DOI: 10.25923/ee1p-zw53

S. P. Wise¹, L. A. K. Barnett², S. Gonzalez^3,4, I. Ortiz⁵, K. Rand^6,7, A. Spear², and J. T. Thorson⁷

¹Economic and Social Science Research Program, Alaska Fisheries Science Center, NOAA, Seattle, WA, USA
²Resource Assessment and Conservation Engineering Division, Alaska Fisheries Science Center, NOAA, Seattle, WA, USA
³Climate and Oceanography, Institute of Marine Research, Bergen, Norway
⁴Bjerknes Centre for Climate Research, Bergen, Norway
⁵Cooperative Institute for Climate, Ocean, and Ecosystem Studies, University of Washington, Seattle, WA, USA
⁶Lynker, Leesburg, VA, USA
⁷Resource Ecology and Fisheries Management, Alaska Fisheries Science Center, NOAA, Seattle, WA, USA

Headlines

• Warming bottom waters, declining sea ice, and rising chlorophyll levels in the Chukchi and northern Bering Seas are restructuring Arctic marine ecosystems.
• Recent shifts in key mid-water-dwelling species reflect changing sea ice, prey dynamics, and the movement of species in the Chukchi and northern Bering Seas.
• Changes in Arctic bottom-dwelling species point to emerging borealization of seafloor ecosystems, with some southern species appearing farther north as conditions warm.
• Chukchi and northern Bering Seas ecosystem shifts—driven by warming, sea ice loss, and species redistribution—affect North Pacific commercial fisheries, Arctic food security, and Indigenous subsistence practices.

Introduction

Arctic marine ecosystems have historically relied on strong biological coupling between the surface/midwater (pelagic) and seafloor (benthic), driven by early spring phytoplankton blooms that export carbon to the seafloor. However, climate change may disrupt this linkage, causing ecosystem change. Ocean warming and sea ice loss can increase mixing across the water column and delay phytoplankton blooms relative to the seasonal timing of sea ice retreat (Kikuchi et al. 2020; Nielsen et al. 2024), which can shift more production to pelagic consumption, reducing food for benthic species on which seabirds, walruses, and other animals rely. Additionally, boreal species may move northward with ocean warming, a process known as borealization, with potential influence on Arctic species.

This research synthesizes 21 years of data to understand how warming oceans and disappearing sea ice are reshaping ecosystems in the northern Bering and Chukchi Seas (NBS-CS). The project, Arctic Integrated Ecosystem Research Program Synthesis, funded by the North Pacific Research Board, is a collaboration between the NOAA Arctic Research Program and NOAA Fisheries. Integrated computer modeling of boreal and Arctic taxa (species groups) is advancing, linking ecological monitoring and remote-sensing data to produce robust assessments of species trends under accelerating climate shifts.

To examine possible ecosystem shifts, we synthesized data in the NBS-CS (Fig. 1) from 2002 to 2022 and developed models to deduce trends. Drawing from 7,560 samples, we selected 27 species groups important for people, ecosystems, and fisheries to model (Table 1). These taxa include commercial species—Pacific cod (Gadus macrocephalus), walleye pollock (Gadus chalcogrammus), yellowfin sole (Limanda aspera), and snow crab (Chionoecetes opilio)—and taxa serving as prey for commercial species, marine mammals, seabirds, and humans, including Indigenous subsistence hunters and fishers.

Environmental drivers

In the Chukchi Sea, bottom temperatures in September show no long-term trends due to high variation among years, but temperatures increased from 2013 to 2019; whereas in the northern Bering Sea, temperatures decreased into the late 2000s and early 2010s before increasing (Fig. 2). Spring sea ice declined in both regions during the study period.

Size-categorized chlorophyll concentrations (indicating plant-like life in the water) increased in the Chukchi Sea and northern Bering Sea from 2013 to 2018 for both large and small sizes (Fig. 2). In the northern Bering Sea, both size classes also increased from 2006 to 2012 before transitioning to a minimum in 2013. These trends were generally consistent with previous observations (Docquier et al. 2024; Frey et al. 2021; Hu et al. 2024).

