ECOSYSTEM ASSESSMENT

Kerim Aydin¹, Jennifer Boldt², Sarah Gaichas¹, Shannon Bartkiw¹, Nick Bond², Troy Buckley¹, Cathy Coon⁴, Shannon Fitzgerald¹, Lowell Fritz⁵, Angie Greig¹, Terry Hiatt¹, Jerry Hoff¹, Kathy Kuletz⁶, Carol Ladd³, Robert Lauth¹, Pat Livingston¹, Michael Martin¹, Franz Mueter⁷, Marcia Muto⁵, Jeff Napp¹, Jim Overland³, Sigrid Salo³, Elizabeth Sinclair⁵, Paul Spencer¹, Phyllis Stabeno³, Ward Testa⁵, Gary Walters¹, Muyin Wang², and Atsushi Yamaguchi⁸

¹Alaska Fisheries Science Center, ²University of Washington, ³Pacific Marine Environmental Lab, ⁴NPFMC, ⁵National Marine Mammal Lab, ⁶USFWS, ⁷University of Alaska, ⁸Hokkaido University, Japan

Contact:  Kerim.Aydin@noaa.gov

Last updated:  November 2008

Introduction

The primary intent of this assessment is to summarize and synthesize historical climate and fishing effects on the shelf and slope regions of the eastern Bering Sea/Aleutian Islands and Gulf of Alaska from an ecosystem perspective and to provide an assessment of the possible future effects of climate and fishing on ecosystem structure and function.  The Ecosystem Considerations section of the Groundfish SAFE provides the historical perspective of status and trends of ecosystem components and ecosystem-level attributes using an indicator approach.  For the purposes of management, this information must be synthesized to provide a coherent view of ecosystems effects in order to clearly recommend precautionary thresholds, if any, required to protect ecosystem integrity. 

The eventual goal of synthesis is to provide succinct indices of current ecosystem conditions reflecting these ecosystem properties.  In order to perform this synthesis, a blend of data analysis and modeling will need to be employed to place measures of current ecosystem states in the context of history and past and future climate.  In this year’s assessment, we derived a ‘short’ list of key indicators to track in the EBS, AI, and GOA, using a stepwise framework, the DPSIR (Drivers, Pressure, Status, Indicators, Response) approach (Elliot 2002).

In applying this framework we have initially determined four objectives based, in part, on stated ecosystem-based management goals of the NPFMC:  maintain predator-prey relationships, maintain diversity, maintain habitat, and incorporate/monitor effects of climate change.  Drivers and pressures pertaining to those objectives were identified and a list of candidate indicators were selected that address each objective and candidate indicators were chosen based on qualities such as, availability, sensitivity, reliability, ease of interpretation, and pertinence for addressing the objectives (Table 1).  In future drafts, we plan to more fully address the human responses (Response portion of the DPSIR approach) to changes in status and impacts.  Use of this DPSIR approach will enable the Ecosystem Assessment to be in line with NOAA’s vision of Integrated Ecosystem Assessments.  For each objective, driver and pressure identified, indicators are briefly described and the status and trends of the indicators are explained.  Where possible, factors that caused those trends are discussed and the potential implications are described.  Some gaps in knowledge are listed for each objective.

Table 1.  Objectives, drivers, pressures and effects, significance thresholds and indicators for fishery and climate induced effects on ecosystem attributes.

Results

A. 
Issue: Predator prey relationships and energy flow
Objectives: Maintain predator prey relationships and energy flow
Drivers: Need for fishing, per capita seafood demand, and concern about climate change
Pressures: Pelagic forage availability, removal of top predators, energy re-direction, energy removal relative to production
Status and Impacts Indices:

1.  Biomass, Catch, and exploitation rates of biological guilds (Bering Sea only).

Contributed by Kerim Aydin and Sarah Gaichas, NMFS

Index:  While species-specific fishing may not wholly account for “ecosystem influences” of fishing, whole-ecosystem indices such as trophic level of the catch may be too coarse, especially if pervasive “fishing down the food web” issues are minor.  Further, as with recent concerns of arrowtooth flounder in the Gulf of Alaska, it is important to evaluate “balance” between different broad biological subcomponents (Guilds) in the ecosystem.   For the EBS, species identified by food web models (Aydin et al. 2008) were separated into 12 guilds by trophic role; the guilds span the trophic levels between phytoplankton and apex predators and include a separate pathway for pelagic and benthic components of the ecosystem (Table 2).

