< Ecosystem Status Report

Eastern Bering Sea Indicators

Indicators presented in this section are intended to provide detailed information and updates on the status and trends of the eastern Bering Sea ecosystem components. These indicators are used to build the Ecosystem Status Report

Last updated: 2017

 
 
  • Contributed by Robert Lauth and Elizabeth Dawson Resource Assessment and Conservation Engineering Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA Contact: bob.lauth@noaa.gov

    Last updated: October 2018

    Description of indicator: Survey operations for the annual AFSC eastern Bering Sea shelf bottom trawl survey in 2017 started on 31 May and ended on 31 July

    Status and trends: Surface temperature mean for the 2018 eastern Bering Sea shelf decreased from the 2017 estimate, while bottom temperature mean increased from the 2017 estimate, but both were still warmer than the long-term time-series mean (Figure 26). The 2018 mean surface temperature was 7.5oC, which was 0.3oC lower than 2017 and 0.9oC above the time-series mean (6.6oC). The mean bottom temperature was 4.3oC, which was 1.4oC higher than 2017, and 1.8oC above the time-series mean (2.5oC). The ‘cold pool’, defined as the area where bottom temperatures are <2 oC, was confined to a very small part of standard EBS survey area (Figure 27). Moreover, it was the lowest areal coverage of the cold pool in the 37-year EBS shelf time-series and the first time that bottom temperatures <0 oC were not observed within the standard EBS survey area (Figure 28).

    Factors influencing observed trends: Warm and cold years are the result of interannual variability in climatic conditions that affect the extent, timing, and retreat of sea ice on the eastern Bering Sea shelf. During warmer than average years, seasonal sea ice generally does not extend as far down the shelf and retreats earlier in the spring.

    Implications: The relatively large interannual fluctuations in bottom temperature on the EBS shelf affect the spatial and temporal distribution of groundfishes and the structure and ecology of the marine community (Kotwicki and Lauth, 2013; Mueter and Litzow, 2008; Spencer, 2008). The timing of phytoplankton and subsequent zooplankton blooms are also affected by the extent of sea ice and timing of its retreat which in turn can affect survival and recruitment in larval and juvenile fishes as well as the energy flow in the system (Hunt et al., 2002; Coyle et al., 2011; Coyle and Gibson, 2017).

    Average summer surface and bottom temperatures
    Figure 26: Average summer surface (green circles) and bottom (blue triangles) temperatures (oC) of
    the eastern Bering Sea shelf collected during the standard bottom trawl surveys from 1982{2018. Water
    temperature samples from each station were weighted by the proportion of their assigned stratum area.
    Dotted lines represent the time-series mean for 1982-2018.

     

     

     

    Map showing the near-bottom temperatures from the 2018 eastern Bering Sea shelf bottom trawl survey
    Figure 27: Map showing the near-bottom temperatures from the 2018 eastern Bering Sea shelf bottom trawl survey.
  • Contributed by Nick Bond (UW/JISAO)
    NOAA/PMEL, Building 3, 7600 Sand Point Way NE, Seattle, WA 98115-6349
    Contact: nicholas.bond@noaa.gov
    Last updated: August 2018

    Summary: The state of the North Pacic atmosphere-ocean system during 2017-2018 was rather similar to that during 2016-17. Both winters featured La Ni~na and weaker than normal Aleutian lows (positive sea level pressure, SLP anomalies). The more prominent sea surface emperature (SST) anomalies during 2017-18 tended to be in the positive sense, with persistent warmth in the subtropical eastern North Pacic, increasing positive anomalies in the Bering Sea, and the expansion of warm waters o the east coast of Asia. The Pacic Decadal Oscillation (PDO) was slightly positive during the past year, with a decline to near zero in the summer of 2018. The climate models used for seasonal weather predictions are indicating about a 70% chance of a weak-moderate El Ni~no for the winter of 2018-19, and warmer than normal SSTs in both the
    western and eastern mid-latitude North Pacic in early 2019.

    Regional Highlights:
    Alaska Peninsula and Aleutian Islands. The weather of this region included suppressed storminess during the fall of 2017 and the following winter of 2017/18. The regional wind anomalies were from the southwest in an overall sense. Based on synthetic data from NOAAs Global Ocean Data Assimilation System (GODAS), the Alaska Stream appears to have been relatively di use, as opposed to concentrated into a narrower, high velocity ow, on the south side of the eastern Aleutian Islands. The eddy activity in this region was on the low side (see leutian Islands Ecosystem Status Report). Bering Sea. The Bering Sea had the least amount of sea ice in the observational record back to 1979. This can be attributed to the delayed start of winter (Being Strait was still open on 1 January) and then very mild temperatures with strong winds from the southwest, particularly in February 2018. An important consequence was a cold pool in summer 2018 of exceedingly small areal extent. The weather during summer 2018 was stormier than usual on the southeast Bering Sea shelf; at the time of this writing it is unknown if those conditions helped sustain primary production later into the warm season than usual. In the region of the M2 mooring the thermal stratication during summer 2018 was somewhat less than observed during recent years; the vertically integrated heat content was the second greatest on record, topped by 2016. Arctic. The winter of 2017/18 was relatively warm in the Arctic, and included an extreme \heat wave" (for the season) in the central Arctic during February. The Arctic's maximum ice extent in mid-March 2018 was the 2nd lowest on record. On the other hand, the decline in sea ice coverage during the late spring and early summer of 2018 was on the slow side, primarily in association with relatively low SLP in the central Arctic and cool and cloudy weather. The west winds accompanying this circulation pattern helped maintain a wide
    band of ice near the coast east of Pt. Barrow. Relatively rapid losses in sea ice concentrations and coverage occurred here in late July 2018. The edge of the pack ice in the Chukchi Sea was well north of its usual

  • Description of indices: The state of the North Pacic climate from autumn 2017 through summer 2018 is summarized in terms of seasonal mean sea surface temperature (SST) and sea level pressure (SLP) anomaly maps. The SST and SLP anomalies are relative to mean conditions over the period of 1981{2010. The SST data are from NOAA's Optimum Interpolation Sea Surface Temperature (OISST) analysis; the SLP data are from the NCEP/NCAR Reanalysis project. Both data sets are made available by NOAA's Earth System Research Laboratory (ESRL) at http://www.esrl.noaa.gov/psd/cgi-bin/data/composites/printpage.pl.

    Status and trends: The eastern portion of the North Pacic ocean experienced during 2014{16 one of the most extreme marine heat waves in the observational record (Scannell et al., 2016); the interval summarized here can be considered a transition period between that event and a more climatologically normal SST distribution on the basin-scale. More detail on the evolution of the SST and SLP from a seasonal perspective is provided directly below.

    The SST during the autumn (Sep{Nov) of 2017 (Figure 14a) was warmer than normal across almost the entire North Pacic Ocean. Greater positive (> 1oC) anomalies occurred in the Chukchi Sea and northwest Bering Sea in the northern and eastern Bering Sea, resulting in a delayed onset of sea ice the following winter. The SST anomalies were negative in the eastern equatorial Pacic in association with the development of La Ni~na. The SLP pattern during autumn 2017 featured prominent positive anomalies over the north central portion of the North Pacic Ocean, with the greatest departures from normal over the open ocean south of the western tip of the Alaska Peninsula (Figure 15a). This SLP distribution implies an enhanced storm track along the east coast of Asia, and  uppressed storminess from the Aleutians into the Gulf of Alaska (GOA).

    The North Pacic atmosphere-ocean system during winter (Dec{Feb) of 2017-18 re ected to large extent a continuation of the previous fall season. The distribution of SST anomalies (Figure 14b) was quite similar, with some additional warming in the subtropical northeastern Pacic extending southwestward from southern California. The equatorial Pacic was characterized by weak/moderate La Ni~na conditions with the strongest negative SST anomalies well east of the dateline. The SLP during this period (Figure 15b) featured an expansion of the pattern of the season before in terms of both magnitude and area, with substantial positive anomalies from about 160oE to western North America north of about 30oN. This relatively high SLP in combination with negative SLP anomalies over the East Siberian Sea resulted in a pressure pattern that supported extremely strong wind anomalies (3 to 4 m s-1) from the southwest across the Bering Sea.


    The distribution of anomalous SST in the North Pacic during spring (Mar{May) of 2018 (Figure 14c) was similar to that during the previous winter season. Exceptions were warming relative to seasonal normal in the eastern Bering Sea and in an east-west band from 25o to 40oN from Japan to the dateline. The SST anomalies in the tropical Pacic were of minor amplitude with the ending of La Ni~na. The SLP anomaly pattern (Figure 15c) for spring 2018 featured bands of lower than normal pressure from eastern Siberia to northwestern Alaska and higher pressure from south of the Aleutian Islands to the GOA, resulting in another season of warm, southwesterly ow anomalies across the Bering Sea. The atmospheric circulation in the northeast Pacic promoted relatively upwelling-favorable winds in the coastal GOA.


    The SST anomaly pattern in the North Pacic during summer (Jun{Aug) 2018 is shown in Figure 14d. Positive anomalies continued in a broad band extending from Japan to the southeastern GOA and from the northern Bering Sea into the Chukchi Sea. In the latter area, particularly strong positive temperature anomalies (exceeding 2oC) developed in the vicinity of Bering Strait. Near normal SSTs were present along most of the west coast of North America from Vancouver Island to southern California. Warmth continued in the subtropical eastern North Pacic from Baja California to the equatorial Pacic east of the dateline, where temperatures were roughly 0.5oC above normal. The distribution of anomalous SLP (Figure 15d) during summer 2018 included mostly just weak anomalies, which is typical for the season. A band of higher than normal pressure extended from the western North Pacic north of about 30oN into the GOA. Lower pressure extended from northwestern Canada across interior Alaska into the Bering Sea.

    Sea Surface temperature anomolies
    Figure 14: SST anomolies for autumn (September{November 2017), winter (December 2017{February 2018), spring (March{May 2018), and
    summer (June{August 2018).

     

  • Contributed by Nick Bond (UW/JISAO)
    NOAA/PMEL, Building 3, 7600 Sand Point Way NE, Seattle, WA 98115-6349
    Contact: nicholas.bond@noaa.gov
    Last updated: September 2018

    Description of indices: Climate indices provide an alternative means of characterizing the state of the
    North Pacic atmosphere-ocean system. The focus here is on ve commonly used indices: the NINO3.4
    index for the state of the El Ni~no/Southern Oscillation (ENSO) phenomenon, Pacic Decadal Oscillation
    (PDO) index (the leading mode of North Pacic SST variability), North Pacic Index (NPI), North Pacic
    Gyre Oscillation (NPGO), and Arctic Oscillation (AO). The time series of these indices from 2008 into
    spring/summer 2018 are plotted in Figure 16.

    Status and trends: The North Pacic atmosphere-ocean climate system was mostly on the warm side
    during 2017{18. This was despite the second fall/winter in a row with a negative value for the NINO3.4
    index in association with a weak/moderate La Ni~na event. The positive state of the PDO (indicating warmer
    than normal SST along the west coast of North America and cooler than normal in the central and western
    North Pacic) that began in early 2014 ended in 2017. This decline is consistent with the typical remote
    eects of ENSO, and in particular the transition from a strong El Ni~no in 2015{16 to the following two
    episodes of La Ni~na. The SST anomaly distribution during spring and summer of 2018 has a minimal
    projection on the characteristic pattern of the PDO. The NPI was strongly positive from fall 2017 into 2018
    due to the relatively high SLP in the region of the Aleutian low. A positive sense for the NPI commonly
    accompanies La Ni~na, its magnitude from late 2017 into 2018 was greater than might be expected.

    The NPGO became strongly negative in 2017, and stayed negative into 2018 (February is the latest month
    for which this index is available). This index has undergone an overall decline from positive values during the
    period of 2008 to 2012. The AO represents a measure of the strength of the polar vortex, with positive values
    signifying anomalously low pressure over the Arctic and high pressure over the Pacic and Atlantic Ocean
    at a latitude of roughly 45oN. It was in a near-neutral state during the last half of 2017 with a transition
    to a positive state in spring 2018 that has continued into summer. A consequence has been relatively low
    pressure in the Arctic during early summer.

    Contributed by Nick Bond (UW/JISAO)
    NOAA/PMEL, Building 3, 7600 Sand Point Way NE, Seattle, WA 98115-6349
    Contact: nicholas.bond@noaa.gov
    Last updated: September 2017

  • Contributed by Nick Bond (UW/JISAO)
    NOAA/PMEL, Building 3, 7600 Sand Point Way NE, Seattle, WA 98115-6349
    Contact: nicholas.bond@noaa.gov
    Last updated: September 2018

    Description of indicator: Seasonal projections of SST from the National Multi-Model Ensemble (NMME)
    are shown in Figures 17. An ensemble approach incorporating dierent models is particularly appropriate
    for seasonal and longer-term simulations; the NMME represents the average of eight climate models. The
    uncertainties and errors in the predictions from any single climate model can be substantial. More detail
    on the NMME, and projections of other variables, are available at the following website: http://www.cpc.
    ncep.noaa.gov/products/NMME/.

    Status and trends: First, the projections from a year ago are reviewed qualitatively. From an overall
    perspective, the SST forecasts were essentially correct with respect to their basin-scale patterns of negative
    and positive SST anomalies. The NMME forecasts included an under-prediction of the magnitudes of some
    of the more prominent anomalies. In particular, Alaskan waters generally ended up warmer than forecast, especially the Bering Sea shelf during late winter and early spring 2018 where there was much less sea ice
    than suggested by the model forecasts made during September 2017.

    Time series of the Nino, PDO, NPI, NPGO, and AO
    Figure 16: Time series of the NINO3.4 (blue), PDO (red), NPI (green), NPGO (purple), and AO
    (turquoise) indices for 2008{2018. Each time series represents monthly values that are normalized using
    a climatology based on the years of 1981{2010, and then smoothed with the application of three-month
    running means. The distance between the horizontal grid lines represents 2 standard deviations. More
    information on these indices is available from NOAA's Earth Systems Laboratory at http://www.esrl.
    noaa.gov/psd/data/climateindices/

    These NMME forecasts of three-month average SST anomalies indicate a continuation of warm conditions
    across virtually all of the North Pacic through the end of the year (Oct{Dec 2018) with a reduction in
    the longitudinal extent of cooler than normal temperatures oshore of the Pacic Northwest (Figure 17a).
    The magnitude of the positive anomalies is projected to be greatest (exceeding 1oC) north of the Kuroshio
    Extension in the western North Pacic and in the northern portion of the Bering Sea. Positive SST anomalies
    are projected in the central and eastern equatorial Pacic. The ensemble model average is strong enough to
    constitute El Ni~no of weak to moderate magnitude. As of early September 2018, the probabilistic forecast
    provided by NOAA's Climate Prediction Center (CPC) in collaboration with the International Research
    Institute for Climate and Society (IRI) for the upcoming fall through winter indicates about a 70% chance of El Ni~no, and otherwise equatorial SSTs in the neutral category. The overall pattern of SST anomalies across
    the North Pacic is maintained through the 3-month periods of December 2018{February 2019 (Figure 17b)
    and February{April 2019 (Figure 17c). There is moderate but by no means a complete consensus among
    the models that the Aleutian low will be deeper than normal (negative SLP anomalies) during the latter
    portion of the winter of 2018{2019. This is a common remote response to El Ni~no, and tends to result in
    relatively warm late winter and early spring weather for Alaska that is liable to be enhanced by the eects
    of the warmth of the waters surrounding Alaska. For the period of February{April 2019, the models are
    projecting little noticeable decline in the magnitude of the equatorial Pacic temperature anomalies even
    though El Ni~no often weakens during the boreal spring. The positive SST anomalies along the west coast of
    North America that are indicated in Figure 17c commonly occur after El Ni~no winters.

