EBS_Assessment
Eastern Bering Sea 2024 Ecosystem Assessment
Elizabeth Siddon
Auke Bay Laboratories, Alaska Fisheries Science Center, NOAA Fisheries
Contact:elizabeth.siddon@noaa.gov
Last updated: November 2024
Introduction
In recent decades, the eastern Bering Sea (EBS) transitioned from an ecosystem governed by interannual variability (1982–2000) into one that experienced multi-year stanzas of warm (2000–2005) and cold (2007–2013) conditions (Baker et al., 2020). In 2014, the EBS entered a warm period that was
unprecedented in terms of magnitude and duration (Figure 20) and that persisted until 2021. This recent warm period included the near-absence of sea ice in the winters of 2017/2018 and 2018/2019 (Figure 36) and subsequent lack of cold pool in summers 2018 and 2019 (Figures 41 and 42) that had distinct impacts to the ecosystems of the southeastern (SEBS) and northern (NBS) Bering Sea shelves.
Since ~2021, oceanographic metrics (e.g., sea ice extent, sea surface temperatures, and bottom tem-peratures) over the EBS shelf have cooled to near average based on respective time series. Currently, the broader North Pacific is predicted to transition from El Ni˜no to La Ni˜na conditions by spring 2025, which is expected to bring continued cooler temperatures to the EBS shelf. This Assessment aims to synthesize recent biological metrics (e.g., zooplankton and fish dynamics) to characterize the current status of the SEBS and NBS ecosystems in response to the cooler oceanographic conditions.
Seasonal sea ice and the resulting cold pool extent are defining features over the Bering Sea shelf. Combined, sea ice and the cold pool create thermal barriers, both horizontally (i.e., north/south) and vertically in the water column. Such thermal barriers affect the spatial distributions of crab and ground-fish (e.g., Thorson et al., 2019; DeFilippo et al., 2023) that have subsequent direct (e.g., habitat expansion) and indirect (e.g., changes in predator/prey dynamics) impacts to managed stocks.
The delineation between the SEBS and NBS is often considered to be at 60oN latitude on the basis of the physical and biological distinctions between these ecological systems, existing research and analyses in these areas, and available data and survey designs. This delineation is supported by broad-scale analyses of the physical oceanography and hydrography (Stabeno et al., 2012a; Baker et al., 2020) and zoogeography of the region (Sigler et al., 2017). For an in-depth review of distinguishing characteristics between these ecoregions, see Baker (2023).
This assessment documents the ecosystem response to the average oceanographic conditions experienced since ~2021, in contrast with the prolonged warm period (2014–2021) and pulse events of 2017/2018 and 2018/2019.
Southeastern Bering Sea
Since 2021, the SEBS has experienced a more neutral thermal state. Over the past year (August 2023–August 2024) many metrics, including sea surface temperature (Figure 23, Figure 30, Figure 34), wintertime sea ice areal extent (Figure 36) and thickness (Figure 39, Figure 40), and cool pool extent (Figure 41, Figure 42) have continued to be near historical averages. While the spatial extent of waters < 2°C in 2024 was only slightly (12.7%) smaller than in 2022 and 2023, the extent of ≤-1oC and ≤0°C isotherms decreased by 54.4% and 75.0%, respectively, and was similar to warm years (see p. 75).
Seasonally, winter atmospheric conditions contribute to determining summer oceanographic conditions over the EBS shelf. Both the North Pacific Index (NPI) and Aleutian Low Index (ALI) provide com-plementary views of the atmospheric pressure system in the North Pacific. During winter 2023–2024, the NPI was average (Figure 2) and the strength (Figure 14) and location (Figure 15) of the Aleutian Low Pressure System were both near climatological averages. Thus, despite delayed formation of sea ice in fall 2023 (Figure 35), cold winds from the Arctic helped advance sea ice to near-normal extent by mid-winter.