For each guild, available time trends of biomass, catch, and exploitation rate (catch/biomass) are presented.  For biomass time trends, stock assessment estimates are used where available for each species within the guild; where no stock assessment models are available survey data is used.  If neither time series are available, the species is assumed to have a constant value equal to the mid-1990s level estimated in Aydin et al. (2008).  Multi-species model estimates are not used here; however, a minimal consumption estimate from diet data and ration estimates was used to calculate a single survey q for forage fish catch in the bottom-trawl surveys (see Aydin et al. 2007 for methods).  Catch data was directly taken from stock assessments or the Catch Accounting System for non-target species.  For 2009-2010, the stock assessment authors’ recommended catch and estimated biomass time series were used; for survey data biomass was assumed to be equal to 2008 levels. 

Status and trends:  Biomass, catch, and explotation rates have beeen within +/- one standard deviation of 1982-2007 levels for all guilds except pelagic foragers; this guild is dominated by walleye pollock (80% of guild biomass in 2007).  The decrease in pollock along with general declines in other forage species has brought the biomass of this group to overall low levels.  Exploitation rate was over one standard deviation above the mean from 2005-2007, however the decreased catchs in 2008, and recommended lower ABC in 2009-2010, has brought the exploitation rate of this guild back towards its long-term mean.  A second trend of interest is that for copepods; the data shown is a strata-weighted average of the data presented in Napp and Yamaguchi (this document).  Standing stock of copepods was low in 2001-2005 but 2006 and 2007 values showed a return towards the mean.

1.  Trophic level of the catch

Contributed by Jennifer Boldt, UW, and Pat Livingston, NMFS

Index:  An index that has been suggested as a measure of overall top-down control of the ecosystem due to fishing is the trophic level of the fishery; in particular, the notion of “fishing down the food web” has been popularized in recent years.  The trophic level of the catch and the Fishery in Balance (FIB) indices have been monitored in the BS, AI, and GOA ecosystems to determine if fisheries have been "fishing-down" the food web by removing top-level predators and subsequently targeting lower trophic level prey.  The FIB index was developed by Pauly et al. (2000) to ascertain whether trophic level catch trends are a reflection of deliberate choice or of a fishing-down the food web effect.  This index declines only when catches do not increase as expected when moving down the food web (i.e., lower trophic levels are more biologically productive), relative to an initial baseline year.  The single metrics of TL or FIB indices, however, may hide details about fishing events..

Status and Trends:  Although there has been a general increase in the amount of catch since the late 1960s in all three areas of Alaska, the trophic level of the catch has been high and relatively stable over the last 25 years.   

Factors Causing Trends:  In general, it appears that fishing events on different species are episodic in the AI and GOA, while pollock steadily dominate catches in the BS throughout the period.

Implications:  Unlike other regions in which this index has been calculated, such as the Northwest Atlantic, the FIB index and the trophic level of the catch in the EBS, AI, and GOA have been relatively constant and suggest an ecological balance in the catch patterns.  Further examination supports the idea that fishing-down the food web is not occurring in Alaska, and there does not appear to be a serial addition of lower-trophic-level fisheries in the BS or GOA.

2.  Bycatch of sensitive top predators

Index:  Groundfish fishery bycatch of sensitive species such as, marine mammals and seabirds, provides an index of the total fishery removal of top predators in ecosystems. 

Status and Trends:  Incidental mortality of pinnipeds in groundfish fisheries was low from 1998-2005, and did not exceed PBRs, and are not expected to have a direct effect on the population status of pinnipeds (Sinclair et al. 2006).  Between 1998 and 2005, an average of 24 harbor seals was taken annually in fisheries in both SEAK and the GOA, and an average of 1 was taken in the BS (Sinclair et al. 2006).  An annual average of 2.6 and 24.6 Steller sea lions were taken in the Eastern and Western Pacific (Sinclair et al. 2006).  Sixteen Northern fur seals on average were taken in the East North Pacific annually Sinclair et al. 2006). 

