An official website of the United States government
Interannual variations in marine net primary production (NPP) contribute to the variability of available living marine resources, as well as influence critical carbon cycle processes. Here we provide a global overview of near‐term (1 to 10 years) potential predictability of marine NPP using a novel set of initialized retrospective decadal forecasts from an Earth System Model. Interannual variations in marine NPP are potentially predictable in many areas of the ocean 1 to 3 years in advance, from temperate waters to the tropics, showing a substantial improvement over a simple persistence forecast. However, some regions, such as the subpolar Southern Ocean, show low potential predictability. We analyze how bottom‐up drivers of marine NPP (nutrients, light, and temperature) contribute to its predictability. Regions where NPP is primarily driven by the physical supply of nutrients (e.g., subtropics) retain higher potential predictability than high‐latitude regions where NPP is controlled by light and/or temperature (e.g., the Southern Ocean). We further examine NPP predictability in the world's Large Marine Ecosystems. With a few exceptions, we show that initialized forecasts improve potential predictability of NPP in Large Marine Ecosystems over a persistence forecast and may aid to manage living marine resources.
more »
« less
This content will become publicly available on March 27, 2026
Quantifying the Potential Predictability of Arctic Marine Primary Production
Abstract Phytoplankton in the Arctic Ocean and sub‐Arctic seas support a rich marine food web that sustains Indigenous communities as well as some of the world's largest fisheries. As sea ice retreat leads to further expansion of these fisheries, there is growing need for predictions of phytoplankton net primary production (NPP), which will likely allow better management of food resources in the region. Here, we use perfect model simulations of the Community Earth System Model version 2 (CESM2) to quantify short‐term (month to 2 years) predictability of Arctic Ocean NPP. Our results indicate that NPP is potentially predictable during the most productive summer months for at least 2 years, largely due to the highly predictable Arctic shelves where fisheries in the Arctic are projected to expand. Sea surface temperatures, which are an important limitation on phytoplankton growth and also are predictable for multiple years, are the most important physical driver of this predictability. Finally, we find that the predictability of NPP in the 2030s is enhanced relative to the 2010s, indicating that the utility of these predictions may increase in the near future. This work indicates that operational forecasts using Earth system models may provide moderately skillful predictions of NPP in the Arctic, possibly aiding in the management of Arctic marine resources.
more »
« less
- Award ID(s):
- 1752724
- PAR ID:
- 10579847
- Publisher / Repository:
- DOI PREFIX: 10.1029
- Date Published:
- Journal Name:
- Journal of Geophysical Research: Oceans
- Volume:
- 130
- Issue:
- 4
- ISSN:
- 2169-9275
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Predictable internal climate variability on decadal timescales (2–10 years) is associated with large‐scale oceanic processes, however these predictable signals may be masked by the noisy climate system. One approach to overcoming this problem is investigating state‐dependent predictability—how differences in prediction skill depend on the initial state of the system. We present a machine learning approach to identify state‐dependent predictability on decadal timescales in the Community Earth System Model version 2 pre‐industrial control simulation by incorporating uncertainty estimates into a regression neural network. We leverage the network's prediction of uncertainty to examine state dependent predictability in sea surface temperatures by focusing on predictions with the lowest uncertainty outputs. In particular, we study two regions of the global ocean—the North Atlantic and North Pacific—and find that skillful initial states identified by the neural network correspond to particular phases of Atlantic multi‐decadal variability and the interdecadal Pacific oscillation.more » « less
-
Abstract We show that explainable neural networks can identify regions of oceanic variability that contribute predictability on decadal timescales in a fully coupled Earth‐system model. The neural networks learn to use sea‐surface temperature anomalies to predict future continental surface temperature anomalies. We then use a neural‐network explainability method called layerwise relevance propagation to infer which oceanic patterns lead to accurate predictions made by the neural networks. In particular, regions within the North Atlantic Ocean and North Pacific Ocean lend the most predictability for surface temperature across continental North America. We apply the proposed methodology to decadal variability, although the concept is generalizable to other timescales of predictability. Furthermore, while our approach focuses on predictable patterns of internal variability within climate models, it should be generalizable to observational data as well. Our study contributes to the growing evidence that explainable neural networks are important tools for advancing geoscientific knowledge.more » « less
-
Abstract We use a modern Earth system model to approximate the relative importance of ice versus temperature on Arctic marine ecosystem dynamics. We show that while the model adequately simulates ice volume, water temperature, air‐sea CO2flux, and annual primary production in the Arctic, itunderestimates upper water column nitrate across the region. This nitrate bias is likely responsible for the apparent underestimation of ice algae production. Despite this shortcoming, the model appears to be a useful tool for exploring the impacts of environmental change on phytoplankton production and carbon dynamics over the Arctic Ocean. Our experiments indicate that under a warmer climate scenario, the percentage of ocean warming that could be apportioned to a reduction in ice area ranged from 11% to 100%, while decreasing ice area could account for 22–100% of the increase in annual ocean primary production. The change to CO2air‐sea flux in response to ice and temperature changes averaged an Arctic‐wide 5.5 Tg C yr−1(3.5%) increase, into the ocean. This increased carbon sink may be short‐lived, as ice cover continues to decrease and the ocean warms. The change in carbon fixation from phytoplankton in response to increased temperatures and reduced ice was generally more than a magnitude larger than the changes to CO2flux, highlighting the importance of fully considering changes to the marine ecosystem when assessing Arctic carbon cycle dynamics. Our work demonstrates the importance of ice dynamics in controlling ocean warming and production and thus the need for well‐behaved ice and BGC models within Earth system models if we hope to accurately predict Arctic changes.more » « less
-
Phytoplankton, a diverse group of small photosynthetic algae inhabiting the sunlit region near the ocean surface, form the base of marine trophic webs ( 1 ). Whereas phytoplankton have evolved in tandem with the climate system for hundreds of millions of years ( 2 ), cumulative greenhouse gas emissions are causing rising ocean temperature, acidification, and oxygen loss at increasing rates ( 3 ). How phytoplankton will respond and adapt to these multistressors in the future ( 4 ), and how this will in turn threaten marine trophic webs and food supply for humans, remain important questions in oceanography. On page 1487 of this issue, Holm et al. ( 5 ) show that a physiological adaptation of phytoplankton to a warming world will lead to a reduction in their nutritional value, with negative consequences for marine ecosystems.more » « less