    Implications: The PDO has also generally been positive during these kinds of periods in the past, but
    the predicted warmth in both the western and eastern portions of the mid-latitude North Pacic does not
    resemble the characteristic pattern of the PDO. An important implication is that the PDO is liable to be
    ill-suited for characterizing the state of the North Pacic in early 2019.

    Inser Figure 17

     

  • Summary. After 3 consecutive warm years (2014–2016), the Bering Sea experienced moderate climate conditions in 2017. Sea level pressures over the Bering Sea were moderate and summer winds were very light from the south giving slightly positive air temperature anomalies (∼+1oC) over all of the Bering Sea. While residual heat maintained above-average water temperatures (both surface and bottom) over the shelf, sea ice extended over much of the southern shelf. This resulted in a larger, although narrow, cold pool over the shelf with weaker thermal gradients at the boundaries. The PDO pattern suggests a possible continuation of moderate conditions into 2018.

    Air temperatures Slightly positive near-surface air temperature anomalies for summer (May– July) in southwest Alaska and the southeastern Bering Sea were +1–1.5oC above those of the eastern regions (Figure 21). Alaska conditions were driven by a moderation of the Pacific Decadal Oscillation (PDO) and a generally western location for the Aleutian Low pressure feature. The moderation of the PDO has resulted in milder, positive SSTs along the coastal Gulf of Alaska with associated low level warm air temperature anomalies. Winds follow the contours of geopotential heights, with east-west gradients associated with the warm temperature regions and the coastal mountains (Figure 22) to the east and the Aleutian Low to the west, giving southerly winds that advect warm temperatures into Alaska. Summer winds were very light from the south giving slightly positive air temperature anomalies over the Bering Sea (Figure 22). Long-term surface air 71 temperatures measured on St. Paul Island (Figure 23) also reflect a moderation of recent warm conditions.

    Insert figure 21

    Sea ice. Seasonal sea ice is a defining characteristic of the Bering Sea shelf. The presence of sea ice influences the timing of the spring bloom and bottom temperatures throughout the year. Over the time series, years with less sea ice coverage occurred in 2001–2005 and years with more extensive sea ice coverage occurred in 2007–2010, 2012, and 2013 (Figure 24). Conditions in 2017 resulted in a more southerly extent of sea ice over the Bering Sea shelf (more similar to 2006). Note the unusual feature of the 2017 sea ice boundary retreating into the Gulf of Anadyr.

    Ocean temperatures. The cold pool (Figure 25), defined by bottom temperatures 2 oC, influences not only near-bottom biological habitat, but also the overall thermal stratification and ultimately the mixing of nutrient-rich water from depth into the euphotic zone during summer. The cold pool extent for summer 2017 was more extensive over the shelf, although more narrowly confined to the middle domain (50–100 m).

    Depth-averaged temperatures. Figure 26 shows the depth-averaged temperatures at mooring M2 located over the southern shelf. The cold years of 2009, 2010, and 2012 cluster together with cooler temperatures; intermediate years of 2006, 2011, and 2017 group together; warm years of 2014–2016, with 2016 being especially warm, cluster together.

    Insert figure 22

    Insert figure 23

    Insert figure 24

    Insert figure 25

    Insert figure 26

     

  • Description of indicator: Oceanographic and fisheries data were collected over the eastern Bering Sea (EBS) shelf during fall 2002–2016 for a multiyear fisheries oceanography research program, Bering-Arctic-SubArctic Integrated Survey (BASIS). Stations were located between 54.5oN and 65oN, at ∼60 km resolution. Bristol Bay stations were sampled from mid-August to early September, while stations in the central and northern EBS were generally sampled from mid-September to early October. Physical oceanographic data were obtained from vertical conductivity-temperaturedepth (CTD) profiles. Mean temperature and salinity above and below the mixed layer depth (MLD) were estimated for each station following methods in Danielson et al. (2011). Normalized anomalies (mean yearly value minus average value over 2002–2016 normalized by standard deviation) of temperature and salinity were separately computed for each Bering Sea Project region (Ortiz et al., 2012) (Figures 29–33). Normalized anomalies of MLD were similarly estimated for middle and outer domain regions (Figure 34). Only station locations sampled 5+ years were included in the analyses (Figure 29).

    Status and trends: Temperatures above and below the MLD (Tabove, Tbelow) were roughly warmer than average in 2002–2005, average in 2006, and cooler than average in 2007–2012 (Figures 30 and 31). In 2014 and 2016, Tabove was high for all regions (with the exception of St. Matthew Island in 2014), whereas in 2015 it was above average in only two regions, likely due to the early onset of fall mixing which deepened the MLD in 2015 (Figure 34). Tbelow was above average primarily in southern regions in 2014 and 2015. In contrast, in 2016 Tbelow was high in both southern and northern regions, similar to the earlier warm periods of 2003–2005. Salinities above and below the MLD (Sabove, Sbelow) for the south middle shelf (regions 3 and 6) were generally higher in warm years (2002–2005, 2014–2016) than in cold years (2006–2012) (Figures 32 and 33). With the exception of 2015, the average MLD varied ∼10 m in the south middle domain (regions 3, 6), 6–7 m in the north middle domain (regions 9, 10), and 13 m in the south outer domain (region 4); variations did not appear to co-vary with warm or cold year periods (Figure 34).

    Insert Figure 29

    Insert Figure 34

    Factors influencing observed trends: Sea ice during winter and spring extended farther to the south as the climate cooled. The cold pool is related to sea ice and thus extends farther south in years with higher sea ice coverage in the southern Bering Sea. The cold pool (located below the MLD) is always present in the northern Bering Sea since ice covers this region each year (Stabeno et al., 2012). The lower bottom salinities near the coast (e.g., inner domain regions and Norton Sound) indicate major freshwater input from the Yukon and Kuskoquim rivers (Figures 29, 32, and 33). Variations in salinity on the middle and outer shelf may be partially related to wind direction, with southeasterly winds producing enhanced on-shelf flows of oceanic water in warm years (Danielson et al., 2012). Therefore, the lower salinity in cold years on the south middle shelf may be due to ice melt and possibly reduced onshore flow of higher salinity waters. Tabove and Sabove are influenced by temporal mixing events relating to episodic wind mixing/storm events, while Tbelow and Sbelow may better reflect longer term climatic shifts. For example, in 2005 (a warm year), Tbelow was warmer than average in the middle domain regions 3, 6, and 9 reflecting the lack of sea ice during spring (Figure 31). In contrast, Tabove was average in these regions (Figure 30), due to high wind mixing in August prior to and during the survey (Eisner et al., 2015).

    Implications: The variations of temperature and salinity between Bering Sea Project regions indicate that water mass properties vary considerably both spatially (horizontally across regions and vertically above and below the MLD) and interannually, and will impact ecosystem dynamics and distributions of zooplankton, fish, and other higher trophic levels. For example, larger more lipid rich zooplankton generally show increases in abundance in both the water column and in forage fish diets in cold compared to warm years (Coyle et al., 2011; Eisner et al., 2014).

     

     

    Contributed by Lisa Eisner, Jeanette Gann, and Kristin Cieciel
    Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: lisa.eisner@noaa.gov
    Last updated: August 2017

  • Description of indicator: Wilderbuer et al. (2002, 2013) summarized the recruitment of winterspawning flatfish in relation to decadal atmospheric forcing, linking favorable recruitment to the direction of wind forcing during spring. OSCURS model time series runs indicated in-shore advection to favorable nursery grounds in Bristol Bay during the 1980s. The pattern change to off-shore in the 1990–97 time series coincided with below-average recruitment for Northern rock sole (Lepidopsetta polyxystra), Arrowtooth flounder (Atheresthes stomias), and Flathead sole (Hippoglossoides elassodon) relative to the 1980s. Favorable springtime winds were present again in the early 2000s which also corresponded with improved recruitment. The time series is updated through 2017 and shown for 2009 through 2017 in Figure 35.

    Status and trends: The 2017 springtime drift pattern was mixed, with winds during the first 60 days of the 90 day drift index being unfavorable off-shore winds that changed to a northerly onshore direction in the last 30 days of the index. This causes some difficulty in interpretation of drift patterns, but they may be more consistent with years of below-average recruitment for winter85 Figure 35: OSCURS (Ocean Surface Current Simulation Model) trajectories from starting point 56oN, 164oW from April 1–June 30 for 2009–2017. spawning flatfish. Only two years out of the past ten have OSCURS runs that are consistent with those which produced above-average recruitment in the original analysis (2008, 2015). The north-northeast drift pattern suggests that larvae may have been advected to favorable, near-shore areas of Bristol Bay by the time of their metamorphosis to a benthic form of juvenile flatfish. Preliminary estimates of Northern rock sole recruitment in recent years are consistent with this larval drift hypothesis. For Arrowtooth flounder and Flathead sole, the correspondence between the springtime drift pattern from OSCURS and estimates of year class strength have weakened since the 1990s. Arrowtooth flounder produced year classes of average strength during some off-shore drift years, suggesting that this species may have different timing for spawning, larval occurrence, and settlement preferences than Northern rock sole. In the case of Flathead sole, the 2001 and 2003 year-classes appear stronger than the weak recruitment that has persisted since the 1990s.

    Insert figure 35

    Implications: The 2017 springtime drift pattern appears to be consistent with years when belowaverage recruitment occurred for Northern rock sole, Arrowtooth flounder, and Flathead sole. Wind patterns in 2008 and 2015 may promote average to above-average recruitment. 2010 featured a mixture of wind direction as there were strong northerly winds for part of the spring but also southerly winds that would suggest increased larval dispersal to Unimak Island and the Alaska Peninsula.

    Contributed by Tom Wilderbuer
    Resource Ecology and Fisheries Management Division, Alaska Fisheries Science Center, National
    Marine Fisheries Service, NOAA
    Contact: tom.wilderbuer@noaa.gov Last updated: August 2017

 
  • Description of indicator: Groups considered to be structural epifauna include: sea whips, corals, anemones, and sponges. Corals are rarely encountered on the eastern Bering Sea shelf so they were not included here. Relative CPUE by weight was calculated and plotted for each species group by year for 1982–2017. Relative CPUE was calculated by setting the largest biomass in the time series to a value of 1 and scaling other annual values proportionally. The standard error (±1) was weighted proportionally to the CPUE to produce a relative standard error.

    Status and trends: Relative catch rates for both sponges and sea anemones remained similar to estimates from 2016, which were lower than the previous 7 years, and sea whip estimates decreased significantly from 2016. These trends should be viewed with caution, however, because the quality and specificity of field identifications and their enumeration have varied over the time series (Stevenson and Hoff, 2009; Stevenson et al., 2016). Moreover, the identification of trends is uncertain given the large variability in relative CPUE (Figure 36).

    Factors influencing observed trends: Further research in several areas would benefit the interpretation of structural epifauna trends including systematics and taxonomy of Bering Sea shelf invertebrates; survey gear selectivity; and the life history characteristics of the epibenthic organisms captured by the survey trawl.

    Implications: Understanding the trends as well as the distribution patterns of structural epifauna is important for modeling habitat to develop spatial management plans for protecting habitat, understanding fishing gear impacts and predicting responses to future climate change (Rooper et al., 2016); however, more research on the eastern Bering Sea shelf will be needed to determine if there are definitive links.

    relative CPUE for benthic epifauna during the May to August time period from 1982{2017.
    Figure 36: AFSC eastern Bering Sea shelf bottom trawl survey relative CPUE for benthic epifauna
    during the May to August time period from 1982{2017.

    Figure 36

    Contributed by Robert Lauth
    Resource Assessment and Conservation Engineering Division, Alaska Fisheries Science Center,
    National Marine Fisheries Service, NOAA
    Contact: bob.lauth@noaa.gov
    Last updated: October 2017

  • Dissolved Total Inorganic Nitrogen Concentrations Above and Below the Pycnocline in the Eastern Bering Sea

    Description of indicator: We present total dissolved inorganic nitrogen (DIN) concentrations (µm) above and below the pycnocline during late summer/early fall 2003–2016 in the eastern Bering Sea. Data are divided by oceanographic domain (inner [0–50 m] and middle [51–100 m]) and further split between the northern and southern shelf at 60oN. The outer domain is left out due to inconsistent sampling. DIN (nitrate, nitrite, and ammonia) above the pycnocline at the surface mixed layer represents what is currently available for primary production at the end of summer (storm activity/wind mixing of deep nutrients to the surface tend to be lower during the summer, and surface nutrient stores are often depleted). Nutrients below the pycnocline represent what is potentially available should wind mixing become strong enough to break down the pycnocline and mix deeper layers to the surface. Sometime during autumn when wind storms increase in frequency and intensity, there is usually a second significant bloom of phytoplankton (though smaller than the spring bloom) after the summer pycnocline breaks down. During this time the deep nutrient stores are brought to the surface. This process is important for sending a new round of energy through the food web just prior to the onset of winter.

    Status and trends: DIN above and below the pycnocline varies from year to year over the eastern Bering Sea shelf in all domains. As expected, the inner domain, which is often more thoroughly mixed, is more highly correlated between the surface and deep waters (with significant correlation in the north, P = 0.002), while in the middle domain there is considerably less correlation (Figure 37). Also as expected, deeper stores of nutrients are more often found at higher concentrations than their surface counterparts. In addition, a significant decreasing trend is seen in surface DIN concentrations within the southern middle domain over time (P < 0.05, Figure 38).

    Factors influencing observed trends: During summer, the strength and frequency of summer storm events and water column stratification influence the amount of nutrients brought to surface waters from depth. Late summer concentrations of DIN at the surface may serve as an indicator of nutrient availability, with higher concentrations seen during windy years and lower stratification, and lower concentrations seen when storm activity is minimal and stratification is high (Gann et al., 2016; Eisner et al., 2016). Accordingly, years with higher water column chlorophyll a concentrations (a proxy for phytoplankton biomass) are associated with higher wind mixing.

    Implications: A decreasing trend over time in surface DIN concentrations could imply that summertime wind mixing events are occurring less frequently and with less intensity in recent years. This trend could limit the amount of primary production available as food for lower trophic levels that may ultimately be adversely affected. Diminished nutrient stores leading to lower production in the upper water column may directly affect food stores for higher trophic levels and lead to slowed growth of age-0 pollock during summer months.