Most seabird bycatch is taken with longline gear (65-94%), although some bycatch is taken with trawls (6-35%) or pots (1%).  The average annual longline bycatch of seabirds is comprised of primarily  fulmars, gulls, and some unidentified birds, albatross, and shearwaters.  Of the total longline seabird bycatch in 2004, 94.3% was caught in the BS, 2.5% in the AI, and 3.2% in the GOA.  Pots catch primarily Northern fulmars, whereas trawl and longline fisheries catch a wider variety of seabirds.  In 2002, total catch of seabirds was 4,694 in the BS, 124 in the AI, and 161 in the GOA (Fitzgerald et al. 2006).  Between 1993 and 2004 the average annual bycatch in the combined Alaskan longline fisheries was 13,144 birds (Fitzgerald et al. 2006).  Over this period the average annual bycatch rates (birds per 1,000 hooks) were 0.065 in the AI and BS areas and 0.021 in the GOA (Fitzgerald et al. 2006).  Those rates have dropped in the last few years, with the running 5-year average now (2000-2004) at 0.035, 0.036, and 0.010 for the AI, BS, and GOA regions respectively. 

Catch of spiny dogfish in groundfish fisheries varies spatially and temporally.  Catches of spiny dogfish were highest in 1998 and 2001 in many areas of the central and western GOA and Prince William Sound (Courtney et al. 2004; Boldt et al. 2003).  Spiny dogfish catch in the BS was low, but also peaked in 2001.  Bycatch in the BS is primarily from along the Alaska Peninsula and along the BS shelf (Courtney et al. 2004; Boldt et al. 2003).  There was no apparent temporal pattern in sleeper shark bycatch in the GOA or PWS (Courtney et al. 2004; Boldt et al. 2003).  Bycatch in the BS was lower and concentrated along the BS shelf.  BS sleeper shark bycatch in 2001 was the highest since 1997 (Courtney et al. 2005; Boldt et al. 2003).  Courtney et al. (2005) state that:  “…a 2% reduction in biomass per year due to fishing is likely less than natural mortality for Pacific sleeper sharks, unless they are extremely long lived. Based upon this risk criterion, Pacific sleeper sharks do not appear to be at risk of overfishing at current levels of incidental catch.”

Factors Causing Trends:  Trends in bycatch may reflect changes in populations due to environmental and/or biological factors, but could also be due to changes in management and bycatch avoidance measures.  Also, seabird mortality in Alaska groundfish fisheries represents only a portion of the fishing mortality that occurs, particularly with the albatrosses.

B.

Issue:  Predator-prey relationships and energy flow
Objective:  Maintain Predator-prey relationships
Driver:  Need for fishing; Per capita seafood demand
Pressure:  Energy redirection
Status and Impacts Indices:

1.  Discards and discard rates

Contributed by Terry Hiatt, NMFS

Index: Estimates of discards for 1994-2002 come from NMFS Alaska Region’s blend data; estimates for 2003-07 come from the Alaska Region’s catch-accounting system.  It should be noted that although these sources provide the best available estimates of discards, the estimates are not necessarily accurate because they are based on visual observations by observers rather than data from direct sampling.

Status and Trends:  In 1998, the amount of managed groundfish species discarded in Federally-managed groundfish fisheries dropped to less than 10% of the total groundfish catch in both the Bering Sea/Aleutian Islands and the Gulf of Alaska.  Discards in the Gulf of Alaska increased somewhat between 1998 and 2003, declined in 2004 and 2005, and have increased again in the last two years.  Discard rates in the Aleutian Islands (AI) dropped significantly in 1997, trended generally upwards from 1998 through 2003, and have declined again over the last four years.  Discards in all three areas are much lower than the amounts observed in 1996 (AI) and 1997 (BS and GOA), before implementation of improved-retention regulations.

Factors Causing Trends:  Decreases in discards are explained by reductions in the discard rates of pollock and Pacific cod that resulted from regulations implemented in 1998 prohibiting discards of these two species.

C.
Issue:  Predator-prey relationships and energy flow
Objective:  Maintain Predator-prey relationships
Driver:  Need for fishing; Per capita seafood demand
Pressure:  Energy redirection
Status and Impacts Indices:

1.  Total catch levels

See next section on invasive species

D.
Issue:  Predator-prey relationships and energy flow
Objective:  Maintain Predator-prey relationships
Driver:  Need for fishing; Per capita seafood demand
Pressure:  Introduction of non-native species
Status and Impacts Indices:

1.  Invasive species observations

Information from Fay (2002)

Index:  Invasive species are those that are not native to Alaska and that could harm the environment, economics, and/or human health of the region (Fay 2002).  The main marine invasive species that are in Alaska or that could potentially be introduced to Alaska include:  Atlantic salmon (Salmo salar), green crab (Carcinus maenas), Chinese mitten crab (Eriocheir sinensis), oyster spat and associated fauna, bacteria, viruses, and parasites.

