EBS Noteworthy (formerly Hot Topics) 2021

Noteworthy (formerly Hot Topics) (pdf)

Here we present items that are new or noteworthy and of potential interest to fisheries managers.

Adult Salmon Run Failures Throughout the Arctic-Yukon-Kuskokwim Region

The Arctic-Yukon-Kuskokwim (AYK) Region experienced unprecedented salmon run failures during the 2021 season. Chinook, chum, and coho salmon runs were extremely weak throughout the entire region (Figures 2, 3, and 4). The 2021 Yukon River salmon season was particularly dire due to concurrent record low run sizes of Chinook, summer chum, fall chum, and coho salmon. For the first time since statehood, all Yukon River salmon directed gillnet fisheries were closed for the entire season, leading to extreme food security and social, cultural, and economic hardships. Even with fishery closures, it is unlikely that the Yukon River salmon run sizes were adequate to meet minimum spawning escapement goals or U.S./Canada Treaty objectives. Fishery restrictions were also required throughout the Kuskokwim and Norton Sound Management Areas, but the situation was mitigated somewhat by Chinook and coho salmon runs within historical ranges and adequate abundances of other salmon species to allow for limited harvest opportunities.

The record low chum salmon runs are of particular concern because chum salmon are the most abundant salmon species returning to the AYK Region and are a critical subsistence, personal use, and commercial resource. Historically, AYK chum salmon have shown resilience and the ability to bounce back from years of low run abundance. Since the mid-2000’s, AYK chum salmon runs have been healthy, characterized by several years of record large run sizes, sustainable fisheries, and consecutive years of meeting or exceeding escapement objectives. Failure of age-4 chum salmon returning to AYK rivers in 2020 forewarned the multiple age-class failure that was observed in 2021. The potential for continued low chum salmon abundance over the coming years should not be overlooked given the changing marine conditions, unprecedented low escapements in 2021, generally low escapements in 2020, and numerous pre-spawn mortality events documented in 2019.

The cause of poor Chinook, chum, and coho salmon runs throughout the AYK Region are not known, but prevailing hypotheses are focused on sub-optimal conditions for growth and survival in the marine environment. AYK salmon species display a wide range of life history strategies and residency times in freshwater and marine environments (Figure 5). The number of spawners

Figure 2: Relative changes in Chinook salmon adult run abundance throughout the Arctic-Yukon- Kuskokwim Region based on three indicator stocks (A), with a focus on the Canadian-origin Yukon River stock (B).

that contributed to the 2021 AYK Chinook, chum, and coho salmon runs were generally within or exceeded escapement goals, and there are no known freshwater environmental factors that easily explain the concurrent poor returns observed in 2021. AYK salmon age and size trends lend some support for marine influences. AYK Chinook salmon have trended towards earlier age-at-maturation and smaller sizes-at-age (e.g., Lewis et al. (2015) and Ohlberger et al. (2018)) and that pattern was again observed in preliminary 2021 data. Yukon River chum salmon and Yukon and Norton Sound coho salmon displayed record low body size, at age, in 2021, suggesting poor growth conditions in the marine environment.

An additional concern for Yukon River Chinook salmon has emerged with the resurgence of ichthyophonus disease after many years of low prevalence. Infection occurs via diet consumed during the marine life stage, and the disease progresses during the adult in-river migration. In 2020

Figure 3: Relative changes in chum salmon adult run abundance throughout the Arctic-Yukon- Kuskokwim Region based on three indicator stocks (A), with a focus on the Yukon River stocks (B).

and 2021, the prevalence of ichthyophonus appears to be near record high levels (based on opportunistic and limited sampling) with currently unknown implications for in-river survival, migration, and spawning success.

The Alaska Department of Fish & Game is addressing the declining salmon runs across the AYK Region through a wide range of applied research initiatives. Efforts include adult salmon tagging programs, increased escapement monitoring, improvements to salmon forecast and total run estimation methods, investigations into the impact of ichthyophonus disease on adult pre-spawn mortality, and expanding the capacity of marine research programs to identify factors that may be affecting the productivity of AYK salmon. Long-term research prioritization and inter-agency collaboration will likely be required to address the needs of salmon fishery management agencies in a changing environment.

Figure 4: Relative changes in coho salmon adult run abundance throughout the Arctic-Yukon-Kuskokwim Region based on three indicator stocks (A), with a focus on the Yukon River stock (B).