    Insert figure 37

    Insert figure 38

    Contributed by Jeanette Gann and Lisa Eisner
    Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: jeanette.gann@noaa.gov
    Last updated: August 2017

  • Description of indicator: BASIS fisheries oceanography surveys were conducted in the eastern Bering Sea mid-August to late September for six warm (2003–2005, 2014–2016), one average (2006), and six cold (2007–2012) years. Variations in chlorophyll a (chla) were used to evaluate spatial and interannual differences in total phytoplankton biomass and size structure (an indication of phytoplankton species). The ratio of large (>10 µm) phytoplankton biomass to total biomass (>10 µm chla / total chla) were estimated from discrete water samples filtered through GFF and 10 µm filters and analyzed with standard fluorometric methods (Parsons et al., 1984). Integrated chla values were estimated from CTD fluorescence profiles, calibrated with discrete chla (GFF) samples. Chla data were averaged over the top 50 m of the water column or to the bottom for shallower stations. Water column stability was estimated over the top 70 m (Simpson et al., 1978). Similarly, a stratification index was estimated at PMEL Mooring 2 (M2) (Ladd and Stabeno, 2012; Eisner et al., 2015). Friction velocity cubed (u*3), a proxy for wind mixing, was obtained from NCEP reanalysis at M2 (courtesy of Nick Bond). Normalized anomalies of temperature, u*3, stratification index, integrated chla, and large size fraction chla are shown for the southeastern Bering Sea middle shelf for 2003–2016 (Figure 39)

    Insert figure 39

    Status and trends: Highest phytoplankton biomass was observed in the south outer shelf (100– 91 200 m) with highest values inshore of Bering Canyon, near the Pribilof Islands, along the Aleutian Islands, north of St. Lawrence Island, and on the south inner shelf (Figure 40). Larger phytoplankton were observed on the inner shelf and near the Pribilof Islands, and smaller phytoplankton on the south middle and outer shelf. Integrated chla varied 3-fold among all years, with the highest values seen in 2005 in the south and 2003 in the north (Figure 41). Typically years with higher integrated chla had a greater fraction of large phytoplankton. The mean size of phytoplankton assemblages were higher in early warm (2003–2005) than in cold (2006–2012) years in the south. In contrast, in more recent warm years (2014–2016) integrated chla was average, whereas large size fraction ratios were below average (Figure 39) especially in 2014 which had the lowest percent large (highest % small) phytoplankton for our time series (Figure 41). This 2014 anomaly was due to an extensive coccolithophore bloom over the north and south middle shelf (see p. 95). Coccolithophores are small phytoplankton cells (2–5 µm) with calcium carbonate plates that give the water a milky aqua appearance. Coccolithophores were also observed in 2015 and 2016 in the south Bering Sea.

    Factors influencing observed trends: Water column stability (or stratification), wind, and temperature can influence interannual and spatial variations in phytoplankton biomass. For the south middle shelf, a positive association was observed between August u*3 (wind mixing 2–3 weeks prior to chla sampling) and integrated chla in the top 50 m (Figure 42). Deep, nutrient-rich waters may be mixed to the surface to fuel production of assemblages of large phytoplankton (e.g., diatoms) during periods of high winds and low water column stability. Phytoplankton growth may be enhanced at higher temperatures, depending on species. For example, the highest chla and largest size fractions were seen in 2005, a period with high August wind mixing, average stability and high water column temperature (Figure 39). The lowest chla and smallest size fractions were observed in 2008, a period with low wind mixing, high stability, and low water column temperature. The low wind mixing in 2014 could also have favored formation of the coccolithophore bloom; these blooms are thought to be associated with low nutrient conditions. Spatially, low chla and small phytoplankton assemblages were seen in the area of highest stability, in the southeastern middle shelf near M2 (Figure 40).

    Figure 40

    Figure 41

    Implications: Phytoplankton dynamics determine the amount and quality of food available to zooplankton and higher trophic levels, and are thus important to ecosystem function. For example, larger phytoplankton assemblages may lead to shorter food webs and a more efficient transfer of energy to seabirds, fish, and marine mammals. The cloudy water associated with coccolithophore blooms may also limit feeding by visual predators (e.g. surface feeding fish and seabirds). Our data help to characterize ecosystem processes during the critical late summer period prior to the over-wintering of key forage fish (e.g., juvenile Walleye pollock, Gadus chalcogrammus; Pacific cod, Gadus macrocephalus; salmonids) (Eisner et al., 2015).

    Insert figure 42

     

    Contributed by Lisa Eisner1 , Kristin Cieciel1 , Jeanette Gann1 , and Carol Ladd2 1Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    2NOAA/PMEL, Building 3, 7600 Sand Point Way NE, Seattle, WA 98115-6349
    Contact: lisa.eisner@noaa.gov
    Last updated: August 2017

  • Description of indicator: Blooms of coccolithophores, a unicellular calcium carbonate-producing phytoplanktonic organism, are easily observed by satellite ocean color instruments due to their high reflectivity (Figure 43). However, in situ measurements in the Bering Sea suggest that the algorithm used by NASA to identify coccolithophores from ocean color is not adequate in the Bering Sea (Iida et al., 2002, 2012). Using methodology developed by Iida et al. (2002, 2012), we identify the number of satellite ocean color pixels associated with coccolithophores. Highly reflective waters in shallow water near the coast can be due to re-suspended diatom frustules rather than coccolithophores (Broerse et al., 2003). Thus, the index is calculated from the region south of 60oN and deeper than 30m depth to avoid contamination by shallow regions around St. Matthew and St. Lawrence Islands, along the Alaskan coast, as well as sediment associated with the Yukon River. Because blooms are often largely confined to either the middle shelf or the inner shelf, two indices are calculated, one for the middle shelf (50–100m depth) and one for the inner shelf (30–50m depth). Using only days that are more than 50% cloud-free, coccolithophore indices were calculated as an average area (km2 ) covered by coccolithophores during the month of September of each year. Blooms are most commonly observed and cloud cover is typically lower during September than other months, thus allowing for better quantification.

    Note that the methodology for calculating the index has changed since the 2016 contribution. In 2016, the period from 1 August to 30 September was used (instead of September only) in calculating the index. This prior estimate eliminated the 1997 data point since SeaWiFS satellite data were not available in August 1997. In addition, we currently use only data from days that are more than 50% cloud free. The 2016 index used a cut-off of 10% cloud free. These two changes do not substantially change interpretation of the index.
    Before 1997, coccolithophore blooms in the eastern Bering Sea were rare. A large bloom (primarily Emiliania huxleyi) occurred in 1997 (Napp and Hunt, 2001; Stockwell et al., 2001) and for several years thereafter. During the 1997 bloom, the bloom was associated with a die-off of short-tailed shearwaters (Puffinus tenuirostris), a seabird commonly seen in these waters (Baduini et al., 2001). It was thought that the bloom may have made it difficult for the shearwaters to see their zooplankton prey from the air (Lovvorn et al., 2001). Since then, coccolithophore blooms in the eastern Bering Sea have become more common. Satellite ocean color data suggest that blooms are only found where water depths are between 20 and 100 m. Blooms typically peak in September and appear to be related to strong stratification (Iida et al., 2012).

    Status and trends: Annual images (Figure 43) show the spatial and temporal variability of coccolithophore blooms in September. Annual indices are obtained from these satellite data by averaging spatially over the inner and middle shelf (Figure 44). Coccolithophore abundance was particularly high during the early part of the record (1997–2000), with an index (averaged over the 3 years) of 120,075 km2 for the middle and inner shelf combined. In 2001, the index dropped to 21,044 km2 and remained low ( less than 50,000 km2 ) through 2006. In 2007, the index rose above 75,000 km2 . A higher index (> 50,000 km2 ) was observed in 2007, 2009, 2011, 2014, and 2016 for the middle shelf and in 2011 and 2014 (> 20,000 km2 ) for the inner shelf. September 2017 exhibited the lowest index of the record with 9 km2 over the middle shelf and 431 km2 over the inner shelf for a total of 440 km2 .

    Factors influencing observed trends: It has been suggested that the strength of density stratification is the key parameter controlling variability of coccolithophore blooms in the eastern Bering Sea (Iida et al., 2012). Stratification influences nutrient supply to the surface layer. Stratification in this region is determined by the relative properties (both temperature and salinity) of two water masses formed in different seasons, the warm surface layer formed in summer and the cold bottom water influenced by ice distributions the previous winter. Thus, the strength of stratification is not solely determined by summer temperatures and warm years can have weak stratification and vice versa (Ladd and Stabeno, 2012).

    Implications: Coccolithophore blooms can have important biogeochemical implications. The Bering Sea can be either a source or a sink of atmospheric CO2, with the magnitude of coccolithophore blooms and the associated calcification playing a role (Iida et al., 2012). In addition, variability in the dominant phytoplankton (diatoms vs. coccolithophores) is likely to influence trophic connections with the smaller coccolithophores resulting in longer trophic chains. Coccolithophores may be a less desirable food source for microzooplankton in this region (Olson and Strom, 2002). As noted previously, the striking milky aquamarine color of the water during a coccolithophore bloom can also reduce foraging success for visual predators.

    Insert figure 43

    Insert figure 44

    Contributed by Carol Ladd1 , Sigrid Salo1 , and Lisa Eisner2
    1NOAA/PMEL, Building 3, 7600 Sand Point Way NE, Seattle, WA 98115-6349
    2Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: carol.ladd@noaa.gov
    Last updated: October 2017

  • Description of indicator: In 2015, EcoFOCI implemented an at sea Rapid Zooplankton Assessment (RZA) to provide a leading indicator of zooplankton composition in Alaska’s Large Marine Ecosystems. The rapid assessment, which is a rough count of zooplankton (from paired 20 and 60-cm oblique bongo tows from 10 m from bottom or 300 m, whichever is shallower), provides preliminary estimates of zooplankton abundance and community structure. The method employed uses coarse categories and standard zooplankton sorting methods (Harris et al., 2005). The categories are small copepods ( 2 mm; example species: Acartia spp., Pseudocalanus spp., and Oithona spp.), large copepods (> 2 mm; example species: Calanus spp. and Neocalanus spp.), and euphausiids ( 15 mm; example species: Thysanoessa spp.). Small copepods were counted from the 153 µm mesh, 20 cm bongo net. Large copepods and euphausiids were counted from the 505 µm mesh, 60 cm bongo net. In 2016, the method was refined and personnel counted a minimum of 100 organisms per sample at sea to improve zooplankton estimates. Other, rarer zooplankton taxa were present but were not sampled effectively with the on-board sampling method. Detailed information on these taxa is provided after in-lab processing protocols have been followed (1+ years post survey). The 70m isobath transect and Unimak Box were sampled April 28–May 8 2017 and sampling halted around mooring M4 due to the presence of ice. In order to provide comparison to yearly RZA data, long-term time series for the inner, middle, and outer domains were developed from archived data. The mean, annual abundance of each RZA category was plotted for the southern inner, middle, and outer shelf of the Bering Sea (Ortiz et al., 2012), and represented primarily April, May, and September as the months with the greatest sampling frequency. No RZA data were available along the inner and outer shelf in 2017 as only the 70 m isobath was sampled. Plotted on the time series were the RZA estimates from the corresponding location and year, presented as an annual mean.

    Status and trends: Each RZA category had similar abundances along the 70 m isobath and had greater abundances in the Unimak Pass area during spring 2017 (Figure 45). Only the small copepods were found in high abundances along the 70 m isobath in spring (Figure 45c). Note that sampling did not proceed further north due to the presence of ice. Large copepods were abundant in the Unimak Pass area during spring and higher abundances were measured in the northernmost stations of the 70 m isobath (Figure 45b). Small copepod abundances were significantly higher in fall of 2017, with a hotspot near St. Matthew Island (Figure 45d). Euphausiids were more prevalent in the Unimak Pass area in spring and near St. Matthew Island in fall (Figures 45e and 45f).
    Large copepod abundances were higher during 2015 and 2016 in the inner shelf when compared to long-term averages, similar to estimates during the 2002–2006 warm period in the middle shelf, and 99 similar to the long-term averages in the outer shelf (Figure 46). Small copepod abundance showed little variability over time, regardless of region (Figure 47). Euphausiid abundance was higher than recent estimates in the inner shelf, similar in the middle shelf, and higher in the outer shelf (Figure 48).

    Factors influencing observed trends: Warm and cold year ’stanzas’ influence zooplankton population dynamics in the Bering Sea (Eisner et al., 2014). Large copepod abundances were notably lower in 2017 along the middle shelf during spring (Figures 45a and 46) and this was typical of a cold year and similar to abundances observed during the cold years of 2005–2009 (Figure 46). Estimates of larger zooplankton abundance in the fall survey remained low, with the exception of the northern portion of the 70 m isobaths (Figure 45b). The slight decline observed in smaller copepod abundance during spring (Figure 45c) may also be due to cooler spring temperatures reducing smaller copepod abundances compared to more recent years (Figure 47); however, abundances were very high during the fall survey (Figure 45d). The low euphausiid abundances observed in the middle shelf (Figures 45e and 45f) appear to be typical of this region during the spring and fall (Figure 48).

    Implications: Smaller copepods form the prey base for late-larval to early juvenile Walleye pollock (Gadus chalcogrammus) during spring (Figure 45c). However, reduced abundances of smaller copepods are not necessarily detrimental as estimated production rates for smaller copepods are similar across warm and cold periods (Kimmel et al., In press). Low abundances of large copepods are less critical in the spring, but very important later in the year (Hunt et al., 2011) (see p. 145). Large copepod abundances were low along the southeastern Bering Sea shelf throughout 2017 (Figures 45a and 45b). This suggests that juvenile pollock did not encounter larger, lipid-rich copepods in the fall of 2017. It has been suggested that euphausiids may compensate for a lack of copepods during fall (Duffy-Anderson et al., 2017) and 2017 data indicate evidence of euphausiid presence on the shelf (Figure 45f) and abundances slightly higher than in recent years (Figure 48). Therefore, pollock may again find adequate prey in order to provision for overwintering.

    Insert Figure 45

    Insert Figure 46

    Insert Figure 47

    Insert Figure 48

    Contributed by Colleen Harpold and David Kimmel
    Resource Assessment and Conservation Engineering Division, Alaska Fisheries Science Center,
    National Marine Fisheries Service, NOAA
    Contact: colleen.harpold@noaa.gov
    Last updated: October 2017

  • Description of indicator: The time series for jellyfishes (primarily Chrysaora melanaster ) relative CPUE by weight was updated for 2017 (Figure 49). Relative CPUE was calculated by setting the largest biomass in the time series to a value of 1 and scaling other annual values proportionally. The standard error (±1) was weighted proportionally to the CPUE to produce a relative standard error

    Insert figure 49

    Status and trends: The relative CPUE for jellyfishes in 2017 increased by 18% from 2016; however, the 2016–17 estimates remain among the lowest observed since 1989. These low CPUE values were within the range of those observed during the first nine years of the time series (1982– 91). There was a period of increasing biomass of jellyfishes throughout the 1990’s (Brodeur et al., 1999) followed by a second period of relatively low CPUEs from 2001 to 2008 and then a second period with relatively higher CPUE values from 2009 to 2015.