Status and Trends:  Currently, Alaska has relatively few aquatic (including marine) invasive species.  Natural spawning of escaped Atlantic salmon has been observed in British Columbian streams, indicating that this could also occur in Alaska.  Chinese mitten crab, native to China, is now established in California and may have spread to the Columbia River (Fay 2002).  Uncertified oyster spat that is imported to Alaska for farming purposes can introduce not only oyster spat (although it is thought that Alaskan waters are too cold for oysters to reproduce), but also other invertebrate larvae, bacteria and viruses (Fay 2002). 

Factors Causing Trends:  The introduction of aquatic invasive species in Alaska can occur in a number of ways, such as those that Fay (2002) lists, including:   “fish farms, the intentional movement of game or bait fish from one aquatic system to another, the movement of large ships and their ballast water from the United States West Coast and Asia, fishing vessels docking at Alaska’s busy commercial fishing ports, construction equipment, trade of live seafood, aquaculture, and contaminated sport angler gear brought to Alaska’s world-renowned fishing sites.” 

Implications:  The potential implications of introductions of non-native species to Alaska marine ecosystems are largely unknown.  Fay (2002), however, states:  “It is thought Atlantic salmon would most likely compete with native steelhead, cutthroat trout, Dolly Varden, and coho salmon, and may also adversely impact other species of Pacific salmon.”  The green crab, which is capable of surviving in Alaskan nearshore waters, could pose a competitive threat to Alaskan tanner and Dungeness crab stocks since they utilize the same nearshore areas as nurseries.  Fay (2002) states:  “With a catadromous life history [the Chinese mitten crab] can move up rivers hundreds of miles where it may displace native fauna, and it is known to feed on salmonid eggs, which could affect salmon recruitment.” Fay (2002) states:  “Little is known about the threat of the movement of bacteria, viruses, and parasites within or to Alaska.  Devastations from the Pacific herring virus in PWS is well known and documented….movement of ballast water from one place to another within Alaska coastal waters could result in injury to other fisheries.  Atlantic Ocean herring disease could also be introduced into Alaska through the import of frozen herring that are used as bait by Alaskan commercial fishers.”

2.  Total catch levels

Index:  Total catch provides an index of how many groundfish fishing vessels are potentially exchanging ballast water resulting in the possible introduction of non-native species. 

Status and Trends:  Total catch in the eastern BS was relatively stable from 1984 to the mid-1990s at approximately 1.7 million t.  In 1999 there was a decrease in catch primarily due to decreased catches of pollock and flatfish, catches then increased to approximately 1.9 million t annually in 2002-2004, and recently in 2007 decreased due to decreases in pollock catch. 

Total catch in the AI is much lower than in the BS and has been more variable (from 61,092 to 190,750 t between 1977 and 2004).  Total catch peaked in 1989, comprised mainly of pollock, and in 1996, comprised of pollock, Pacific cod, Atka mackerel, and rockfish.  Pollock were a large proportion of catches from the late 1970s to the early 1990s.  In 2007, cod catches increased. 

In the GOA, total catch has ranged from less than 50,000 t in the 1950s to highs of 384,242 t in 1965, which was associated with high rockfish catches, and 377,809 t in 1984, which was associated with high pollock catches.  Since the 1985, total catch has varied between 180,301 t (1987) and 307,525 t (1992).  Catches of pollock and Pacific cod determine the major patterns in catch variability.

Factors Causing Trends:  Pollock and flatfish catches drive the catch trends in the Bering Sea.   Catch trends in the AI are driven by catches of pollock, Pacific cod, Atka mackerel, and rockfish.  In the GOA, catch trends are driven by catches of pollock and Pacific cod.  The potential for introductions of invasive species through groundfish fishery ballast water exchange likely increased in the 1960s with increased catches.

Implications:  The effects of the introduction of invasive species via the movement of large ships and their ballast water in Alaska marine ecosystems is largely unknown.