Winds can impact transport and surface (upper ~30-40 m) drift of early life stages of crab and ground-fish. December 2023 had significant along-shelf winds to the southeast, and weaker but more sustained winds to the southeast from March to May 2024 (Figure 19; note: winds to the southeast also oc-curred in March and May of 2023). Such winds favor offshore transport, which has been correlated with below-average recruitment for some winter-spawning flatfish (e.g., northern rock sole, arrowtooth flounder) because larvae are transported away from suitable nursery habitat (Wilderbuer et al., 2002, 2013). Beginning in May and continuing through summer 2024, persistent storms resulted in a deeper mixed layer, which brought up deeper, cooler water from depth, such that SSTs remained cooler through at least August 2024 (Figure 27). Sea surface temperatures (SSTs) and bottom temperatures were near the long-term averages in all regions by summer 2024. Notable deviations include (i) warm SSTs in the outer domain from fall 2023 through spring 2024 and (ii) unusually warm bottom temperatures in the northern outer domain since spring 2024 that may indicate an intrusion of shelf water (Figure 34).
Measures of benthic productivity showed mixed signs over the SEBS in 2024. Increases were observed for sedentary sea anemones and sea pens (Figure 43), as well as motile eelpouts and poachers (Figure 116). The biomass of the motile epifauna guild (e.g., echinoderms, crabs) remains above the long term mean, buoyed by above-average biomass of echinoderms (Figure 2). Crab populations in the EBS remain low, though relative increases were observed for tanner crab and snow crab in 2024 (Figure 118). Decreases were observed for sedentary sponges (Figure 43), as well as motile sea stars (Figure 116). The biomass of the benthic forager guild (e.g., small-mouthed flatfishes) remains below the time series mean (Figure 2). The condition of small-mouthed flatfish was also mixed in 2024 (Figure 95), potentially reflecting species-level differences in metabolic demand and/or spatial overlap with prey resources.
The community of St. Paul Island has been collecting regular (~weekly) CTD observations for nearly the past decade, including chlorophyll a concentrations (Figure 45). June–August 2024 had some of the highest chlorophyll a concentrations on record with values ~10 µg/L in June and July, which was considerably higher than 2023. At mooring M2, the peak of the 2024 spring bloom was slightly later than average, but the fall bloom occurred unusually early (Figure 46). The fall bloom started in early September, ~ 1 month earlier than usual (Sigler et al., 2014). Frequent storm events during summer 2024 resulted in weaker water-column stratification (P. Stabeno, pers comm). A large storm in late August likely caused water column mixing, which introduced nutrients to the surface, and initiated the early fall bloom. Weak stratification and the early fall bloom likely contributed to a lesser coccolithophore bloom (J. Nielsen, pers comm). The fall bloom may provide a sustained prey resource for zooplankton through the fall.
The Rapid Zooplankton Assessment in spring 2024 noted moderate abundance of small copepods, but low abundance of large copepods and near-zero abundance of euphausiids, which is typical for the spring. In summer, small copepods remained abundant throughout the region. Large copepods remained in low abundance while euphausiids increased, especially towards the northern portion of the SEBS (see p. 89). Euphausiid density during the summer acoustic survey declined in 2024 to the second-lowest value in the time series (Figure 61). In fall, both small and large copepods as well as euphausiids were in low abundance, but increased towards the north (Figure 53). Euphausiids had significantly higher lipid content in 2024 relative to 2022 (Figure 59). The biomass of jellyfish remained low to average in 2024 (Figures 62, 63), representing no significant change in competitive pressure for planktivorous predators like pollock.