    Factors influencing observed trends: The fluctuations in jellyfish biomass and their impacts on forage fish, juvenile Walleye pollock (Gadus chalcogrammus), and salmon in relation to other 105 biophysical indices were investigated by Cieciel et al. (2009) and Brodeur et al. (2002, 2008). Ice cover, sea-surface temperatures in the spring and summer, and wind mixing all have been shown to influence jellyfish biomass and affect jellyfish sensitivity to prey availability (Brodeur et al., 2008).

    Implications: Jellyfish are an important predator and prey. Large jellyfish blooms can impact survival of juvenile and forage fishes. Monitoring fluctuations in jellyfish abundance is important for understanding ecological impacts to juvenile and forage fishes and higher trophic levels.

    Contributed by Robert Lauth
    Resource Assessment and Conservation Engineering Division, Alaska Fisheries Science Center,
    National Marine Fisheries Service, NOAA
    Contact: bob.lauth@noaa.gov
    Last updated: October 2017

  • Description of indicator: Pelagic jellyfish were sampled using a trawl net towed in the upper 20 m of the eastern Bering Sea during the Alaska Fisheries Science Center’s Bering Arctic Subarctic Integrated Surveys (BASIS) during late summer, 2004–2016. Stations were approximately 30 nautical miles apart and a trawl was towed for approximately 30 minutes. Area swept was estimated from horizontal net opening and distance towed.
    Jellyfish catch was estimated in kilograms. Surveys were not conducted in the south 60oN) during 2013 and 2015 and north (≥60oN) during 2008; jellyfish densities in these areas were estimated using geostatistical modeling methods (Thorson et al., 2015). All jellyfish medusae caught in the surface trawl (top 18–20 m of the water column) were sorted by species and subsampled for bell diameter and wet weight. Six species are commonly caught with the surface trawl: Aequorea sp., Chrysaora melanaster, Cyanea capillata, Aurelia labiata, Phacellocephora camtschatica, and Staurophora mertensi. Biomass is calculated for each species and compared across species and oceanographic domains on the Bering Sea shelf.
    Abundance and distribution (center of gravity and area occupied) were estimated for each jellyfish species using the VAST package for multispecies version 1.1.0 (Thorson et al., 2015; Thorson and Kristensen, 2016; Thorson et al., 2016a,b) in RStudio version 1.0.136 and R software version 3.3.0 (R Development Core Team, 2016). The abundance index is a standardized geostatistical index developed by Thorson et al. (2015); Thorson and Kristensen (2016); Thorson et al. (2016a,b) to estimate indices of abundance for stock assessments. We specified a gamma distribution and estimated spatial and spatio-temporal variation for both encounter probability and positive catch rate components.

    Status and trends: Temporal trends in the estimated abundance of jellyfish indicated an increase of smaller sized jellyfish (Aequorea, Aurelia, and Cyanea) and a decrease in the larger jellyfish (Chrysaora) in the eastern Bering Sea during 2016 (Table 4). Starting in 2014, notable increases in jellyfish species composition were observed for all taxa except Chrysaora and continued through 2016. The larger jellyfish was typically more abundant during the 2007–2013 cold stanza, while the smaller sized species were more abundant during the 2004–2006 and 2014–2016 warm stanzas, with the exception of the 2014 warm year. In 2016, Aurelia exceeded the typically most abundant Chrysaora (Table 4).
    The distribution of jellyfish varied among species and years. Yearly distributions throughout the sample grid for all species have been patchy and highly variable (example plots shown for Aequorea [Figure 50] and Aurelia [Figure 51]). Despite uneven distributions throughout oceanographic domains, highest concentrations of all species were found to occur in the middle domain. Center of gravity plots indicate no warm and cold year trend in the distribution of jellyfish (plots not shown). Area occupied was higher for all species during 2016 than the long-term average (Figure 52), except for Aurelia. Aequorea and Aurelia were the only species with a trend of an expanded distribution during warm years and contracted distribution during cold years, with the exception of 2016 for Aurelia (Figures 50, 51, and 52).

    Factors causing observed trends: Shifts in abundance of single large sized jellyfish in cold years to multiple smaller sized species in warm years indicate that there could possibly be a shift to multiple taxa present in the future during warm stanzas. The cause for the shifts in biomass and distribution do not seem to rely solely on physical ocean factors (temperature and salinity). These shifts could also be a result of environmental forcing earlier in the growing season or during an earlier life history stage (polyp), which may influence large medusae biomasses and abundances (Purcell et al., 2009).

    Implications: Significant increases in jellyfish biomass may redirect energy pathways in the eastern Bering Sea food web through jellyfish predation on zooplankton and larval fish, and could result in limited carbon transfer to higher trophic levels (Condon et al., 2011).

    Insert table 4

    Insert figure 50

    Insert figure 51

    Insert figure 52

    Contributed by Kristen Cieciel and Ellen Yasumiishi
    Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: kristin.cieciel@noaa.gov
    Last updated: August 2017

  • There are no updates to Ichthyoplankton indicators in this year’s report. See the contribution archive for previous indicator submissions at: http://access.afsc.noaa.gov/reem/ecoweb/ index.php

  • Description of indicator: Pelagic fish were sampled using a trawl net towed in the upper 20 m of the eastern Bering Sea during the Alaska Fisheries Science Center’s Bering Arctic Subarctic Integrated Surveys (BASIS) during late summer, 2002–2016. Stations were approximately 30 nautical miles apart and a trawl was towed for approximately 30 minutes. Area swept was estimated from horizontal net opening and distance towed.
    Fish catch was estimated in kilograms. Surveys were not conducted in the south (60oN) during 2013 and 2015 and north (≥60oN) during 2008 but fish densities in these areas were estimated using geostatistical modeling methods (Thorson et al., 2015). Four forage fish are commonly captured in the trawl: Capelin (Mallotus villosus), Pacific herring (Clupea pallasii), Sand lance (Ammodytes hexapterus), and Pacific sandfish (Trichodon trichodon).
    Abundance and distribution (center of gravity and area occupied) were estimated using the VAST package for multispecies version 1.1.0 (Thorson et al., 2015; Thorson and Kristensen, 2016; Thorson et al., 2016a,b) in RStudio version 1.0.136 and R software version 3.3.0 (R Development Core Team, 2016). The abundance index is a standardized geostatistical index developed by Thorson et al. to estimate indices of abundance for stock assessments. We specified a gamma distribution and estimated spatial and spatio-temporal variation for both encounter probability and positive catch rate components.

    Status and trends: Temporal trends in the estimated abundance of these forage fish species indicate a decline during 2016 (Table 5). Pacific herring were the most abundant species followed Capelin, Pacific sandfish, and Sand lance. Trends in abundance did not track the recent warm (2002–2005, 2014–2016) and cold (2007–2013) years.
    The distribution of forage fish in pelagic waters varied among species and years. Capelin were distributed on the central and northern Bering Sea shelf (Figure 53). Pacific herring were distributed in the northeastern Bering Sea middle and inner domains (0–100 m bottom depth) (Figure 54). Sand lance were captured primarily in the inner domain of the eastern Bering Sea shelf (plot not shown), while Pacific sandfish were distributed on the southeastern Bering Sea shelf (Figure 55).
    Center of gravity indicated that Pacific sandfish was distributed farther west during warm stanzas (2002–2005 and 2014–2016) and farther east during the cold stanza (2008–2013). No warm and cold year trend in the latitudinal or longitudinal distribution were observed in the distribution of Capelin, Pacific herring, or Sand lance in the survey area (center of gravity plots not shown). Area occupied indicated that these fish did not expand or contract their ranges during warm years relative to cold years (plots not shown).

    Insert table 5

    Factors influencing observed trends: Forage fish had lower abundances during 2016, the third consecutive warm year, indicating poor environmental conditions for the growth and survival of forage fish in the eastern Bering Sea. However, over the 15 year time series, trends in the abundances of forage fish did not coincide with warm or cold conditions.

    Implications: Recent declines in the abundance of forage fish in pelagic waters during late summer implies poor conditions for growth and survival of pelagic fish species in our survey area during August and September. Lower forage fish abundance may impact the feeding and survival of birds, fish, and marine mammals that rely on them for prey.

    Figure 53

    Figure 54

    Figure 55

    Contributed by Ellen Yasumiishi, Kristin Cieciel, Alex Andrews, and Elizabeth Siddon
    Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: ellen.yasumiishi@noaa.gov
    Last updated: August 2017

  • Juvenile Chinook Salmon Abundance in the Northern Bering Sea

    Description of indicator: An index of juvenile Chinook salmon (Oncorhynchus tshawytscha) abundance was constructed for the Canadian-origin (Upper Yukon) stock group of the Yukon River, 2003–2017. Juvenile (first year at sea) abundance is estimated during late-summer (typically September) during surface trawl and oceanographic surveys in the northern Bering Sea. Estimates are based on trawl CPUE data, estimates of genetic stock composition, and mixed layer depth. Abundance for the Canadian-origin stock group have ranged from 0.7 million to 2.9 million juveniles with an overall average of 1.7 million juvenile Chinook salmon from 2003–2017 (Figure 56).

    Insert figure 56

    Status and trends: Abundance estimates in 2017 are preliminary and are based on average stock composition and mixed layer depth corrections; final estimates will be available in the spring of 2018. The preliminary estimate of Canadian-origin juvenile Chinook salmon in the northern Bering Sea in 2017 is 1.3 million juveniles, which is below the overall average of 1.7 million.

    Factors influencing observed trends: Changes in the early life-history (freshwater and early marine) survival, as indicated by the number of juveniles-per-spawner (Figure 57), is the primary factor impacting juvenile abundance in the northern Bering Sea. Preliminary estimates of juvenilesper-spawner in 2017 is the lowest we have observed since 2003. The number of spawning adults is also an important contributing factor to the number of juveniles present in the northern Bering Sea.

    Insert Figure 57

    Implications: Juvenile abundance is significantly correlated (r = 0.87, p less than 0.001) (Figure 58) with adult returns, indicating that much of the year-to-year variability in survival of Yukon River Chinook salmon occurs during their early life stages (freshwater and initial marine). The Canadianorigin stock group of Chinook salmon is the largest stock group of Chinook salmon in the Yukon River and has a complex management framework, directed by both domestic and international (US/Canada) management policies and decisions. Juvenile abundance data are used to assist these 119 pre-season fisheries management decisions in the Yukon River. Juvenile Chinook salmon abundance also has important implications for abundance-based bycatch caps for Chinook salmon in the eastern Bering Sea Walleye pollock (Gadus chalcogrammus) fishery, as low juvenile abundance increases the probability of reduced bycatch caps three to four years in the future.

    Figure 58

     

    Contributed by Jim Murphy1 and Kathrine Howard2
    1Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    2Alaska Department of Fish and Game
    Contact: jim.murphy@noaa.gov
    Last updated: October 2017

  • Description of indicator: Juvenile Pacific salmon were sampled using a trawl net towed in the upper 20 m of the eastern Bering Sea during the Alaska Fisheries Science Center’s Bering Arctic Subarctic Integrated Surveys (BASIS) during late summer, 2002–2016. Stations were approximately 30 nautical miles apart and a trawl was towed for approximately 30 minutes. Area swept was estimated from horizontal net opening and distance towed.
    Fish catch was estimated in kilograms. Surveys were not conducted in the south (less than 60oN) during 2013 and 2015 and north (≥60oN) during 2008 but fish densities in these areas were estimated using geostatistical modeling methods (Thorson et al., 2015). As juveniles during the first year at sea, four of the five salmon species were commonly captured in the trawl: Chinook salmon (Oncorhynchus tshawytscha), chum salmon (O. keta), pink salmon (O. gorbuscha), and sockeye salmon (O. nerka). Abundance and distribution (center of gravity and area occupied) were estimated using the VAST package for multispecies version 1.1.0 (Thorson et al., 2015; Thorson and Kristensen, 2016; Thorson et al., 2016a,b) in RStudio version 1.0.136 and R software version 3.3.0 (R Development Core Team, 2016). The abundance index is a standardized geostatistical index developed by Thorson et al. to estimate indices of abundance for stock assessments. We specified a gamma distribution and estimated spatial and spatio-temporal variation for both encounter probability and positive catch rate components.

    Status and trends: Temporal trends in the estimated abundance of juvenile salmon indicated a recent increase in the eastern Bering Sea (Table 6). Juvenile sockeye were the most abundant species followed pink, chum, and Chinook salmon. Both juvenile pink and sockeye salmon had an alternating year pattern with higher abundances in even-numbered years. Juvenile salmon were typically more abundant during warm years (2002–2005 and 2014–2016) than during cold years (2007–2013), with the exception of higher juvenile pink and chum salmon abundances during 2007 and 2009.
    The distribution of juvenile salmon varied among species and years. Chinook were concentrated in the inner domain ( less than 50m) of the north and southeastern Bering Sea indicating an origin of Norton Sound (Yukon River) in the north and the Kuskokwim River in the south. Chum salmon were most abundant around Nunivak Island (60oN) and were likely from the Kuskokwim River. Sockeye salmon were abundant in the south indicating primarily Bristol Bay origin. Center of gravity indicated that juvenile Chinook, chum, and pink salmon were farther south during warm years, while juvenile sockeye salmon were distributed farther north and west in warm years (Figure 59). Area occupied indicated that all juvenile salmon species expanded their distribution in 2016 relative to 2015, except for pink salmon (Figure 60). Juvenile sockeye and Chinook salmon were the only species that occupied a smaller area during cold years and a larger area in warm years.

    Factors influencing observed trends: Higher abundances of juvenile salmon during recent warm years indicate improved environmental conditions for the survival in the eastern Bering Sea during summer and/or in freshwater rivers and lakes of western Alaska. Juvenile sockeye salmon responded to warming with an expansion in their range and a distribution farther north. The northern-origin juveniles distributed farther south in warm years, possibly in search of food such as age-0 Walleye pollock (Gadus chalcogrammus) during years with low abundances of large zooplankton (Coyle et al., 2011).

    Implications: Recent increases in the abundance of juvenile salmon in our survey area during later summer implies improved conditions for growth and survival of salmon from western Alaska lakes and rivers and/or a change in the distribution of juvenile salmon into our survey area during August and September. Juvenile indices may be an early indication for the numbers of returning adults to the region of origin.