Gaps in predator-prey relationship knowledge:

Information or indicators that would improve our understanding of predator-prey relationships in Alaska marine ecosystems includes:

1.  a time series of zooplankton biomass in the GOA and AI

2.  a time series of forage fish species in all areas

3.  an indicator of the degree of spatial and temporal concentration of groundfish fisheries

E.
Issue:  Habitat
Objective:  Maintain habitat
Driver:  Need for fishing; Per capita seafood demand
Pressure:  Habitat loss/degradation due to fishing gear effects on benthic habitat, HAPC biota, and other species
Status and Impacts Indices:

1.  Areas closed to bottom trawling in the EBS/ AI and GOA

Contributed by John Olson, NFMS

Index and Status:  Many trawl closures have been implemented to protect benthic habitat or reduce bycatch of prohibited species (i.e., salmon, crab, herring, and halibut).  Some of the trawl closures are in effect year-round while others are seasonal.  In general, year-round trawl closures have been implemented to protect vulnerable benthic habitat.  Seasonal closures are used to reduce bycatch by closing areas where and when bycatch rates have historically been high.  Additional measures to protect declining western stocks of the Steller Sea Lion began in 1991 with some simple restrictions based on rookery and haulout locations, to specific fishery restrictions 2000 and 2001.  For 2001, over 90,000 nmi of the EEZ off Alaska was closed to trawling year-round.  Additionally 40,000 nmi were closed on a seasonal basis.  State waters (0-3nm) are also closed to bottom trawling in most areas.  Closures implemented in 2006 as part of protection for Essential Fish Habitat encompass a large portion of the Aleutian Islands.  The largest of these closures is called the Aleutian Islands Habitat Conservation area and closes 279,000 nmi to bottom trawling year round. Five new closures implemented in 2008 as part of protection for Essential Fish Habitat encompass a large portion of the northern Bering Sea. These five closures add 134,500 nm2 to the area closed to bottom trawling year round. By implementing these closures, almost 50% of Alaska’s EEZ is closed to bottom trawling. 

2.  Fishing effort

Contributed by John Olson, NMFS

Index:  Fishing effort is an indicator of damage to or removal of Habitat Areas of Particular Concern (HAPC) biota, modification of nonliving substrate, damage to small epifauna and infauna, and reduction in benthic biodiversity by trawl or fixed gear.  Intensive fishing in an area can result in a change in species diversity by attracting opportunistic fish species which feed on animals that have been disturbed in the wake of the tow, or by reducing the suitability of habitat used by some species.  Trends in fishing effort will reflect changes due to temporal, geographic, and market variability of fisheries as well as management actions.  Bottom trawl and hook and line effort are measured as the number of observed days fished; whereas, pot fishing effort is measured as the number of observed pots fished.  Observed fishing effort is used as an indicator of total fishing effort.  It should be noted, however, that most of the vessels using pot gear are catcher vessels either under 60’ or between 60’-125’.  These vessels either do not require an observer present or only on 30% of the fishing days. 

Status, Trends:  In general, bottom trawl effort in the Gulf of Alaska and Aleutian Islands has been relatively low since 2004, with a slight increase in 2007.  Bottom trawl effort in the Bering Sea remained relatively stable from 2001 through 2006 and decreased in 2007.  Hook and line effort in the Bering Sea increased from 1990 to 2004 before it decreased in 2005-2007.  In the Aleutian Islands, hook and line effort has been relatively low for the last 5 years.  In the Gulf of Alaska hook and line effort has been relatively stable over the last 10 years.  Pelagic trawl effort in the BS was relatively stable during 1999-2006 with a small increase in 2007.  There has been very little or no pelagic trawl effort in the AI in recent years.  Pelagic trawl effort in the GOA increased slightly in 2007.  The observed pot fishing effort has been relatively stable in the BS, GOA, and AI in the last few years.

Factors Causing Trends:  Some of the reduction in bottom trawl effort in the Bering Sea after 1997 can be attributed to changes in the structure of the groundfish fisheries due to rationalization.  As of 1999, only pelagic trawls can be used in the Bering Sea pollock fisheries.  Fluctuations in bottom trawl effort track well with overall landings of primary bottom trawl target species, such as flatfish and to a lesser extent pollock and cod. 