Pelagic (surface and mid-water) trends

In the Chukchi Sea, most pelagic taxa showed no change in estimated population size (Fig. 3); however, Pseudocalanus copepods declined from 2010 to 2012 before increasing across 2013 to 2019, and Calanus marshallae/glacialis increased from 2010 to 2016 before declining. These trends are likely due to reduced sea ice and increased northward water transport (Spear and Kimmel, personal communication). Pink salmon (Oncorhynchus gorbuscha) and sockeye salmon (Oncorhynchus nerka) showed unusual recent high years, which likely reflects population increases related to shifting distributions.

In the northern Bering Sea, patterns were mixed (Fig. 3): Pseudocalanus increased, while northern sea nettles (Chrysaora melanaster) increased slightly overall. Pacific herring (Clupea pallasii), pink, and sockeye salmon were largely unchanged. Capelin (Mallotus villosus) and Calanus marshallae/glacialis had one anomalous high year. The data offer little support that pelagic species increased in response to pelagic-benthic decoupling.

Benthic (sea bottom) trends

In the Chukchi Sea, time series for the benthic infauna (sediment dwelling species) show an increase only in bivalvia (Fig. 4). Typical Arctic epibenthic taxa, species living on or just above the seafloor, had mixed responses (Figs. 4 and 5), and also sparse data in many cases. Chionoecetes, including snow crab, decreased over time, while Arctic cod (Boreogadus saida) were unchanged, as was the saffron cod (Eleginus gracilis) which resides in both Arctic and boreal waters (Arctic-boreal). Of the epibenthic and boreal species, pollock increased while yellowfin sole and Pacific cod were unchanged, as was the Arctic-boreal Alaska plaice (Pleuronectes quadrituberculatus). Epibenthic invertebrates, sea cucumbers, sea squirts, and brittle stars showed no trend, while sea urchins slightly decreased.

Among infaunal taxa in the northern Bering Sea, bivalvia and polychaetes showed no overall population trend despite a slight initial decrease for bivalves (Fig. 4). Crustaceans trended downward, with 2009 as an anomalously high year. This is consistent with documentation of decreasing infaunal biomass between 1990 and 2003 (Grebmeier et al. 2006) and amphipods from 1983 to 2000 (Moore et al. 2003). Other epibenthic invertebrates showed no change. Epibenthic fishes were sampled in more years in the northern Bering Sea than the Chukchi Sea (Fig. 5). Two of the Arctic and Arctic-boreal epibenthic fishes in the northern Bering Sea—Arctic cod and saffron cod—declined (although the Arctic cod pattern relies primarily on an anomalously high first survey year) while Alaska plaice was unchanged. Boreal epibenthic fishes—pollock and Pacific cod—increased into the late 2010s before declining, while yellowfin sole slightly increased.

In both regions, epibenthic changes were not consistent with pelagic-benthic decoupling, as we did not observe the reductions in benthic biomass or prey availability that would be expected under such a shift. Instead, changes in species composition marked by increasing boreal-affiliated taxa (e.g., walleye pollock, yellowfin sole) and declining Arctic-boreal and Arctic-affiliated taxa (e.g., saffron cod, snow crab) provide some evidence for borealization.

Borealization

In the Chukchi Sea, one of five typically Arctic taxa time series declined (snow crab). Two of the six boreal taxa increased: Pseudocalanus and pollock, while sockeye and pink salmon had recent anomalously high years but showed no consistent long-term trend.

In the northern Bering Sea, the saffron cod population decreased, while Arctic cod remained low after an initial high year. Calanus marshallae/glaciallis showed no trend aside from a recent high year. Two of the six boreal taxa increased: Pseudocalanus and yellowfin sole, while pollock and Pacific cod initially increased into the warm late 2010s and then declined as temperatures moderated.

In summary, approximately one-third of the boreal taxa examined in NBS-CS increased over the time series and one-third experienced recent anomalous highs, and one-third of Arctic taxa declined, which is consistent with borealization. Warming temperatures, declining sea ice, and shifting productivity in the Chukchi and northern Bering Seas drive ecosystem changes with significant implications for fisheries, food security, and Indigenous subsistence.