    Insert table 6

    Figure 59

    Figure 60

    Contributed by Ellen Yasumiishi, Kristin Cieciel, Jim Murphy, Alex Andrews, and Elizabeth Siddon
    Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: ellen.yasumiishi@noaa.gov
    Last updated: August 2017

  • Spatial and Temporal Trends in the Abundance and Distribution of Groundfish in the Eastern Bering Sea During Late Summer, 2002–2016

    Description of indicator: Groundfish were sampled using a trawl net towed in the upper 20 m of the eastern Bering Sea during the Alaska Fisheries Science Center’s Bering Arctic Subarctic Integrated Surveys (BASIS) during late summer, 2002–2016. Stations were approximately 30 nautical miles apart and a trawl was towed for approximately 30 minutes. Area swept was estimated from horizontal net opening and distance towed.
    Fish catch was estimated in kilograms. Surveys were not conducted in the south (less than60oN) during 2013 and 2015 and north (≥60oN) during 2008 but fish densities in these areas were estimated using geostatistical modeling methods (Thorson et al., 2015). Four species were commonly caught with the surface trawl: age-0 Pacific cod (Gadus macrocephalus), age-0 Walleye pollock (Gadus chalcogrammus), Atka mackerel (Pleurogrammus monopterygius), and Yellowfin sole (Limanda aspera).
    Abundance and distribution (center of gravity and area occupied) were estimated using the VAST package for multispecies version 1.1.0 (Thorson et al., 2015; Thorson and Kristensen, 2016; Thorson et al., 2016a,b) in RStudio version 0.99.896 and R software version 3.3.0 (R Development Core Team, 2016). The abundance index is a standardized geostatistical index developed by Thorson et al. to estimate indices of abundance for stock assessments. We specified a gamma distribution and estimated spatial and spatio-temporal variation for both encounter probability and positive catch rate components.

    Status and trends: Temporal trends in the estimated abundance of these groundfish species indicated a decline in 2016 (Table 7). Age-0 pollock were the most abundant groundfish species in the survey area followed by Yellowfin sole, Atka mackerel, and then Pacific cod.
    The distribution of groundfish in pelagic waters varied among species and years. Age-0 Pacific cod were distributed on the southern Bering Sea shelf near Unimak Pass (Figure 61). Age-0 pollock were the most widely distributed species; they were primarily in the middle domain of the southeastern shelf, but distributed farther north during warm years (Figure 62). Atka mackerel were captured primarily in the outer domain of the southeastern Bering Sea shelf (plots not shown). Yellowfin sole were distributed in the southern inner and middle domains (plots not shown).
    Temporal trends in the distribution (center of gravity) indicated that age-0 pollock were distributed farther north during recent warm years. No warm and cold year trend was observed in the distribution of age-0 Pacific cod or Yellowfin sole. Atka mackerel were generally distributed farther north during warm stanzas and farther south during the cold stanza (plots not shown). Area occupied (plots not shown) indicated that age-0 pollock had an expanded range during warm years relative to cold years (Figure 62).

    Insert table 7

    Figure 61

    Factors influencing observed trends: Lower abundances of groundfish in pelagic waters during 2016, the third consecutive warm year, indicate poor environmental conditions for the growth and survival in the eastern Bering Sea during summer or movement out of the survey area. Age-0 pollock appeared to respond to warming with an expansion in their range and a distribution farther north. Movement of age-0 pollock and Atka mackerel farther north during warm years indicate a response to warming by changing their distribution.

    Implications: Lower abundances of groundfish in surface waters during 2016 indicate a change in productivity of pelagic waters. Age-0 pollock were distributed primarily in the southeastern Bering Sea middle domain, but were farther north during warm years with higher population densities. This is possibly a response to a search for prey during years of low lipid-rich taxa (e.g., Calanus spp.; Coyle et al. (2011)).

    Figure 62

    Contributed by Ellen Yasumiishi, Kristin Cieciel, Alex Andrews, and Elizabeth Siddon
    Auke Bay Laboratories, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: ellen.yasumiishi@noaa.gov
    Last updated: August 2017

  • Description of indicator: Length-weight residuals are an indicator of somatic growth (Brodeur et al., 2004) and, therefore, a measure of fish condition. Fish condition is an indicator of how heavy a fish is per unit body length, and may be an indicator of ecosystem productivity. Positive length-weight residuals indicate fish are in better condition (i.e., heavier per unit length) whereas negative residuals indicate fish are in poorer condition (i.e., lighter per unit length). Fish condition may affect fish growth and subsequent survival (Paul et al., 1997; Boldt and Haldorson, 2004).
    The AFSC eastern Bering Sea shelf bottom trawl survey data were utilized to acquire lengths and weights of individual fish for Walleye pollock (Gadus chalcogrammus), Pacific cod (Gadus macrocephalus), Arrowtooth flounder (Atheresthes stomias), Yellowfin sole (Limanda aspera), Flathead sole (Hippoglossoides elassodon), Northern rock sole (Lepidopsetta polyxystra), and Alaska plaice (Pleuronectes quadrituberculatus). Only summer standard survey strata and stations were included in analyses, no corner stations were included (Figure 63). Survey strata 31 and 32 were combined as stratum 30; strata 61 and 62 were combined as stratum 60; strata 41, 42, and 43 were combined as stratum 40. Strata 82 and 90 were excluded from analyses because they are not standard survey strata.
    Length-weight relationships for each of the seven species were estimated with a linear regression of log-transformed values over all years where data were available (during 1982–2013). Additionally, length-weight relationships for age-1+ pollock (length from 100–250 mm) were also calculated independent from the adult life history stages. Predicted log-transformed weights were calculated and subtracted from measured log-transformed weights to calculate residuals for each fish. Lengthweight residuals were averaged for the entire EBS and for the 6 strata sampled in the standard summer survey. Temporal and spatial patterns in residuals were examined.

    Figure 63

    Status and trends: Length-weight residuals have varied over time for all species with a few notable patterns (Figure 64). Residuals for all species where there were data were negative in 1999, a cold year in the Bering Sea. Residuals became positive or more positive in 2002 for five of the seven species examined. Flatfish residuals were generally positive from 2002 to 2004 or 2005 depending on species. Age-1 pollock and Pacific cod residuals were positive from 2001 to 2004 or 2005. In 2008, all species except Flathead sole and pollock had negative residuals. There has been a distinct negative trend in Pacific cod since a peak value in 2003. Condition of pollock older than age 1 in 2017 was the second lowest on record. Age-1 pollock and older pollock were not well correlated in most years. Length-weight residuals for all species were less in 2017 than in 2016 indicating poorer condition in the most recent year (Arrowtooth flounder was the only exception).
    Spatial trends in residuals were also apparent for some species. Generally, fish were in better condition on the outer shelf (strata 50 and 60; Figure 65). For all species except Yellowfin sole (which did not occur in outer shelf strata), residuals were almost always positive on the northern outer shelf (stratum 60; Figure 65). For Yellowfin sole, residuals were positive in the outermost shelf strata in which they occurred (stratum 40) except in 1999. In addition to having positive residuals on the outer shelf, gadids tended to have negative residuals on the inner shelf (Figure 65). Pollock residuals were generally positive in strata 50 and 60 and negative in strata 10, 20, and 40. Cod residuals were generally positive in stratum 60 and negative in strata 10 and 20. Spatial patterns in flatfish residuals were also apparent but varied among species. Alaska plaice residuals were almost always negative in stratum 40. Flathead sole residuals were often positive in strata 40 (Figure 64).

    Factors influencing observed trends: One potential factor causing the observed temporal variability in length-weight residuals is temperature. The year 1999 was a particularly cold year in the Bering Sea and also a year of negative length-weight residuals for all groundfish examined (where data existed). Despite the abundant large crustacean zooplankton and relatively high microzooplankton productivity present in 1999 (Hunt et al., 2008) the spatial distribution of some groundfish species is affected by temperatures and a cold year may, therefore, have affected the spatial overlap of fish and their prey. Cold temperatures may have also affected fish energy requirements and prey productivity. Conversely, the continuing warmer than normal 2016 temperatures across the Bering Sea shelf may have resulted in negative trends for length-weight residuals.
    Other factors that could affect length-weight residuals include survey sampling timing and fish migration. The date of the first length-weight data collected annually varied from late May to early June (except 1998, where the first data available were collected in late July). Also, the bottom trawl survey is conducted throughout the summer months, and as the summer progresses, we would expect fish condition to improve. Since the survey begins on the inner shelf and progresses to the outer shelf, the higher fish condition observed on the outer shelf may be due to the fact that they are sampled later in the summer. We also expect that some fish will undergo seasonal and, for some species, ontogenetic migrations through the survey months. For example, seasonal migrations of pollock occur from overwintering areas along the outer shelf to shallow waters (90–140 m) for spawning (Witherell, 2000). Pacific cod concentrate on the shelf edge and upper slope (100–250 m) in the winter, and move to shallower waters (generally less than 100 m) in the summer (Witherell, 2000). Arrowtooth flounder are distributed throughout the continental shelf until age 4, then, at older ages, disperse to occupy both the shelf and the slope (Witherell, 2000). Flathead sole overwinteralong the outer shelf, and move to shallower waters (20–180 m) in the spring (Witherell, 2000). Yellowfin sole concentrate on the outer shelf in the winter, and move to very shallow waters (less than30 m) to spawn and feed in the summer (Witherell, 2000). How these migrations affect the length-weight residuals is unknown at this time.

    Figure 64

    Figure 65

    Implications: A fish’s condition may have implications for its survival. For example, in Prince William Sound, the condition of Pacific herring prior to the winter may in part determine their survival (Paul and Paul, 1998). The condition of Bering Sea groundfish may therefore partially contribute to their survival and recruitment. In the future, as years are added to the time series, the relationship between length-weight residuals and subsequent survival can be examined further. It is likely, however, that the relationship is more complex than a simple correlation. Also important to consider is the fact that condition of all sizes of fish were examined and used to predict survival.
    Perhaps, it would be better to examine the condition of juvenile fish, not yet recruited to the fishery, or the condition of adult fish and correlations with survival. This work has not yet been done for the 2017 bottom trawl survey data, but we are preparing a manuscript describing the juvenile-adult condition correlation and further splitting of juvenile and adult fishes and anticipate including it in the 2018 Ecosystem Status Report.
    The reduced condition for all species in 2017 compared to 2016 is a potential cause for concern and may be a leading indicator of poor overwinter survival and the potential for smaller stocks in 2018. It should be noted anecdotally that the commercial fishery was finding pollock in poorer condition during the summer season as well.

    Contributed by Jennifer Boldt1 , Chris Rooper2 , and Jerry Hoff2 1Fisheries and Oceans Canada, Pacific Biological Station, 3190 Hammond Bay Rd, Nanaimo, BC, Canada V9T 6N7
    2Resource Assessment and Conservation Engineering Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: chris.rooper@noaa.gov
    Last updated: October 2017

  • Description of indicator: We report trends in age-1 total mortality for Walleye pollock (Gadus chalcogrammus), Pacific cod (Gadus macrocephalus), and Arrowtooth flounder (Atheresthes stomias) from the eastern Bering Sea. Total mortality rates are based on residual mortality inputs (M1) and model estimates of annual predation mortality (M2) produced from the multi-species statistical catch-at-age assessment model known as CEATTLE (Climate-Enhanced, Age-based model with Temperature-specific Trophic Linkages and Energetics). See Appendix 1 of the BSAI Walleye pollock stock assessment (Ianelli et al., 2017), Holsman et al. (2016), Holsman and Aydin (2015), Ianelli et al. (2015), and Jurado-Molina et al. (2005) for more information.

    Status and trends: Estimated age-1 natural mortality (i.e., M1+M2) for pollock, Pacific cod, and Arrowtooth flounder peaked in 2016 and, for the third year in a row, remained elevated in 2017 at levels above those observed since the late 1980’s (Figure 66). At 1.69 yr-1 age-1 mortality estimated by the model was greatest for pollock relative to Pacific cod or Arrowtooth flounder. Age1 mortality was lower for Pacific cod and Arrowtooth flounder, with total age-1 natural mortality stable at around 0.69 and 0.65 yr-1, respectively, although both were above long-term means in 2015–2017

    Factors influencing observed trends: Temporal patterns in natural mortality reflect annuallyvarying changes in predation mortality that primarily impact age-1 fish (but also impact ages 2 and 3 fish in the model). Pollock are primarily consumed by older conspecifics, and pollock cannibalism accounted for 55% (on average) of total predation mortality for age-1 pollock except for 2006–2008 when predation by Arrowtooth flounder exceeded cannibalism as the largest source of predation mortality of age-1 pollock (Figure 67).
    Combined annual predation demand (annual ration) of pollock, Pacific cod, and Arrowtooth flounder in 2017 was 6.07 million t, down slightly from the 7.65 million t annual average during the warm years of 2014–2016. Pollock represent approximately 78% of the model estimates of combined prey consumed with 4.84 consumed annually by all three predators in the model (Figure 68).

    Figure 66

    Implications: We find evidence for recent elevated rates of predation mortality on age-1 pollock, Pacific cod, and Arrowtooth flounder. This pattern may reflect higher metabolic (and energetic) demand of predators under warm conditions combined with maturing large 2010–2012 age classes of pollock and Pacific cod that have increased predator demand in the eastern Bering Sea (Holsman and Aydin, 2015; Spencer et al., In press; Hunsicker et al., 2013; Zador et al., 2011). This pattern may also explain low model estimates of recruitment of eastern Bering Sea pollock and Pacific cod in recent years.
    Between 1980 and 1993, the relatively high natural mortality rates reflect patterns in combined annual demand for prey by all three predators that was highest in the mid 1980’s (collectively 7.84 million t per year), and in recent years (collectively 7.25 million t per year). The peak in predation mortality of age-1 pollock in 2006 corresponds to the maturation of a large age class of 5–7 year old pollock and 2 year old Pacific cod that dominated the age composition of the two species in 2006. Similarly, the recent peaks in mortality in 2016 reflect maturation of the large 2012 year class of pollock.

    Figure 67

    Figure 68

    Contributed by Kirstin Holsman, Jim Ianelli, and Kerim Aydin
    Resource Ecology and Fishery Management Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: kirstin.holsman@noaa.gov
    Last updated: October 2017

  • Age-0 Recruitment of Pacific Cod (Gadus macrocephalus) in the Eastern Bering Sea as Predicted by the Average of the North Pacific Index from October through December

    Description of indicator: The North Pacific Index (NPI) was developed by Trenberth and Hurrell (1994), and represents the area-weighted sea level pressure over the region 30oN–65oN, 160oE– 140oW. Monthly values of the NPI since January 1899 are reported at https://climatedataguide. ucar.edu/sites/default/files/npindex_monthly.txt. Specifically, the indicator used in this analysis is the average of the monthly NPI values from October–December in each year.
    In the 2012 assessment of the eastern Bering Sea (EBS) stock of Pacific cod (Gadus macrocephalus) (Thompson and Lauth, 2012), annual log-scale recruitment deviations (from the mean) estimated by the assessment model were regressed against each of several environmental indices summarized by Zador et al. (2011). The highest univariate correlation was obtained for the spring–summer NPI. Further investigations were conducted with monthly NPI data from the website referenced above. The best univariate model obtained in the 2012 analysis was a linear regression of recruitment deviations from 1977–2011 against the October–December average NPI (from the same year). Vestfals et al. (2014) also noted a positive correlation between Pacific cod recruitment and the NPI, although not the October–December average NPI in particular.