As the numerically dominant forage fish in the EBS, age-0 pollock are an important component of available forage over the SEBS shelf to piscivorous predators such as Pacific cod, pollock, seabirds, and marine mammals. In spring 2024, larval pollock abundance was the highest of years sampled (2012, 2014, 2016, 2018, 2024; Figure 66), with larval condition highest in the southeast and lowest to the northwest (Figure 67). By late summer, age-0 pollock CPUE estimates were low in the middle domain (Figure 71), but at the same time in the inner domain, age-0 pollock were the most numerous non-salmonid species collected in the ADF&G nearshore survey (Figure 7). In the middle domain, age-0 pollock were distributed shallower (Figure 73), similar to a warm year, even though the mixed layer depth was deeper (Figure 27) and SSTs were cooler. Since 2022, with the cooler SSTs, pollock weights and energy density have been low while % lipid has been average (see p. 126). Juvenile and adult Pacific herring and capelin are predominantly caught in the NBS (Andrews et al., 2015). As such, both species were in low abundance in the SEBS (Figures 78 and 79). Conversely, the 2024 forecast for Togiak herring was the fifth highest on record, but was 32% lower than the 2023 forecast (see p. 134). Quantitative linkages among ecosystem drivers and forage fish biomass were explored (see p. 115) using dynamic structural equation modeling (DSEM) to illustrate how environmental changes might affect the availability of different forage fish species.
Salmon have unique, species-specific life histories that can extend throughout the Bering Sea ecosystem and into the Gulf of Alaska. Species have shown contrasting responses to the recent return to cooler average temperatures following persistent warm conditions from 2014–2021. Western Alaska chum salmon, for example, occupy the EBS in summer as juveniles before overwintering in the Gulf of Alaska, therefore their dynamics as juveniles are reflective of the pelagic environment in the EBS. These stocks collapsed during the warm period, driven by changes in prey and subsequent energetic condition (Figure 54, Figure 86, Farley Jr et al., 2024). In the SEBS, juvenile chum salmon fish condition has remained below average in 2022 and 2024 (Figure 86). Conversely, juvenile chum salmon condition has improved in the NBS since 2021 (Figure 87) and the 2024 juvenile abundance estimate from the NBS surface trawl survey was the highest on record (Figure 89). These divergent trends indicate better pelagic foraging conditions for juvenile chum salmon in the NBS than the SEBS. Bristol Bay sockeye salmon life history strategies and population dynamics reflect their freshwater rearing (i.e., lakes) and marine migratory pathways that favor warm conditions. In cooler years, Bristol Bay sockeye delay offshore migration to remain in warmer nearshore waters (Farley et al., 2007) while in warmer years they migrate offshore more rapidly, avoiding extreme marine heatwave conditions over the shelf. Thus, in the recent cooler (i.e., average) conditions, sockeye salmon may have remained nearshore and reflect nearshore foraging conditions. Adult run sizes in Bristol Bay in 2023 and 2024 have been closer to long-term averages (Figure 90). In 2024, juvenile sockeye salmon abundance and fish condition was low in the SEBS. Chinook salmon population dynamics are not as straightforwardly linked to marine conditions; populations have been declining more broadly since the early 2000s, indicating a more complex set of stressors that may be constraining production of these stocks (see p. 139).
Groundfish condition (i.e., length-weight residuals) can provide insights into the foraging conditions for both benthic (e.g., small-mouthed flatfishes) and pelagic (e.g., pollock) foragers. The condition of benthic foragers has been mixed since 2021, showing no clear trends of increasing or decreasing (Figure 95), with estimates of biomass also being mixed (yellowfin sole +8%, northern rock sole +4%, and Alaska plaice -3%) in 2024. Trends in benthic infaunal prey are indirectly assessed via the motile epifauna guild which has remained above the time-series mean since 2010 (Figure 2). The condition of pelagic foragers has decreased and/or remained below the time-series average since ~2021 (Figure 95), yet estimates of biomass from the bottom trawl survey increased (pollock +74%, arrowtooth flounder +26%) or showed a slight decline (Pacific cod -4%) in 20245. The revised Oscillating Control Hypothesis (Hunt et al., 2011) would predict a return of large, lipid-rich copepods under cooler conditions. It is important to note that thermal conditions over the EBS shelf have largely cooled to average, not cold, conditions (e.g., Figure 30). This may partially explain why the thermal conditions over the SEBS shelf since 2021 have not yet prompted a return of large, lipid-rich copepods or euphausiids over the shelf.