Methods and data

Data were from surveys and models tracking changes in the Arctic marine environment. Most benthic species data came from the Distributed Biological Observatory (DBO). Plankton and fish data came from the Ecosystem and Fisheries Oceanographic Coordinated Investigations (EcoFOCI) (Mordy et al. 2023). Additional data on fish, invertebrates, and crab came from the NOAA Alaska Fisheries Science Center Groundfish Assessment Program (GAP). Chlorophyll data were from DBO and EcoFOCI surveys (Eisner 2016). Zooplankton were sampled by EcoFOCI (Kimmel et al. 2023). Pelagic fish and jellyfish were sampled by EcoFOCI (Murphy et al. 2023). Infauna were sampled with a Van Veen grab (Grebmeier et al. 2018) on Pacific Marine Arctic Regional Synthesis (PacMARS) (Grebmeier and Cooper 2024) and DBO surveys. Epibenthic fish and invertebrates were sampled in the Chukchi Sea (Iken et al. 2019; Markowitz et al. 2024). Bottom water temperature data were from the Pacific Arctic Regional Ocean Modeling System (Danielson et al. 2020), based on the Regional Ocean Modeling System (Curchitser et al. 2013). Spring sea ice concentration was calculated from the National Snow and Ice Data Center (NSIDC) monthly sea ice index (Fetterer et al. 2017). Trends in environmental variables and chlorophyll were defined by comparing annual values to the standard deviation of the long-term mean (Fig. 2).

Time series of zooplankton, fish, and invertebrate population size were standardized using the R-package tinyVAST (Thorson et al. 2025). In total, we fit 44 spatio-temporal generalized linear mixed models (ST-GLMM) for the 27 taxa to estimate the indices of population size (Figs. 3-5). Each ST-GLMM estimated a log-linked Tweedie distribution for biological samples while estimating an annually varying intercept as a fixed effect, a spatial latent variable using the stochastic partial differential equation (SPDE) approximation (Lindgren et al. 2011), and a spatio-temporal variable that follows a random walk over time. After fitting the ST-GLMM, we calculated the epsilon-corrected, area-expanded population size index over a given region (Thorson and Kristensen 2016). Strong trends in estimated population size indices were identified when confidence intervals (2 standard deviations, approximating a 95% CI) had little to no overlap among years.

Acknowledgments

Funding by the North Pacific Research Board through its Arctic Integrated Ecosystem Research Program created the unique opportunity to bring together diverse datasets, perspectives, and disciplines. This work was made possible by a multidisciplinary team and by the vision and sustained commitment of Dr. Elizabeth A. Logerwell.

References

Curchitser, E. N., H. P. Batchelder, D. B. Haidvogel, J. Fiechter, and J. Runge, 2013: Advances in physical, biological, and coupled ocean models during the US GLOBEC program. Oceanography, 26(4), 52-67, https://doi.org/10.5670/oceanog.2013.75.

Danielson, S. L., and Coauthors, 2020: Oceanic routing of wind-sourced energy along the Arctic continental shelves. Front. Mar. Sci., 7, 509, https://doi.org/10.3389/fmars.2020.00509.

Docquier, D., F. Massonnet, F. Ragone, A. Sticker, T. Fichefet, and S. Vannitsem, 2024: Drivers of summer Arctic sea-ice extent at interannual time scale in CMIP6 large ensembles revealed by information flow. Sci. Rep., 14, 24236, https://doi.org/10.1038/s41598-024-76056-y.

Eisner, L., 2016: The Bering Sea: Current status and recent trends. PICES press, 24(1), 51-54.

Fetterer, F., K. Knowles, W. N. Meier, M. Savoie, A. K. Windnagel, and T. Stafford, 2025: Sea Ice Index, G02135, Version 4. NSIDC: National Snow and Ice Data Center, Boulder, CO, USA, accessed 2 October 2025, https://doi.org/10.7265/a98x-0f50.

Frey, K. E., J. C. Comiso, L. W. Cooper, J. M. Grebmeier, and L. V. Stock, 2021: Arctic ocean primary productivity: The response of marine algae to climate warming and sea ice decline. Arctic Report Card 2021, T. A. Moon, M. L. Druckenmiller, and R. L. Thoman, Eds., https://doi.org/10.25923/kxhb-dw16.

Grebmeier, J. M., and L. W. Cooper, 2024: Benthic observations and ecosystem connections in the Pacific Arctic over temporal and spatial scales. AGU Fall Meeting Abstracts, Vol. 2024, OS21D-0626, doi: 10.1002/essoar.10515159.1, https://agu.confex.com/agu/agu24/meetingapp.cgi/Paper/1686070.