    Figure 69

    Status and trends: In the 2016 assessment of the EBS stock of Pacific cod (Thompson and Lauth, 2016), the 2015 average October–December NPI was reported as being barely positive (z-score = 0.018). The 1977–2015 time series is shown in Figure 69. The trend depends on the range of years considered. If the regression starts in 2014, the trend is positive; if the regression starts anywhere from 2006 through 2013, the trend is negative; and if the regression starts anywhere from 1991 through 2005, the trend is positive.
    In each assessment since 2012, the regression analysis has been updated. The regression in the 2016 assessment (Thompson and Lauth, 2016) resulted in a correlation of 0.55 (R2=0.30). The time series, regression line, and 95% confidence interval from the 2016 regression are shown in Figure 70. According to this regression, the probability of the 2015 year class being higher than the median for the time series is 51%. However, the datum for 2015 (magenta diamond in Figure 70) falls quite a bit below the predicted value from the regression. This marks the first time in the last 11 years (cohorts) that the sign of the difference from the mean estimated by the assessment model differs from the sign predicted by the regression (although the difference from the mean for 2015 predicted by the regression is extremely close to zero [0.014]).

    Figure 70

    Factors influencing observed trends: Two years, 1990 and 2002 (green and yellow diamonds in Figure 70), turned out to be far more influential than any other year in determining the magnitude of the estimated slope, and both of these influences were negative. Therefore, the positive slope is not due to the influence of outliers; if anything, the outliers are making the relationship appear less strong than would be the case without them. Circulation patterns over the EBS shelf vary with large-scale climate drivers such as the Pacific Decadal Oscillation (see p. 68). The strength of the Bering Slope Current is correlated with the NPI (Vestfals et al., 2014) with higher NPI values related to weaker along-shelf transport. The positive relationship between the NPI and Pacific cod recruitment may indicate that weaker circulation leads to better retention of age-0 fish in suitable nursery habitats (Vestfals et al., 2014).

    Implications: Potential uses of the estimated relationship in the context of fishery management are: 1) as an independent means of corroborating initial estimates of year class strength, which are not made until the year class reaches age-1 (the first age at which the fish show up in the EBS shelf bottom trawl survey); 2) as a determinant of year class strength within the stock assessment model itself; and 3) in the event that the average October–December NPI can be forecast into the future, as a means of forecasting future year class strengths.

    Contributed by Grant Thompson
    Resource Ecology and Fisheries Management, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: grant.thompson@noaa.gov
    Last updated: October 2017

  • Description of indicator: “Miscellaneous” species fall into three groups: eelpouts (Zoarcidae), poachers (Agonidae), and sea stars (Asteroidea). The three species comprising the eelpout group are the wattled eelpout (Lycodes palearis) and shortfin eelpout (L. brevipes) and to a lesser extent the marbled eelpout (L. raridens). The biomass of poachers is dominated by a single species, the sturgeon poacher (Podothecus acipenserinus) and to a lesser extent the sawback poacher (Leptagonus frenatus). The composition of sea stars in shelf trawl catches are dominated by the purple-orange sea star (Asterias amurensis), which is found primarily in the inner/middle shelf regions, and the common mud star (Ctenodiscus crispatus), which is primarily an inhabitant of the outer shelf. Relative CPUE by weight was calculated and plotted for each species or species group by year for 1982–2017. Relative CPUE was calculated by setting the largest biomass in the time series to a value of 1 and scaling other annual values proportionally. The standard error (±1) was weighted proportionally to the CPUE to produce a relative standard error.

    Status and trends: The 2017 relative CPUE for eelpouts decreased by 15% from 2016, but was still among the highest estimates over the last 11 years. The poacher group CPUE decreased by 30% since 2016 and by 48% since 2015. The 2017 poacher estimate ranked as the lowest since 2001. Only during a single 3-year time period from 1984 to 1986 were poacher estimates significantly lower. The sea stars as a group increased by 21% from 2016 to 2017, and the 2017 CPUE ranked as the second highest since 1982 (Figure 81).

    Factors causing observed trends: Determining whether these trends represent real responses to environmental change or are simply an artifact of standardized survey sampling methodology (e.g., temperature dependent catchability) will require more specific research on survey trawl gear selectivity relative to interannual differences in bottom temperatures and on the life history characteristics of these epibenthic species.

    Implications: Eelpouts have important roles in the energy flow within benthic communities. For example, eelpouts are a common prey item of Arrowtooth flounder (Atheresthes stomias). However, it is not known at present whether these changes in CPUE are related to changes in energy flow.

    Figure 81

    Contributed by Robert Lauth
    Resource Assessment and Conservation Engineering Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: bob.lauth@noaa.gov
    Last updated: October 2017

  • Seabird Monitoring Summary from Alaska Maritime National Wildlife Refuge

    Description of indicator: The Alaska Maritime National Wildlife Refuge has monitored seabirds at colonies around Alaska in most years since the early- to mid-1970’s. Time series of annual breeding success and phenology (among other parameters) are available from over a dozen species at eight Refuge sites in the Gulf of Alaska, Aleutian Islands, and Bering and Chukchi Seas. Monitored colonies in the eastern Bering Sea include St. Paul and St. George Islands. Here, we focus on cliff-nesting, primarily fish-eating species: black-legged kittiwake (Rissa tridactyla), red-legged kittwake (R. brevirostris), common murre (Uria aalge), thick-billed murre (U. lomvia), and redfaced cormorants (Phalacrocorax urile). Reproductive success is defined as the proportion of nest sites with eggs (or just eggs for murres that do not build nests) that fledged a chick.

    Status and trends: Cliff-nesting seabirds showed overall poor reproductive success in 2017 at both islands, with the exception of nearshore-feeding red-faced cormorants (Figure 84). This was the third consecutive year of poor reproduction for both black-legged and red-legged kittiwakes. This was the second year of poor reproduction for both common and thick-billed murres; common murres at St. George had some reproductive success (0.33) but fewer than normal birds showed up to breed. Mean hatching dates were late across the board for those species that hatched any chicks.

    Factors influencing observed trends: In general, these species appear to have had negative responses to the marine heat wave in the Northeast Pacific over the past few years, with widespread reproductive failures, die-offs, and low attendance at breeding colonies. Kittiwakes began to fail during the first year of the heatwave in 2015, while the murres did not show negative responses until 2016. This pattern may reflect differences in natural history, with murres able to buffer reproductive success in poor conditions to some degree (Burger and Piatt, 1990). Historically, kittiwakes fail to fledge any chicks about one in six years at the Pribilof Islands, whereas for murres this was unprecedented before 2016.

    Implications: Reproductive activity of central-place foraging seabirds can reflect ecosystem conditions at multiple spatial and temporal scales. For piscivorous species that feed at higher trophic levels, continued reduced reproductive success may indicate that the ecosystem has not yet shifted back from warm conditions and/or there is a lagged response of the prey. Despite environmental changes returning back to more neutral conditions (see p. 63), seabird foraging conditions do not appear to have recovered in the eastern Bering Sea. In contrast, the improvement in attendance and minimal reproductive activity among murres in the Gulf of Alaska during 2017 indicates some improvement in foraging conditions for those species.

    Figure 84

    Contributed by Heather Renner and Marc Romano Alaska Maritime National Wildlife Refuge, 95 Sterling Highway, Suite 1, Homer, AK 99603 Contact: heather renner@fws.gov Last updated: October 2017

  • The Marine Mammal Protection Act requires stock assessment reports to be reviewed annually for stocks designated as strategic, annually for stocks where there is significant new information available, and at least once every 3 years for all other stocks. Each stock assessment includes, when available, a description of the stock’s geographic range, a minimum population estimate, current population trends, current and maximum net productivity rates, optimum sustainable population levels and allowable removal levels, as well as estimates of annual human-caused mortality and serious injury through interactions with commercial fisheries and subsistence hunters. The most recent (2014) Alaska Marine Mammal stock assessment was released in August 2015 and can be downloaded at http://www.nmfs.noaa.gov/pr/sars/region.htm.

    Northern Fur Seal (Callorhinus ursinus) Pup Production in the Bering Sea

    Description of indicator: The northern fur seal (Callorhinus ursinus) ranges throughout the North Pacific Ocean from southern California north to the Bering Sea and west to the Okhotsk Sea and Honshu Island, Japan. Breeding in the U.S. is restricted to only a few sites: the Pribilof Islands and Bogoslof Island in Alaska, and San Miguel and the Farallon Islands off California (Muto et al., 2016). Two separate stocks of northern fur seals are recognized within U.S. waters: an Eastern Pacific stock (Pribilof and Bogoslof Islands) and a California stock.
    Northern fur seals were listed as depleted under the Marine Mammal Protection Act (MMPA) in 1988 because population levels had declined to less than 50% of levels observed in the late 1950s, with no compelling evidence that carrying capacity had changed (NMFS, 2007). Fisheries regulations were implemented in 1994 (50 CFR 679.22(a) (6)) to create a Pribilof Islands Area Habitat Conservation Zone (no fishing with trawl permitted), in part to protect northern fur seals. Under the MMPA, this stock remains listed as “depleted” until population levels reach at least the lower limit of its optimum sustainable population (estimated at 60% of carrying capacity). A Conservation Plan for the northern fur seal was written to delineate reasonable actions to protect the species (NMFS, 2007). Pup production of northern fur seals on Pribilof and Bogoslof Islands is estimated by the Marine Mammal Laboratory biennially using a mark-recapture method (shearsampling) on 1–2 month old pups. The most recent pup production estimate for the Pribilof Islands was conducted during August 2016; pup production on Bogoslof Island was assessed in August 2015.

    Status and trends: We estimated 80,641 (standard error [SE] = 717) pups were born on St. Paul Island and 20,490 (SE = 460) pups were born on St. George Island in 2016. The observed pup mortality rates were 2.7% on St. Paul Island and 1.1% on St. George Island. The total estimated number of pups born on St. Paul Island in 2016 (not including Sea Lion Rock) was 12.1% less than in 2014, which was 5.3% less than in 2012 (Towell et al., In press). On St. George Island there was an 8.2% increase between 2014 and 2016, following a 17.0% increase between 2012 and 2014 (Figure 85).
    Pup production has been declining since 1998 at an average annual rate of 4.12% (SE = 0.40%, P less than 0.01) on St. Paul Island and shows no significant trend (SE = 0.57%, P = 0.13) on St. George Island over the same time period. The overall rate of decline on the Pribilof Islands (excluding Sea Lion Rock) was 3.50% (SE = 0.40%, P less than 0.01) annually from 1998 to 2016.
    Since 2002, pup production has been lower than was estimated in 1921 on St. Paul Island and in 1918 on St. George Island, when the populations were recovering at 8% annually from a pelagic harvest that ended in the early 20th century. On a positive note, St. George Island pup production has shown an increase for two censuses in a row, an increase of 26.6% in 2016 from 2012.

    Figure 85

    Factors influencing observed trends: While overall pup production has declined on the Pribilof Islands, it has increased on Bogoslof Island. The last Bogoslof survey occurred in August 2015 at which time pup production had increased at approximately 10.1% (SE = 1.08) per year since 1997. This rate is faster than what could be expected from a completely closed population of fur seals, indicating that at least some of the increase is due to females moving from the Pribilof Islands (presumably) to Bogoslof Island to give birth and breed. However, recent volcanic activity (December 2017 to September 2017) will likely impact pup production at Bogoslof Island this season. Additionally, declines observed on the Pribilof Islands are much greater than the increase in numbers on Bogoslof Island, indicating that the decline on the Pribilof Islands cannot be due entirely to emigration.

    Implications: Differences in trends between the largely shelf-foraging Pribilof fur seals and the pelagic-foraging Bogoslof fur seals likely reflect differences in their summer foraging success, and are unlikely related to large-scale changes in the North Pacific Ocean (e.g., regime shifts, Pacific Decadal Oscillation), since these populations both occupy the same habitats in the North Pacific Ocean during the fall, winter, and spring.

     

    Contributed by Rod Towell, Rolf Ream, John Bengtson, Michael Williams, and Jeremy Sterling Marine Mammal Laboratory, Alaska Fisheries Science Center, NOAA Contact: rod.towell@noaa.gov Last updated: October 2017

  • Aggregated Catch-Per-Unit-Effort of Fish and Invertebrates in Bottom Trawl Surveys on the Eastern Bering Sea Shelf, 1982–2017

    Description of indicator: The index provides a measure of the overall biomass of demersal and benthic fish and invertebrate species. We obtained catch-per-unit-effort (CPUE in kg ha) of fish and major invertebrate taxa for each successful haul completed during standardized bottom trawl surveys on the eastern Bering Sea shelf (EBS), 1982–2017. Total CPUE for each haul was computed as the sum of the CPUEs of all fish and major invertebrate taxa. To obtain an index of average CPUE by year across the survey region, we modeled log-transformed total CPUE (N = 13,338 hauls) as a smooth function of depth, Julian Day and location (latitude / longitude) with year-specific intercepts using Generalized Additive Models following Mueter and Norcross (2002). Hauls were weighted based on the area represented by each station. The CPUE index does not account for gear or vessel differences, which are confounded with interannual differences and may affect results prior to 1988.

    Status and trends: Total log(CPUE) in the EBS shows an apparent long-term increase from 1982– 2005, followed by a decrease from 2005 to 2009, increased CPUE in 2010–2013, and a substantial increase in 2014 to the highest observed value in the time series (Figure 86). Estimated means prior to 1988 may be biased due to unknown gear effects and because annual differences are confounded with changes in mean sampling date, which varied from as early as June 15 in 1999 to as late as July 16 in 1985. On average, sampling occurred about a week earlier since the 2000s compared to the 1980s.

    Factors influencing observed trends: Commercially harvested species accounted for approximately 95% of survey catches. Fishing is expected to be a major factor determining trends in survey CPUE, but environmental variability is likely to account for a substantial proportion of the observed variability in CPUE through variations in recruitment, growth, and distribution. The increase in survey CPUE in the early 2000s primarily resulted from increased abundances of Walleye pollock (Gadus chalcogrammus) and a number of flatfish species (Arrowtooth flounder, Atheresthes stomias; Yellowfin sole, Limanda aspera; Rock sole, Lepidopsetta bilineata; and Alaska plaice, Pleuronectes quadrituberculatus) due to strong recruitments in the 1990s. Decreases in 2006–2009 and subsequent increases are largely a result of fluctuations in pollock recruitment and abundance. Increases in pollock and Pacific cod biomass in 2010 resulted in the observed increase in log(CPUE). Models including bottom temperature suggest that, in the EBS, CPUE is greatly reduced at low temperatures (less than 1oC) as evident in reduced CPUEs in 1999 and 2006–2009, when the cold pool covered a substantial portion of the shelf. Overall, there is a moderate positive relationship between average bottom temperatures and CPUE in the same year (r = 0.53, p = 0.0089), but not in the following years. The reduction in CPUE during cold periods is likely due to a combination of actual changes in abundance, temperature-dependent changes in catchability of certain species (e.g. flatfish, crab), and changes in distribution as a result of the extensive cold pool displacing species into shallower (e.g., red king crab) or deeper (e.g., Arrowtooth flounder) waters. The increase in total CPUE in the Bering Sea in 2014 was largely due to an increase in pollock catches in the bottom trawl survey. CPUE decreased in 2015 and has remained stable since then

    Figure 86

    Implications: This indicator can help address concerns about maintaining adequate prey for upper trophic level species and other ecosystem components. Relatively stable or increasing trends in the total biomass of demersal fish and invertebrates, together with a relatively constant size composition of commercial species, suggest that the prey base has remained stable over recent decades, but displays substantial fluctuations over time, largely as a result of variability in pollock biomass. Decreasing CPUE in the eastern Bering Sea in the early 2000s was a concern, but biomass has increased as a result of several strong year classes of pollock entering the survey.