Hook and line effort in both the Bering Sea and Aleutian Islands occurs mainly for Pacific cod, Greenland turbot, and sablefish.  The predominant hook and line fisheries in the Gulf of Alaska are composed of sablefish and Pacific cod.  In southeast Alaska, there is a demersal rockfish fishery dominant species include yelloweye rockfish (90%), with lesser catches of quillback rockfish.  Sablefish has been an IFQ fishery since 1995, which has reduced the number of vessels, crowding, gear conflicts and gear loss, and increased efficiency. 

The pot fishery occurs mainly for Pacific cod which form dense spawning aggregations in the winter months.  In the Bering Sea, fluctuations in the pot cod fishery may be dependent on the duration and timing of crab fisheries.  There is also a state-managed fishery in State waters.  

There are spatial variations in fishing effort in the BS, GOA, and AI (see fishing effort contributions, this report).  Spatial changes in fisheries effort may in part be affected by fishing closure areas (i.e., Steller sea lion protection measures) as well as changes in markets and increased bycatch rates of non-target species.

Implications:  The effects of changes in fishing effort on habitat and HAPC biota are largely unknown.  It is possible that the reduction in bottom trawl effort in all three ecosystems could result in decreased habitat loss/degradation due to fishing gear effects on benthic habitat, HAPC biota, and other species; whereas, increases in hook and line and pot fisheries could have the opposite effect.  The footprint of habitat damage likely varies with gear (type, weight, towing speed, depth of penetration), the physical and biological characteristics of the areas fished, recovery rates of HAPC biota in the areas fished, and management changes that result in spatial changes in fishing effort (NMFS 2007; http://www.nmfs.noaa.gov/pr/permits/eis/steller.htm).

3.  HAPC biota catch

Index:  In addition to prohibited and target species catches, groundfish fisheries also catch non-target species.  HAPC biota (seapens/whips, sponges, anemones, corals, tunicates) comprise a portion of the non-target species catches.  HAPC biota are taxa which form living substrate, and are identified by NMFS as meeting the criteria for special consideration in resource management.  HAPC biota are used by fish, including commercially important groundfish, as habitat.  Bycatch of HAPC species in both trawl and longline gear is of concern.  Concentrations of HAPC species often occur in nearshore shallow areas but also are found in offshore deep water areas with substrata of high microhabitat diversity.  Trends in fishery catches of HAPC biota may be indicators of total HAPC biota removals.  In addition to tracking removal of HAPC biota, fishery catches of HAPC biota may also reflect changes in management actions, fishing effort, the spatial distribution of the fishery, and/or in HAPC biota abundance; however, distinguishing between these is not possible and not the purpose of this index here.  Catches are estimated based on visual observations by observers rather than from direct sampling; therefore, may be less accurate than target fish catch estimates.

Status, Trends, and Factors Causing Trends:  In the BSAI, catches of HAPC biota decreased 2003-2007.  The catch of HAPC biota in the GOA is approximately 50 times lower than in the BSAI and has varied annually. 

Factors Causing Trends:  Benthic tunicates comprise the majority of HAPC biota catches in the BSAI, caught mainly by the flatfish fishery; this catch has decreased since 2004.  Sea anemones comprise the majority of HAPC biota catch in the GOA and they are caught primarily in the flatfish fishery. 

Implications:  The reduction in HAPC biota catches imply that removal of those taxa by fishing gear has been reduced in the BSAI and been relatively stable in the GOA in recent years.  The cause of this decrease is largely unknown but could be due to a combination of factors, such as the reduction in bottom trawl fishing effort in the Bering Sea, variation in gear (type, weight, towing speed, depth of penetration), changes in areas fished and the physical and biological characteristics of the areas, recovery rates of HAPC biota in the areas fished (NMFS 2007; http://www.nmfs.noaa.gov/pr/permits/eis/steller.htm).

4.  HAPC biota survey CPUE

Contributed by Michael Martin and Robert Lauth, NMFS

Index:  As mentioned above, HAPC biota are taxa that form living substrate which are used by fish, including commercially important groundfish, as habitat.  HAPC biota include seapens/whips, sponges, anemones, corals, and tunicates.  NMFS bottom trawl survey catches of HAPC biota provide one potential indicator of HAPC biota abundance trends.  Sampling is done over the same large areas annually in the BS and biennially in the AI and GOA.  This is, however, not the ideal indicator of abundance trends because the survey gear is not designed for efficient capture of all HAPC biota, it does not perform well in many of the areas where these groups are thought to be more prevalent and survey effort is quite limited in these areas as a result, catches are highly variable, and the survey gear and onboard sampling techniques have changed over time.  Examination of the frequency of occurrence in hauls may address some of these issues (see HAPC biota for the three regions, this report).   