Figure 5: Common life history strategies displayed by Arctic-Yukon-Kuskokwim Region Chinook, chum, and coho salmon. Shaded boxes highlight the years during which salmon returning in 2021 were subject to freshwater and marine environments. Note: The 2021 salmon runs to AYK were impacted by environmental conditions experienced during the brood year (BY) spawning, freshwater (FW) rearing, and marine (SW) growth phases. AYK Chinook salmon typically spend 1 year in FW and 2–5 years in SW. Chum salmon out-migrate immediately after emergence and typically spend 3–4 years in the marine environment. Coho salmon typically spend 2 years in FW and 1 in SW.

Contributed by Zachary W. Liller,Alaska Department of Fish & Game Arctic-Yukon-Kuskokwim Region Division of Commercial Fisheries

Getting to the Bottom of it: an Exploration of ROMS Bottom Temperatures

The Bering Sea ROMS model (Bering10K) originated as a subdomain of the larger Northeast Pacific ROMS model (NEP5) approximately a decade ago. Since that time, several peer-reviewed publications have documented its subsequent development along with quantitative and qualitative validation of the model’s performance against physical and biogeochemical observations from the region. First, we highlight several of these publications that focus on various aspects of the coupled ocean-ice-biogeochemical model complex.

Because sea ice dynamics play such a key role in biophysical processes in this region, successful simulation of sea ice advance and retreat, and the interannual variations in both timing and magnitude of sea ice processes, were a necessary precursor for a regional model to be useful in the Bering Sea region. Danielson et al. (2011) discusses the performance of a 35-year hindcast simulation of the NEP5 model, focusing on modes of variability within the ocean and sea ice modules. While this paper predates the Bering10K model, many of its skill metrics, especially those demonstrat- ing successful simulation of currents, stratification, tidal harmonics, and sea ice concentration, are applicable to the smaller Bering Sea domain as well.

Hermann et al. (2013) is the first paper to focus specifically on the Bering10K model. Model validation focused on climatological circulation patterns compared to in situ drifters, as well as water column hydrography (seasonal and interannual patterns in stratification, mixed layer depth, temperature, etc.) compared to long-term moorings located in the middle shelf region.

Biogeochemistry within the Bering10K model is currently simulated using the custom BESTNPZ model (Gibson and Spitz, 2011). Kearney et al. (2020) provides an in-depth evaluation of biophysi- cal and biological metrics related to the implementation of this model within the three-dimensional Bering10K model. This includes a quantitative comparison of cold pool extent within the model compared to measurements from the annual groundfish survey. We also revisit many of the earlier validation metrics related to sea ice extent, mixed layer depth, stratification, and currents, and compare simulated primary production with both satellite-derived and in situ measurements. This paper also includes a history of the updates that were made to the coupled regional model over its 10-year history.

Finally, Kearney (2021) provides a more in-depth look at how the simulated surface and bottom temperatures within the Bering10K model compare to the data collected from the annual conti- nental shelf groundfish survey over its 40-year history. This technical report expands upon the details underlying the cold pool validation metrics presented in Kearney et al. (2020), and presents several maps of skill metrics (such as bias, correlation, RMSE, etc.) for both bottom and surface temperature across the southeastern and northern shelf regions.

Next, we present several comparative examples of satellite-derived sea surface temperature and ROMS bottom temperature data that facilitate examination of spatial (i.e., vertical differences) and temporal (i.e., phenological) impacts of temperature dynamics between surface and bottom waters on organisms in the eastern Bering Sea.

Spatial patterns are evident when looking across depth-defined strata (i.e., inner vs. middle/outer shelf) (Figure 6). The left hand panels show temperature variability in the inner domain (10–50m) while the right hand panels show variability in the middle/outer domain (50–200m). The impact of mixing in the inner domain results in bottom waters having larger swings in temperature, whereas

stronger stratification in the middle/outer domain leads to more stable bottom water temperatures.

Figure 6: ROMS bottom temperatures averaged within depth-defined strata for the Northern and Southeastern Bering Sea regions.

Temporal patterns (within and across years) of water temperature dynamics have impacts on individual species’ phenological responses and subsequent match/mismatch patterns. In Figure 7, the bottom panel shows warm waters persisted at depth through fall 2019 whereas surface waters were relatively cooler. Both surface and bottom temperatures were closer to the long-term mean in 2021.

Figure 8 highlights a potentially phenologically important trigger for organisms’ early life history development as well as horizontal and vertical distributions. The relative timing each year when the surface water temperature drops below bottom water temperature has varied over the time series, with greater variability in the inner domain than in the middle/outer domains.