    Contributed by Franz Mueter1 and Robert Lauth2
    1University of Alaska Fairbanks, 17101 Point Lena Loop Road, Juneau, AK 99801
    2Resource Ecology and Fisheries Management Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    Contact: fmueter@alaska.edu
    Last updated: October 2017

 

Indicators presented in this section are intended to provide a summary of the status of several ecosystem-scale indicators related to fishing and human economic and social well-being. These indicators are organized around objective categories derived from U.S. legislation and current management practices

  • Time Trends in Groundfish Discards

    Description of indicator: Estimates of groundfish discards for 1993–2002 are sourced from NMFS Alaska Region’s blend data, while estimates for 2003 and later come from the Alaska Region’s Catch Accounting System. These sources, which are based on observer data in combination with industry landing and production reports, provide the best available estimates of groundfish discards. Discard rates as shown here are calculated as the weight of groundfish discards divided by the total (i.e., retained and discarded) catch weight for the relevant area-gear-target sector. Where rates are described below for species or species groups, they represent the total discarded weight of the species/species group divided by the total catch weight of the species/species group for the relevant area-gear-target sector. These estimates include only catch of FMP-managed groundfish species on FMP-managed groundfish targets: not included are groundfish discards in the halibut fishery and discards of non-FMP groundfish species, such as forage fish and species managed under prohibited species catch (PSC) limits.

    Status and trends: Since 1993 discard rates of groundfish species in federally-managed Alaskan groundfish fisheries have generally declined in both pollock and non-pollock trawl fisheries in the Bering Sea (BS). Discard rates in the BS pollock trawl sector declined from 20% to about 1% in 1998 and have remained at or below this level. Rates in the non-pollock trawl sector have declined from a high of 50% in 1994 and have remained below 8% since 2011. Discard rates and volumes in the BS fixed gear sector have been stable relative to trawl sectors but have trended slightly upward since 2012, with the 2016 rate (14.3%) representing the highest annual rate since 1998 and the 2016 discard weight (26.7K metric trons), the highest annual weight over the entire time series (Figure 94).

    Figure 94

    Factors influencing observed trends: Since the early 1990s fisheries managers in North Pacific groundfish fisheries have employed various measures to address the problem of discards, including:
    Limited access privilege programs (LAPP) that reduce economic discards by removing the race for fish In-season closure of fisheries once target or bycatch species quotas are reached Minimum retention and utilization standards for certain fisheries Maximum retainable amounts (MRAs), which specify the amounts of “bycatch only” species that harvesters may retain relative to other groundfish species that remain open to directed fishing. MRAs reduce regulatory discards by allowing for limited retention of species harvested incidentally in directed fisheries.

    In the Bering Sea, management and conservation measures aimed at reducing bycatch have contributed to an overall decline in groundfish discards over time. Pollock roe stripping, wherein harvesters extract only the highest value pollock product and discard all of the remaining fish, was prohibited in 1991. Throughout the 1990s, declines in total catch and discard of non-pollock groundfish in the pollock fishery coincided with the phasing out of bottom trawl gear in favor of pelagic gear, which allows for cleaner pollock catches. Full retention requirements for pollock and Pacific cod were implemented in 1998 for all vessels fishing for groundfish. Pollock discards declined significantly across both trawl gear sectors and have been effectively nonexistent in the trawl pollock fishery since it was rationalized in 2000 and became subject to more comprehensive observer coverage. Between 1997 and 1998 annual discard rates for cod fell from 13% to 1% in the non-pollock trawl sector and from 50% to 3% in the pollock trawl sector.
    Low retention rates in the non-AFA trawl catcher processor (head and gut) fleet prompted adoption of Amendments 79 and 80 to the BSAI Groundfish FMP in 2008. Amendment 79 established a Groundfish Retention Standard (GRS) Program with minimum retention and utilization requirements for vessels at least 125 feet LOA (industry-internal monitoring of retention rates has since replaced the GRS Program). Amendment 80 expanded the GRS program to all vessels in the head and gut fleet and established a cooperative-based LAPP with fixed allocations of certain nonpollock groundfish species. These allocations are intended to eliminate the race for fish and remove the economic incentive to discard less valuable species caught in the multi-species flatfish fishery. Groundfish discard rates in the trawl flatfish fishery fell from 23% to 12% between 2007 and 2008 and have continued on a gradual decline since then.
    Since 2003 across all Bering Sea sectors combined, discard rates for species groups historically managed together as the “other groundfish” assemblage (skate, sculpin, shark, squid, and octopus) have ranged from 65% to 80%, with skates representing the majority of discards by weight. In the fixed gear sector other groundfish typically account for at least 70% of total groundfish discards annually. Fluctuations in discard volumes and rates for these species may be driven by changes in market conditions and in fishing behavior within the directed fisheries in which these species are incidentally caught. For example, low octopus catch from 2007–2010 may be attributable to lower processor demand for food-grade octopus and decreases in cod pot-fishing effort stemming from declines in cod prices.

    Implications: Characterizing fishery bycatch, which includes discards of groundfish, is an important component of ecosystem-based management. Discards add to the total human impact on biomass without providing a benefit to the Nation and as such are seen as “contrary to responsible stewardship and sustainable utilization of marine resources” (Kelleher, 2005). Bycatch in general 178 constrains the utilization of commercial species (resulting in forgone income) and increases the uncertainty around total fishing-related mortality, making it more difficult to assess stocks, define overfishing levels, and monitor fisheries for overfishing. Although ecosystem effects of discards are not fully understood, discards of whole fish and offal have the potential to alter energy flow within ecosystems and have been observed to result in changes to habitat (e.g., oxygen depletion in the benthic environment) and community structure (e.g., increases in scavenger populations).
    Minimizing fishery discards is recognized as an ecological, economic, and moral imperative in various multilateral initiatives and in National Standard 9 of the Magnuson-Stevens Fishery Conservation and Management Act. Over the last three decades, management measures in North Pacific groundfish fisheries have generally been effective in increasing groundfish retention and utilization and reducing discards. Monitoring discards and discard rates provides a way to assess the continuing efficacy of such measures.

    Contributed by Jean Lee
    Resource Ecology and Fisheries Management Division, AFSC, NMFS, NOAA
    Alaska Fisheries Information Network, Pacific States Marine Fisheries Commission
    Contact: jean.lee@noaa.gov
    Last updated: September 2017

  • Area Disturbed by Trawl Fishing Gear in the Eastern Bering Sea

    Description of indicator: Fishing gear can impact habitat used by a fish species for the processes of spawning, breeding, feeding, or growth to maturity. This indicator uses output from the Fishing Effects model to estimate the habitat reduction of geological and biological features over the eastern Bering Sea domain, utilizing spatially explicit VMS data. The time series for this indicator is available since 2003, when widespread VMS data became available.

    Status and trends: Habitat impacts due to fishing gear (pelagic and non-pelagic trawl, longline, and pot) interactions have decreased steadily from a high of about 3.5% between 2003–2008 to the present level of about 2.3% in the eastern Bering Sea (Figures 98 and 99).

    Figure 98

    Factors influencing observed trends: Trends in seafloor area disturbed can be affected by numerous variables, such as fish abundance and distribution, management actions (e.g., closed areas), changes in the structure of the fisheries due to rationalization, increased fishing skills (e.g., increased ability to find fish), markets for fish products, and changes in vessel horsepower and fishing gear.

    Between 2003 and 2008, variability in habitat reduction was driven largely by the seasonality of fishing in the eastern Bering Sea. In 2008, Amendment 80 was implemented, which allocated BSAI Yellowfin sole, Flathead sole, Rock sole, Atka mackerel, and Aleutian Islands Pacific ocean perch to the head and gut trawl catcher processor sector, and allowed qualified vessels to form cooperatives. The formation of cooperatives reduced overall effort in the fleet while maintaining catch levels. In 2010, trawl sweep gear modifications were implemented on non-pelagic trawls in the eastern Bering Sea, resulting in less gear contacting the seafloor and less habitat impact

    Figure 99

    Implications: Habitat impacts vary with the biological and geological characteristics of the areas fished, recovery rates of those biological and geological structures, and management changes that result in spatial redistribution of fishing effort. Although the impacts of fishing across the domain are very low, it is possible that localized impacts may be occurring. The issue of local impacts is an area of ongoing development with the Fishing Effects model. The 2015 EFH 5-year review was completed in 2017, and AFSC stock assessment authors considered habitat impacts to managed species for the first time. In no cases was the effects of fishing on Essential Fish Habitat considered to be more than minimal or not temporary. The EFH 5-year Review Summary Report has been published as a Processed Report and can be found here: ftp://ftp.library.noaa.gov/noaa_ documents.lib/NMFS/TM_NMFS_AFKR/TM_NMFS_FAKR_15.pdf.

    Contributed by John V. Olson
    Habitat Conservation Division, Alaska Regional Office, National Marine Fisheries Service, NOAA
    Contact: john.v.olson@noaa.gov
    Last updated: October 2017

  • Fish Stock Sustainability Index and Status of Groundfish, Crab, Salmon, and Scallop Stocks

    Description of indicator: The Fish Stock Sustainability Index (FSSI) is a performance measure for the sustainability of fish stocks selected for their importance to commercial and recreational fisheries (http://www.nmfs.noaa.gov/sfa/fisheries_eco/status_of_fisheries). The FSSI will increase as overfishing is ended and stocks rebuild to the level that provides maximum sustainable yield. The FSSI is calculated by awarding points for each fish stock based on the following rules: 1. Stock has known status determinations:
    (a) overfishing level is defined = 0.5 (b) overfished level is defined = 0.5 2. Fishing mortality rate is below the “overfishing” level defined for the stock = 1.0 3. Biomass is above the “overfished” level defined for the stock = 1.0 4. Biomass is at or above 80% of the biomass that produces maximum sustainable yield (BMSY) = 1.0 (this point is in addition to the point awarded for being above the “overfished” level) The maximum score for each stock is 4.
    In the Alaska Region, there are 36 FSSI stocks and an overall FSSI of 144 would be achieved if every stock scored the maximum value, 4. Over time, the number of stocks included in the FSSI has changed as stocks have been added and removed from Fishery Management Plans (FMPs). Prior to 2015 there were 35 FSSI stocks and maximum possible score of 140. To keep FSSI scores for Alaska comparable across years we report the total Alaska FSSI as a percentage of the maximum possible score (i.e., 100%). Additionally, there are 29 non-FSSI stocks, two ecosystem component species complexes, and Pacific halibut which are managed under an international agreement. None of the non-FSSI stocks are known to be overfished, approaching an overfished condition, or subject to overfishing. For more information on non-FSSI stocks see the Status of U.S. Fisheries webpage.
    Within the BSAI region there are 22 FSSI stocks. The assessment for sablefish is based on aggregated data from the GOA and BSAI regions. In previous FSSI contributions, the sablefish FSSI score was included among BSAI species. Starting with this year’s contribution, sablefish is removed from the BSAI FSSI contribution and is now included in the GOA FSSI contribution (see the Gulf of Alaska Ecosystem Considerations Report).

    Table 10

    Status and trends: As of June 30, 2017, no BSAI groundfish stock or stock complex is subjected to overfishing, is considered to be overfished, or to be approaching an overfished condition (Table 10). Among BSAI crab stocks, the Pribilof Islands blue king crab stock is considered to be overfished and is subject to overfishing. This stock is in year 3 of a rebuilding plan.
    The current overall Alaska FSSI is 132.5 out of a possible 144, or 92%, based on updates through June 2017 and is unchanged from last year (Figure 100). The overall Alaska FSSI has generally trended upwards from 80% in 2006 to 92% in 2017. The BSAI groundfish FSSI score is 56 out of a maximum possible 56, and BSAI king and tanner crabs are 25.5 out of a possible 32. The overall Bering Sea/Aleutian Islands score is 81.5 out of a maximum possible score of 88 (Table 11). Since 2006 the BSAI overall FSSI has increased from 74% up to 93% in 2017 (Figure 101).

    Figure 100

    Factors influencing observed trends: Although the overall Alaska FSSI score is unchanged from last year, there have been changes in the FSSI scores for one BSAI crab stock and one BSAI groundfish stock. The Pribilof Islands blue king crab stock lost one point when it was determined they were subject to overfishing. The primary driver of decline for this stock is thought to be changes in environmental conditions that negatively affect reproduction. One point was gained when the biomass of the BSAI Greenland halibut stock increased to greater than 80% of BMSY.
    The two point changes offset and the result in no net change to the overall Alaska FSSI. Other crab groups in the BSAI region with FSSI scores less than 4 are golden king crab-Aleutian Islands (FSSI=1.5) and blue king crab-St. Matthew’s Island (FSSI=3). Neither of these king crab stocks are subject to overfishing. It is unknown if the golden king crab-Aleutian Islands stock is overfished and BMSY is not estimated.

    Figure 101

    Implications: The majority of Alaska groundfish fisheries appear to be sustainably managed. A single stock in the BSAI is considered to be overfished and subject to overfishing (Pribilof Islands blue king crab). No other stocks or stock complexes in the BSAI are known to be approaching an overfished condition.

    Table 11

    Contributed by Andy Whitehouse
    Joint Institute for the Study of the Atmosphere and Ocean (JISAO), University of Washington, Seattle, WA
    Contact: andy.whitehouse@noaa.gov
    Last updated: August 2017

  • Economic Indicators in the Eastern Bering Sea Ecosystem - Landings

    Description of indicator: Landings are a baseline metric for characterizing commercial economic production in the eastern Bering Sea. Landings are the retained catch of fish and are plotted here by functional group (Figure 102). While many species comprise a functional group, it is the handful of species that fishermen target that dominate the economic metrics in each group. The primary target species in the apex predators’ functional group are Pacific cod (Gadus macrocephalus), Pacific halibut (Hippoglossus stenolepis), Sablefish (Anoplopoma fimbria), and Arrowtooth flounder (Atheresthes stomias). The primary target species in the pelagic foragers’ functional group are Walleye pollock (Gadus chalcogrammus), Atka mackerel (Pleurogrammus monopterygius), and Pacific ocean perch (Sebastes alutus). The primary target species in the benthic foragers’ functional group are Yellowfin sole (Limanda aspera), Rock sole (Lepidopsetta bilineata), and Flathead sole (Hippoglossoides elassodon). The primary target species in the salmonid functional group are Chinook (Oncorhynchus tshawytscha), sockeye (O. nerka), and pink (O. gorbuscha) salmon. The primary target species in the motile epifauna functional group are king, bairdi, and snow crab. Because of significant differences in the relative scale of landings across functional group, landings are plotted on a log scale (figures based on Fissel et al. (2016)).