Status, Trends:  Despite the caveats, a few general patterns are clearly discernible.  The CPUE of HAPC biota is highest in the Aleutian Islands.  In the AI, HAPC biota CPUE has been variable, but relatively stable for the last 5 survey years.  The CPUE of HAPC biota in the Bering Sea peaked in the late 1990s to the early 2000s, and has decreased since then.  In the BS, over the last eight years, sea whip and sea anemone CPUE has increased, whereas, sponge CPUE has decreased.  Both the mean CPUE and frequency of occurrence of gorgonians seem to have decreased since 1994 in the AI; this is opposite the trends seen in stony corals over the same time period.  HAPC biota CPUE in the GOA have been relatively low and stable, with a slight decline during the last 4 survey years.  The frequency of occurrence of sponge and sea anemones in the GOA, however, seems to have increased since 1984.   

Factors Causing Trends:  Trends in both the BS and AI are driven primarily by sponge CPUE.   Sea anemone and sponge CPUE drive trends observed in the GOA.  Prior to 1990, Japanese vessels using large tire gear performed the majority of tows in both the AI and GOA.  This allowed these vessels to sample in areas considered untrawlable with current survey gear, so damage to HAPC biota likely exceeded later years, even though catches were generally smaller.  This gear difference is thought to largely account for the abrupt change in relative abundance patterns after 1987.  There are also regional trends within each of the three ecosystems (see HAPC biota for the three regions, this report). 

Implications:  Survey catches of HAPC biota may not necessarily reflect population abundance trends; therefore, the implications of survey catch trends of HAPC biota are largely unknown.  The population trends of HAPC biota are not necessarily represented by survey catches because surveys are currently unable able to devote effort to sampling untrawlable areas that have the highest HAPC biota abundance, especially in the AI.

Gaps in habitat knowledge:

Information or indicators that would improve our understanding of habitat in Alaska marine ecosystems includes:

1.  habitat disturbance as a function of fishing intensity

2.  HAPC biota population abundance and distribution, particularly in areas currently untrawlable with standard survey gear.

3.  the importance of HAPC biota as habitat for different species and life stages of fish

4.  the relationship between physical factors such as sediment type, bathymetry, and oceanography and the abundance and distribution of HAPC biota.

5.  an index that reflects the amount of fish habitat that is damaged, such as:  proportion of habitat damaged by fishing gear, or the area (km²) with HAPC biota closed to fishing relative to the area with HAPC biota that is open to fishing.

F.
Issue:  Diversity
Objective:  Maintain Diversity
Driver:  Need for fishing; Per capita seafood demand
Pressure:  Effect of fishing on diversity
Status and Impacts Indices:

1.  Groundfish survey species richness and diversity

Contributed by Franz Mueter, University of Alaska

Indices:  The number of species and the proportions of species in an ecosystem can be affected by fishing in a variety of ways, including the removal of species and the removal of invertebrate species that provide fish habitat (e.g., sponge).  The effect of fishing on species richness and diversity are poorly understood at present. Because fishing primarily reduces the relative abundance of some of the dominant species in the system, species diversity is expected to increase relative to the unfished state.  However, changes in local species richness and diversity are strongly confounded with natural variability in spatial distribution and relative abundance.  The Shannon-Wiener diversity index and species richness index are standard indices of the numbers and proportions of species.  Utilizing the NMFS standard bottom trawl survey data, the average number of fish and major invertebrate taxa per haul and the average Shannon index of diversity (based on weight CPUE; Magurran 1988) by haul were computed for the GOA (west of 147°N) and EBS.  Indices were based on a total of 53 taxa in the GOA and 46 taxa in the EBS (Table 1 in Mueter & Litzow 2008). Taxa were included at the lowest possible taxonomic level, i.e. at a level that was consistently identified throughout all surveys. Indices were computed following Mueter & Norcross (2002). Briefly, annual average indices of local richness and diversity were estimated by first computing each index on a per-haul basis, then estimating annual averages by modeling haul-specific indices as a function of geographic location, depth, date of sampling, area swept, and year.