Grebmeier, J. M., and Coauthors, 2006: A major ecosystem shift in the northern Bering Sea. Science, 311(5766), 1461-1464, https://doi.org/10.1126/science.1121365.

Grebmeier, J. M., K. E. Frey, L. W. Cooper, and M. Kędra, 2018: Trends in benthic macrofaunal populations, seasonal sea ice persistence, and bottom water temperatures in the Bering Strait region. Oceanography, 31(2), 136-151, https://doi.org/10.5670/oceanog.2018.224.

Hu, Z.-Z., M. J. McPhaden, B. Huang, J. Zhu, and Y. Liu, 2024: Accelerated warming in the North Pacific since 2013. Nat. Climate Change, 14(9), 929-931, https://doi.org/10.1038/s41558-024-02088-x.

Iken, K., F. Mueter, J. M. Grebmeier, L. W. Cooper, S. L. Danielson, and B. A. Bluhm, 2019: Developing an observational design for epibenthos and fish assemblages in the Chukchi Sea. Deep Sea Res. II: Top. Stud. Oceanogr., 162, 180-190, https://doi.org/10.1016/j.dsr2.2018.11.005.

Kikuchi, G., H. Abe, T. Hirawake, and M. Sampei, 2020: Distinctive spring phytoplankton bloom in the Bering Strait in 2018: A year of historically minimum sea ice extent. Deep Sea Res. II: Top. Stud. Oceanogr., 181-182, 104905, https://doi.org/10.1016/j.dsr2.2020.104905.

Kimmel, D. G., L. B. Eisner, and A. I. Pinchuk, 2023: The northern Bering Sea zooplankton community response to variability in sea ice: Evidence from a series of warm and cold periods. Mar. Ecol. Prog. Ser., 705, 21-42, https://doi.org/10.3354/meps14237.

Lindgren, F., H. Rue, and J. Lindström, 2011: An explicit link between Gaussian fields and Gaussian Markov random fields: the stochastic partial differential equation approach. J. Roy. Stat. Soc. B, 73(4), 423-498, https://doi.org/10.1111/j.1467-9868.2011.00777.x.

Markowitz, E. H., E. J. Dawson, S. N. Wassermann, C. B. Anderson, S. K. Rohan, N. E. Charriere, and D. E. Stevenson, 2024: Results of the 2023 eastern and northern Bering Sea continental shelf bottom trawl survey of groundfish and invertebrate fauna. NOAA Tech. Memo. NMFS-AFSC-487, U.S. Dept. of Commerce, 242 pp., https://doi.org/10.25923/2mry-yx09.

Moore, S. E., J. M. Grebmeier, and J. R. Davies, 2003: Gray whale distribution relative to forage habitat in the northern Bering Sea: current conditions and retrospective summary. Can. J. Zool., 81(4), 734-742, https://doi.org/10.1139/z03-043.

Mordy, C. W., and Coauthors, 2023: Progress of Fisheries-Oceanography Coordinated Investigations in the Gulf of Alaska and Aleutian Passes. Oceanography, 36(2-3), 94-100, https://doi.org/10.5670/oceanog.2023.218.

Murphy, J., and Coauthors, 2023: Northern Bering Sea ecosystem and surface trawl cruise report, 2021: NOAA Tech. Memo. NMFS-AFSC-479, U.S. Dept. of Commerce, 136 pp., https://doi.org/10.25923/mbyq-xc41.

Nielsen, J. M., and Coauthors, 2024: Spring phytoplankton bloom phenology during recent climate warming on the Bering Sea shelf. Prog. Oceanogr., 220, 103176, https://doi.org/10.1016/j.pocean.2023.103176.

Thorson, J. T., K. Y. Aydin, M. L. H. Cheng, B. S. Dias, D. G. Kimmel, and K. Kristensen, 2025: Bottom-up interactions in state-space age-structured models using mass-balance dynamics. Fish Fish., 26(6), 1048-1064, https://doi.org/10.1111/faf.70016.

Thorson, J. T., and K. Kristensen, 2016: Implementing a generic method for bias correction in statistical models using random effects, with spatial and population dynamics examples. Fish. Res., 175, 66-74, https://doi.org/10.1016/j.fishres.2015.11.016.

December 4, 2025