    Status and trends: Landings in the eastern Bering Sea are predominantly from the pelagic forager functional group. The primary species landed within this group is pollock whose landings are an order of magnitude larger than that of any other species or functional group. Trends in the landings of the apex predator functional group are primarily driven by TAC levels in the Pacific cod stock which has remained healthy and has remained slightly higher since 2011 than before. Landings were increasing up to 2008 in the flatfish fisheries which make up the benthic foragers functional group. Total flatfish catches are well below their respective TACs and stocks remain healthy. EBS salmon landings have remained largely stable from 2004–2016 with a temporary decline from 2011–2013. Landings in the crab stocks which comprise the motile epifauna group have trended up gradually since 2003 reflecting the increasing health of the stocks following rationalization of the crab fisheries.

    Factors influencing observed trends: Between 2008–2010 conservation-based reductions in the pollock Total Allowable Catch (TAC) resulted in reduced landings for the pelagic forager functional group. In 2008 Amendment 80 to the BSAI groundfish FMP was implemented rationalizing the major flatfish fisheries which resulted in significant reductions in bycatch.
    Total catch of the groundfish that comprise the pelagic forager, apex predators, and benthic foragers’ functional groups in the Bering Sea is capped at 2 million metric tons. The sum of the Allowable Biological Catches (ABC) for these groups are typically above the cap and TACs are reduced from the ABC by the North Pacific Fishery Management Council to meet the cap requirement. This cap system influences interpretation of trends in landings relative to their underlying stocks as changes in landings may not be the direct result of changes in biomass

    Figure 102

    Implications: Landings depict one aspect of the raw stresses from harvesting imposed on the eastern Bering Sea ecosystem’s functional groups through fishing. This information can be useful in identifying areas where harvesting may be impacting different functional groups in times where the functional groups within the ecosystem might be constrained. What is clear from Figure 102 is that pelagic foragers have been by far the largest share of total landings over the 2003–2016 period, while motile epifauna represent the smallest share. Monitoring the trends in landings stratified by ecosystem functional group provides insight on the fishing-related stresses on ecosystems. The ultimate impact that these stresses have on the ecosystem cannot be discerned from these metrics alone and must be viewed within the context of what the ecosystem can provide.

    Contributed by Benjamin Fissel1 , Jean Lee1,2, and Steve Kasperski1
    1Resource Ecology and Fishery Management Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    2Alaska Fisheries Information Network, Pacific States Marine Fisheries Commission Contact: Ben.Fissel@noaa.gov
    Last updated: October 2017

  • Economic Indicators in the Eastern Bering Sea Ecosystem - Value and Unit Value

    Description of indicator: Three plots are used to characterize economic value in an ecosystem context for the eastern Bering Sea. Ex-vessel value is the un-processed value of the retained catch (Figure 105). Ex-vessel value can informally be thought of as the revenue that fishermen receive from the catch.
    First-wholesale value is the revenue from the catch after primary processing by a processor (Figure 106). First-wholesale value is a more comprehensive measure of value to the fishing industry as it includes ex-vessel value as well as the value-added revenue from processing which goes to processing sector.
    The first-wholesale value to total catch unit value is the ratio of value to biomass extracted as a result of commercial fish harvesting (Figure 107). The measure of biomass extracted in this index included retained catch, discards, and prohibited species catch. This metric answers the question: “how much revenue is the fishing industry receiving per-unit biomass extracted from the ecosystem?” The first-wholesale to total catch unit value is analogous to a volumetrically weighted average price across functional groups which is inclusive of discards. However, discards represent a relatively small fraction of total catch. Because of the comparatively larger volume and value from pelagic foragers’ the unit value index is more heavily weighted towards this group.
    Figures 105 and 106 are plotted by functional group. While many species comprise a functional group, it is the handful of species that fishermen target that dominate the economic metrics in each group. The primary target species in the apex predators functional group are Pacific cod (Gadus macrocephalus), Pacific halibut (Hippoglossus stenolepis), Sablefish (Anoplopoma fimbria), and Arrowtooth flounder (Atheresthes stomias). The primary target species in the pelagic foragers functional group are Walleye pollock (Gadus chalcogrammus), Atka mackerel (Pleurogrammus monopterygius), and Pacific ocean perch (Sebastes alutus). The primary target species in the benthic foragers functional group are Yellowfin sole (Limanda aspera), Rock sole (Lepidopsetta bilineata), and Flathead sole (Hippoglossoides elassodon). The primary target species in the salmonid functional group are Chinook (Oncorhynchus tshawytscha), sockeye (O. nerka), and pink (O. gorbuscha) salmon. The primary target species in the motile epifauna functional group are king, bairdi, and snow crab. Because of significant differences in the relative scale of value across functional group, value is plotted in logs. Revenues in Figures 105–107 have been adjusted for inflation using the GDP chain-type deflator (figures based on Fissel et al. (2016)).

    Status and trends: Ex-vessel value is the revenue from landings, consequently trends in ex-vessel value and landings are closely connected. Ex-vessel value is highest in the pelagic forager functional group because of the volume of landings in the pollock fishery. Benthic forager flatfish revenues were increasing from 2000–2008 with increase landings volume but recent declines in value have been the result of decreased prices. Value in the motile epifauna group has been increasing with crab landings. The generally increasing trend in salmon value is the result of generally stable landings and strong prices.
    Differences in the relative level of the indices between the landings and ex-vessel value in Figure 105 reflect differences in the average prices of the species that make up the functional group. Hence, landings of benthic forager flatfish may be larger than salmon, but salmon ex-vessel value is higher because it commands a higher price.
    First-wholesale value was generally increasing for each of the functional groups up to about 2008– 2010 with stable or increasing landings and gradually increasing prices. After this, variation in landings or in prices has had differential impacts. The value of the pelagic forager group has been relatively stable with the exception of 2008–2009 when landings we low. Since 2013 prices for pollock have decreased as global pollock supply has been high, but increased landings have had the combined effect of only marginal decreases in value. First-wholesale value dipped in the apex predator group with a decrease in Pacific cod prices in 2009, but prices rebounded after and stable landings resulted in fairly stable revenue. Benthic forager first-wholesale value decreased from 2012 to 2015 with decreases in flatfish prices as demand for these products plateaued with significant supply. Decreased landings in 2012 brought down salmon value but a price increase buoyed value in 2013 as landings continued to decline after which landing and value have remained at roughly 2010 levels. Value in the motile epifauna group continued to increase with increasing crab prices through 2012 but has since stabilized and value has decreased slightly with marginal reductions in landing.
    The unit value index increased from 2003–2008 with generally increasing prices across all functional groups. Pollock prices fell somewhat in 2013 with significant global pollock supply. Salmon and motile epifauna prices also rose in 2010 and have shown significant volatility since. Apex predator prices dipped in 2009, rebounded in 2010–2011, declined in 2013, and have since leveled out. Benthic forager prices declined through 2009, increased from 2009–2012 and decreased after before leveling out in 2014. The cumulative effect of this price changes is that the first-wholesale unit value index increased to 2008, was relatively volatile at this high level through 2012 then decreased somewhat in 2013 and has vacillated at approximately that level since.

    Figure 106

    Factors influencing observed trends: The reduction in revenue from 2008–2010 was the result of conservation based reductions in the pollock Total Allowable Catch (TAC). Supply reductions in the pollock fishery which began in 2008 resulted in increased first-wholesale prices which account for the significant increase in the 2008 unit value and the relatively high level maintained through 2012. In recent years, increasing global supply has put downward pressure on minimally processed whitefish product prices which has filtered through to the ex-vessel market. As a result, revenue has decreased since 2013 in the pelagic forager and apex predator groups despite strong landings.
    Ex-vessel prices are influenced by a multitude of potential factors including demand for processed products, the volume of supply (both from the fishery and globally), the first-wholesale price, inflation, fishing costs, and bargain power between processors and fishermen. However, annual variation in the ex-vessel prices tends to be smaller than variations in catch and short to medium term variation in the landings and ex-vessel revenue indices appear similar.
    First-wholesale value is the revenue from the sale of processed fish. Some fish, in particular pollock and Pacific cod, are processed into numerous product forms which can influence the generation of revenue by the processing sector. Level shifts in the relative location of the first-wholesale indices compared to the ex-vessel indices are influenced by differences in the amount and types of value-added processing in each functional group

    Figure 107

    Implications: The economic metrics displayed here provide perspective on how the human component of the ecosystem utilizes and receives value from the fish species within the eastern Bering Sea ecosystem. Ex-vessel and first-wholesale value metrics are a measure of the ultimate value from the raw resources extracted and how humans add value to the harvest for their own uses. In contrast to the landings metrics that are heavily dominated by the pelagic forager functional group, ex-vessel and first-wholesale revenues are more evenly distributed across functional groups, which indicates the importance of the groups with lower landings and higher prices to the fishing sector.
    Situations in which the value of a functional group are decreasing but catches are increasing indicate that the per-unit value of additional catch to humans is declining. This information can be useful in identifying areas where fishing effort could be reallocated across functional groups in times where the functional groups within the ecosystem might be constrained while maintaining value to the human component of the ecosystem. Monitoring the economic trends stratified by ecosystem functional group provides insight on the fishing related stresses on ecosystems and the economic factors that influence observed fishing patterns. The ultimate impact that these stresses have on the ecosystem cannot be discerned from these metrics alone and must be viewed within the context of what the ecosystem can provide.

    Contributed by Benjamin Fissel1 , Jean Lee1,2, and Steve Kasperski1 1Resource Ecology and Fishery Management Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA 2Alaska Fisheries Information Network, Pacific States Marine Fisheries Commission Contact: Ben.Fissel@noaa.gov Last updated: October 2017

  • Saltwater Recreational Fishing Participation in the Eastern Bering Sea: Number of Anglers and Fishing Days

    Description of indicator: Federal fisheries management objectives include managing healthy ecosystems in part to provide recreational fishing opportunities. We use saltwater fishing participation to represent trends in recreational fishing in Alaska. The magnitude of recreational saltwater fishing participation is captured by (a) the days fished and (b) the number of anglers. The Alaska Department of Fish and Game (ADF&G) conducts an annual survey of anglers to collect information on participation, catch, and harvest (e.g., Jennings et al. (2015); Romberg (2016)). Annual estimates of the total number of saltwater anglers are available from 1996 to 2015. Estimates of the total number of saltwater fishing days are available from 1981 through 2015. For the purposes of this indicator, ADF&G Sport Fishing Areas A to H and J to Q correspond to the Gulf of Alaska, while Areas R-Z comprise the eastern Bering Sea (EBS) (see http://www.adfg.alaska.gov/sf/sportfishingsurvey/index.cfm?ADFG=main.home).

    Status and trends: In the EBS, the total number of days fished has remained under 30,000, reflecting the low level of saltwater sport fishing that occurs in the region. Since the mid-1990s, there have only been two years with more than 15,000 fishing days in saltwater. In recent years, the annual fishing days has been just shy of 10,000 fishing days (Figure 108). The annual number of saltwater anglers fishing in the EBS has declined overall since the mid-1990s and is currently at about 2,000 anglers (Figure 109).

    Figure 109

    Factors influencing observed trends: The amount of saltwater recreational fishing occurring in the EBS is a small fraction of the amount in the Gulf of Alaska, in large part due to the remoteness of the EBS fishing locations and absence of large population centers. The difficulty in accessing fishing locations in the EBS for non-resident anglers means few non-residents fish in the region. The lower resident population sizes of EBS communities result in relatively low numbers of resident anglers as well.
    Beyond geographic constraints, saltwater recreational fishing participation in Alaska generally is influenced by a number of factors, including fishing regulations for target species, social and economic factors affecting the angler and the angler’s household, and expected fishing conditions (e.g., stock size, timing and size of runs, weather, etc.). Pacific halibut (Hippoglossus stenolepis) and Pacific salmon (Oncorhynchus spp.) are the most common target species, with other species less frequently being the principal target but being caught on trips targeting halibut or salmon. Fishing regulations for these fish influence decisions about whether or not to fish, where to fish, what species to fish for, and by what means to fish (e.g., unguided or guided fishing).
    Fishing regulations in the Pacific halibut sport fishery were first established in 1973, but have changed significantly over the years in the EBS (Meyer, 2010). Beginning in 2014, Southcentral Alaska charter boat anglers, which includes those in Bristol Bay and the Alaska Peninsula, began facing charter-specific bag and size limit and other restrictions (see https://alaskafisheries. noaa.gov/fisheries/2c-3a-halibut-regs). Under the Halibut Catch Sharing Plan (CSP), which went into effect during 2014, the management tools used to regulate harvest of Pacific halibut in the recreational sport sector are evaluated annually (79 Federal Register 13906).
    ADF&G manages Pacific salmon in Alaska primarily through a policy that involves maintaining spawning habitats and ensuring escapement levels through area closures (Heard, 2009). Allocation between the commercial and recreation sectors is set by the Alaska Board of Fish and can have a profound influence on observed trends.
    Macroeconomic factors, such as economy-wide recessions, likely affect participation patterns in saltwater fishing in Alaska, though national-level factors are less likely to impact recreational fishing levels in the EBS due to the low number of non-resident anglers. Instead, the declining trend in the numbers of anglers since the mid-1990s may be related to demographic trends in communities in the EBS, such as the net out-migration of EBS residents in the last decade. The increase in the number of anglers and the number of fishing days in recent years (2013–2015) may be a consequence of households in the EBS turning to saltwater recreational fishing as a secondary food source as the state economy has been in a recession (ADLWD, 2017a). While conditions in the (primarily state and local) economy are likely to explain some of the observed trends, the statistics generally reflect micro-level decisions made by individual anglers (e.g., Lew and Larson (2011, 2012, 2015)).

    Implications: Monitoring the number of saltwater anglers and fishing days provides a general measure of fishing effort and participation in the saltwater sport fishery and can reflect changes in ecosystem conditions, target stock status, management, economic factors, demographic trends, and other economic, social, and cultural factors. Generally, Alaska is well known for its sport fishing opportunities and draws anglers both from within and from outside Alaska. In the EBS, however, saltwater recreational fishing effort is currently low. As a result, it likely represents a trivial source of extraction for sport-caught species like Pacific halibut, Pacific salmon, and rockfish. Nevertheless, studies have indicated saltwater fishing in Alaska is valuable to anglers (e.g., Lew and Larson (2011, 2012, 2015)) and contributes to the economy by creating jobs and generating sales to fishing and non-fishing businesses and income to households (Lovell et al., 2013). Recent estimates of the annual fishing days and total saltwater anglers in the EBS suggest the number of saltwater anglers and the number of fishing days are increasing slightly. Without significant changes in the demographics of the region or the ecological, economic, management, or socio-cultural factors that are likely to influence EBS-level participation in saltwater recreational fishing, it is likely that saltwater recreational fishing will remain at, or near, recently observed levels.

    Contributed by Daniel K. Lew1 and Jean Lee1,2
    1Resource Ecology and Fishery Management Division, Alaska Fisheries Science Center, National Marine Fisheries Service, NOAA
    2Alaska Fisheries Information Network, Pacific States Marine Fisheries Commission
    Contact: Dan.Lew@noaa.gov
    Last updated: August 2017

  • text and graph will go here

     

     

     

     

  • text and graph will go here