Individual groundfish species may be more or less able to shift their distribution to find preferred thermal conditions or more abundant and/or energetically favorable forage. For example, based on Food Habits Lab stomach content analysis, pollock consumption of copepods increased from 2023 to 2024, replacing euphausiids as the greatest percent by weight in the diets (K. Aydin, pers comm). This may be explained by the spatial distribution of the pollock population in 2024, which was concentrated over the northwest outer domain (L. Barnett, pers comm), where large ‘oceanic’ copepods occur (euphausiids mainly occur over the middle domain). Additionally, rates of cannibalism have been low between 2021–2024; the lowest year for cannibalism on record was 2018 (K. Aydin, pers comm). More generally, from 2010–2024 adult and juvenile pollock and P. cod growth potential across the EBS (SEBS and NEBS) remains below the long-term average (1982–2010), likely due to metabolic demands that have increased faster than consumption rates or changes in energetic density (see p. 168).
Metrics of stability in the fish community (for species regularly caught in the SEBS bottom trawl survey) indicate overall stability and resilience, although there are anomalous peaks in individual species (e.g., capelin, sablefish). Trends in mean lifespan (Figure 123) show little year-to-year variability and give no indication of shifts between short-lived and longer-lived species. The mean length and the stability of the groundfish community remained above average in 2024 (1982–2024; Figures 124, 125).
Seabirds are indicators of secondary productivity and shifts in prey availability that may similarly affect commercial fish populations. Species that experienced recent population losses (e.g., least auklets) have not rebounded. Overall, reproductive success was mixed for both fish-eating and plankton-eating species, but generally higher for species on St. George Island, similar to 2023 (Figures 119 and 120). This may indicate differences in local availability of small schooling forage fish and zooplankton, respectively, in feeding areas utilized by seabirds of each island. High rates of colony disturbance by bald eagles at St. Paul Island also contributed to the reduced reproductive success there (M. Rustand and H. Renner, pers comm). No major seabird die-off events were observed in 2024 (Figure 121).
We track emerging stressors like ocean acidification (OA), as well as emerging science tools and quanti-tative applications to better understand dynamics and potential impacts to the EBS ecosystem. Metrics of OA (pH and Ωarag) continued a multi-decadal decline, indicating more corrosive bottom-water con-ditions for marine calcifiers, though values have improved slightly since 2022. At this time, there is no evidence that OA can be linked to recent declines in crab populations. It is worth noting that Oarag is approaching the threshold value (<1.0) for pteropod shell dissolution that could have subsequent biological significance through the food web (Figure 128). Environmental DNA (eDNA) can provide single-species and community level information that could be used in fisheries assessments and manage-ment (see p. 29). While eDNA cannot replace some of the biological data collected from traditional survey methods, there are numerous opportunities to expand eDNA collections and leverage eDNA data for fisheries research and management. Several current and on-going collaborations are addressing gadid populations in the Arctic, groundfish communities in the EBS, northern fur seal and Steller sea lion diet studies in the Aleutian Islands, and ice seal surveys along the ice edge during spring break-up. Boreal-ization, or the broad reorganization of Arctic ecosystems to a more temperate physical and biological state, is included as a new index for the SEBS (Figure 2). The index includes nine time series, from physical to biological, and may be useful for summarizing climatic and ecological changes in the SEBS. The borealization index has reverted to values similar to the time series mean during 2022–2024 (Figure 8b).