Contributed by Kelly Kearney, University of Washington, Cooperative Institute for Climate, Ocean, and Ecosystem Studies [CICOES] and NOAA Fisheries, Alaska Fisheries Science Center Jordan Watson, NOAA Fisheries, Alaska Fisheries Science Center, Auke Bay Laboratories, Matt Callahan, Pacific States Marine Fisheries Commission Tyler Hennon, University of Alaska Fairbanks, College of Fisheries and Ocean Sciences

Figure 7: Phenology of SST (satellite-derived, source: Coral Reef Watch) and bottom temperature (derived from ROMS) in the northern and southeastern Bering Sea. Years are plotted Sept–Aug with Sept–Dec appended to the subsequent year. Frequency of data is daily for SST and weekly for ROMS. Depths are filtered to between 10–200m. Note different y-axis scales for SST and bottom temperatures.

Figure 8: Relative timing within each year during which the average regional surface temperatures dropped below the bottom temperatures. Years are plotted Sept–Aug with Sept–Dec appended to the subsequent year. Missing values in 2019–2021 in SEBS regions demonstrate that the surface temperature never dropped below the bottom temperature in that year.

Ice seal Unusual Mortality Event: an update

On September 12, 2019, NOAA Fisheries declared an Unusual Mortality Event (UME) for three species of ice seals in Arctic waters of Alaska. This declaration resulted from elevated numbers of dead bearded, ringed, and spotted seals on Alaska shores, which were reported, beginning in June 2018, from Kotlik in the northern Bering Sea to Utqiag˙vik in the northern Chukchi Sea. The UME investigation continued in 2020 through 2021.

Prior to the 2019 ice seal UME declaration, strandings during 2010–2017 averaged 29 ice seals annually. For the next four years, confirmed strandings included 111 ice seals (June 1–December 31, 2018), 164 seals (2019), 34 seals (2020), and 41 seals (2021) (Table 1, Figure 9). These bearded, ringed, spotted, and unidentified stranded seals were confirmed from dedicated surveys and opportunistic sightings.

Table 1: Confirmed ice seal strandings related to the UME in the Bering and Chukchi seas.

Year Bearded Ringed Spotted Unidentified Total
2010–2017 Average

2018a 35 29 20 27 111
2019 50 35 26 53 164
2020 9 7 7 11 34
2021 10 20 7 4 41
TOTALb 104 91 60 95 350

a 1 June–31 December 2018

b 1 June 2018–8 October 2021

The reports from 2018–2019 indicated several seals were emaciated at the time of death. Stranding reports for seals from 2020–2021, however, did not identify emaciation as a factor in the seals’ condition. Most photographs and reports identified the 2020–2021 stranded seals as healthy and robust.

In 2020, strandings in June (15 seals) were at the same level as in July and August (16 seals). In 2021, there were fewer strandings during June (14 seals), than in July and August (68 seals). Of note, photos and/or skin samples confirmed only 50% of stranding reports in 2020 and 47% of the stranding reports during 2021.

The ongoing COVID-19 pandemic severely limited NOAA’s ability to travel to collect tissue samples, morphometric data, and/or conduct surveys during 2020–2021. Coastal residents were essential for the documentation and reporting of strandings while conducting their normal daily activities (ATV, boating, and walking). However, because of the remote locations, decomposition of the stranded seals, and a lack of traveling biologists/veterinarians, sampling and full necropsies of ice seals remained uncommon.

The increased mortality of seals reported during 2018–2019 coincided with the dramatic reduction in the extent, quality, and duration of sea ice habitat for pupping and nursing in the northern Bering Sea during both years (Boveng et al., 2020). The increased mortality of young seals during June 2018–2019 could also indicate impacts from the effects of a transitioning ecosystem, such as competition for prey. In a study by the Alaska Fisheries Science Center, spotted seal pups

image Figure 9: Number of confirmed ice seal strandings in Alaska by month.

Figure 9: Number of confirmed ice seal strandings in Alaska by month.

and ribbon seals 4 of all age classes declined in body condition over a longer period (2007–2018), coincident with a decline in Bering Sea ice extent, quality, and duration (Boveng et al., 2020). The ice seal UME of 2018–2019 may therefore reflect an ecological shock from those two extreme years superimposed on a longer-term trend.

Contributed by Barbara Mahoney, NOAA Fisheries, Protected Resources Division Peter Boveng, NOAA Fisheries, Marine Mammal Laboratory, Gay Sheffield, University of Alaska Fairbanks, Alaska Sea Grant


4 Ribbon seals were not among the identified strandings in the UME, but they typically are reported in much lower numbers than bearded, ringed, and spotted seals, likely due to their smaller population and more offshore habits.