocean and climate models are among the most important tools scientists use to predict how Earth’s climate will change in the future. Yet many widely used models still overlook a critical part of the marine ecosystem: the dynamic interaction between fish and plankton.
Most large-scale ocean and climate models simulate plankton and fish separately: in many cases, fish feed on plankton in the model – but plankton don’t “feel” the impact – the feedback loop is missing.
A new PML-led study demonstrates that integrating fish directly into marine ecosystem models fundamentally alters predictions of ocean carbon uptake, revealing a 7% reduction in CO2 uptake within a shelf environment
Dr Helen Powley, Modeller at PML, and lead author, said:
“The missing feedback loop matters: fish feed on plankton communities, and plankton play a central role in regulating how much carbon dioxide the ocean absorbs from the atmosphere. Without representing the real interconnected processes, models risk underestimating how marine ecosystems respond to environmental change – and how strongly the ocean can help us buffer climate warming.”
In the study, Dr Powley, alongside PML scientists Dr Rebecca Millington, Dr Yuri Artioli, Dr Robert Wilson and Dr Jorn Bruggeman of Bolding & Bruggeman, created a fully two-way coupled model – linking physics, plankton, and fish in a single system – revealing how top-down pressure from fish cascades through the food web and alters the marine carbon cycle.
Using the FABM framework, this study brings together the following models:
- PML’s internationally recognised ERSEM (European Regional Sea Ecosystem Model)
- The hydrodynamic model NEMO (Nucleus for European Modelling of the Ocean)
- A size-spectrum fish communities model MIZER (Multi-Species Size Spectrum Modelling in R)
Linking the models created a complete a more realistic system: with interlinked processes from sunlight and nutrients through plankton to fish and fisheries.
The model was applied to the North West European Shelf, one of the world’s most productive and heavily fished regions, and an important carbon sink. Switching from traditional one-way coupling to two-way feedbacks had striking consequences – when fish were allowed to feedback into the plankton system:
- The shelfs ability to absorb atmospheric CO₂ declined by 7%
- A decline in fish biomass by 20%
- Fish landings fell by more than one third
- Mesozooplankton – a key marine food web group – biomass reduced by 17%
- Carbon cycling across the shelf weakened
Dr Powley said:
“These changes were driven largely by shifts in mesozooplankton dynamics: because mesozooplankton play a central role in transferring carbon through the marine food web and into the deep ocean, altering their mortality and grazing pressure had system-wide consequences.”
“Crucially, the reduction in air–sea CO₂ uptake caused by ecological feedbacks was larger than the change caused by including terrestrial dissolved organic carbon in previous modelling studies.”
Fish are increasingly recognised as part of the carbon cycle, since fishing removes carbon and nutrients from the ocean, but the study shows that the indirect ecological effects of fish predation on plankton may be even more important than the carbon physically removed through catches.
Dr Powley added:
“Fish are not just passive responders to environmental change – they are active agents shaping the ocean’s carbon balance. As pressures from climate change and fishing intensify, understanding the full web of interactions in marine ecosystems is essential. Without two-way coupling, models may overestimate fish biomass, misrepresent plankton dynamics, miscalculate carbon sequestration and underestimate ecosystem responses to fishing and climate change.”
The new study comes as, this week, UNESCO released a new report revealing a critical lack of understanding of how the ocean absorbs and stores carbon – described as a ‘major blind spot that could undermine global climate predictions’.
‘This glaring uncertainty about our planet’s largest carbon sink threatens to skew current climate predictions, and hamper our ability to develop effective mitigation and adaptation strategies in the coming decades.’ [Source: UNESCO]
Dr Powley is presenting the research this week at the Ocean Sciences Meeting in Glasgow. Join the session, ‘Challenging one-way coupled biogeochemical to fisheries modelling: new insights from a two-way coupled model representing photons to fisheries on the North-West European Shelf’ – 11.30am 27th February, Hall 1. More information here >>
The new model code can be found here and will be used in the EU OceanICU project where it will be combined with different biogeochemical models to predict carbon cycle changes on the EU Digital Twin of the Ocean.