Northern Bering Sea
Similar to the SEBS, the NBS has returned to more neutral thermal conditions since ~2021. While the extent of sea ice has been steadily increasing in the NBS since 2018 (Figure 3), ice thickness increased dramatically in the Bering Strait region (i.e., a step-change increase) since 2021 (Figure 39b) and has been above average through 2024. Ice thickness may also be a proxy for ice residency over the shelf, which may be related to the abundance of ice algae that contributes to the productivity of the NBS ecosystem. Additionally, the proportion of open-water phytoplankton blooms has been low since 2021 (2024 was the lowest since 2008) (Figure 3), therefore the proportion of ice-associated blooms has been higher. Previous research indicates that a higher proportion of ice-associated blooms is thought to result in a higher abundance of pelagic secondary producers in the summer and fall (Coyle et al., 2008; Kimmel et al., 2018). In fact, the abundance of large copepods measured in fall over the NBS shelf has increased to the time series mean (2000–2024) since 2021 (Figure 3). In 2024, large copepods were patchy with the highest values north and south of St. Lawrence Island (Figure 55) and their lipid content was significantly higher in 2024 compared to 2023 (Figure 58). Jellyfish biomass increased in 2023 and remained high in 2024 (Figure 62). Taken together, these indicators show that pelagic forage has increased in the NBS since 2021.
Measures of pelagic productivity in the NBS include age-0 pollock, herring, capelin, and juvenile salmonids CPUE. Age-0 pollock CPUE estimates have remained low compared to those in the SEBS (Figure 71). Age-0 pollock weight has been below average in 2022–2024 and while % lipid increased from below average in 2021 to above average in 2024. Age-0 pollock energy density has decreased from above average in 2021 to below average in 2024 as fish are smaller overall (see p. 126). Herring also remained low in the NBS, but capelin increased dramatically from 2023 to 2024 (Figure 79). Juvenile salmonid condition, measured as energy density anomalies, varied among species in the NBS in 2024 (Figure 87). For juvenile pink, chum, and coho salmon condition decreased from positive in 2023 to average in 2024. Juvenile Chinook salmon condition increased from average to positive in 2024. The abundance of juvenile Chinook salmon was at record low in 2024 (Figure 88) while fall juvenile chum salmon was at a record high in 2024 (Figure 89). Trends in pelagic productivity have been mixed since 2021, indicating a potential lagged response to the average thermal conditions.
In the NBS, we track harmful algal blooms (HABs) as an emerging stressor to the ecosystem as well as people’s nutritional, cultural, and economic needs. Recent oceanographic changes in the EBS has made conditions more favorable for HAB species, particularly the dinoflagellate Alexandrium catenella and diatoms in the genus Pseudo-nitzschia (Anderson et al., 2012). In October of 2023, no harvested bowhead whales contained domoic acid, but 90% contained low levels of saxitoxin (Figure 131). This confirms a consistent trend of higher prevalence of saxitoxin than domoic acid in Arctic food webs observed in all regions including the Bering Strait (see p. 215).
Looking ahead through spring 2025, the expected transition to La Niña is projected to bring continued cooler conditions to the EBS shelf with SST anomalies within 0.25°C of normal (Figure 16). Relatively cool SSTs during the early ice season (Oct 15 - Dec 15) may contribute to earlier formation of sea ice than has been observed over the last several years. However, recent storms (e.g., October 20–22, 2024) in the NBS and Bering Strait region, which caused extreme flooding in Shishmaref, may now entrain relatively warmer water into the surface layer and delay sea ice formation. The NBS has shown some indications of ‘recovery’ from the prolonged warm period (2014–2021). Some trends remain mixed since the return to neutral thermal conditions in 2021. If continued neutral or cool conditions persist, it will be informative to observe how benthic and pelagic indicators of crab and groundfish populations respond to those conditions in the NBS in 2025.
5https://meetings.npfmc.org/CommentReview/DownloadFile?p=f2da5d6f-17ae-4af0-961a-112377153dc6.pdf&fileName=2024%20EBS%20Bottom%20Trawl%20Survey%20Presentation.pdf