Status and Trends:  Average species richness and diversity of the groundfish community in the Gulf of Alaska increased from 1990 to 1999 with both indices peaking in 1999 and sharply decreasing between 1999 and 2001.  Species richness and diversity on the Eastern Bering Sea shelf have undergone significant variations from 1982 to 2006.  The average number of species per haul has increased by one to two species since 1995, while the Shannon Index increased from 1985 through 1998 and decreased sharply in 1999.

Factors Causing Trends:  The average number of species per haul depends on the spatial distribution of individual species (taxa).  If species are, on average, more widely distributed in the sampling area the number of species per haul increases.  Spatial shifts in distribution from year to year lead to high variability in local species richness in certain areas, for example along the 100m contour in the Eastern Bering Sea.  These shifts appear to be the primary drivers of changes in species richness.  Local species diversity is a function of how many species are caught in a hauls and how evenly CPUE is distributed among the species.  In the GOA both average species diversity and local richness showed very similar trends, suggesting that relative species composition (evenness) was relatively stable.  In contrast, trends in species diversity in the EBS differed markedly from those in richness.  For example, low species diversity in the EBS in 2003 occurred in spite of high average richness, primarily because of the high dominance of walleye pollock, which increased from an average of 18% of the catch per haul in 1995-98 to 30% in 2003, but decreased again to an average of 21% in 2004.  The increase in species richness, which was particularly pronounced on the middle shelf, has been attributed to subarctic species spreading into the former cold pool area as the extent of the cold pool has decreased over recent decades (Mueter & Litzow 2008).  However, species diversity has been low in recent years, compared to the 1990s, which suggests that species remain patchily distributed such that a given haul may be dominated by one or a few species.

2.  Size Diversity

Contributed by Jennifer Boldt, University of Washington, and Shannon Bartkiw, Pat Livingston, Jerry Hoff, and Gary Walters, AFSC

Index:  Marine food web relationships are strongly influenced by animal size.  One important indicator of the diversity of animal size in the food web is the slope of the community size spectrum (CSS).  The CSS examines the relationship between abundance and size of animals in a community, and has been found to explain some fishing-induced changes at a system-wide level.  For example, in an exploited fish assemblage, larger fish generally suffer higher fishing mortality than smaller individuals and this may be one factor causing the size distribution to become skewed toward the smaller end of the spectrum (Zwanenburg 2000), leading to a decrease in the slope of the size relationship over time with increasing fishing pressure.  The community size spectrum slopes and heights were estimated for the Bering Sea fish community using data from standard NMFS bottom trawl survey, 1982-2006 (Boldt et al., in review).

Status and Trends:  There were no linear trends or step-changes in the eastern Bering Sea fish CSS heights (Boldt et al., in review).  The EBS CSS slopes did not have a significant linear trend, but significant step changes indicate the slope was lower (less negative) during 1984-2005 (Boldt et al., in review).  

Factors Causing Trends:  Changes in CSS slopes and intercepts reflect changes in fish size and abundance, respectively, and can be due to fishing intensity and/or climate variability.  CSS slopes and heights vary temporally for different groups of taxa that are exposed to different levels of exploitation (Boldt et al., in review).  These changes in CSS slopes and heights were not due to significant shifts in species composition and not correlated with fishing intensity or bottom temperature variability (Boldt et al., in review). 

Implications:  Unlike other marine ecosystems, the eastern Bering Sea CSS indicates that there has not been a linear decreasing trend in groundfish size or abundance during 1982-2006 (Boldt et al., in review).  In fact, there were more large fish in the latter part of the times series, which is contrary to expectations if fishing were removing large individuals. 

3.  Groundfish Status

Index:  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/statusoffisheries/SOSmain.htm).  The FSSI will increase as overfishing is ended and stocks rebuild to the level that provides maximum sustainable yield. The FSSI is calculated by assigning a score for each fish stock based on the following rules:

            1. Stock has known status determinations:

                        a) overfishing 0.5

                         b) overfished 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 maximum sustainable yield (MSY) 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. The value of the FSSI is the sum of the individual stock scores.  In the Alaska Region, there are 35 FSSI stocks and an overall FSSI of 140 would be achieved if every stock scored the maximum value, 4.

Status and Trends:  The current overall Alaska FSSI is 114.5 of a possible 140, based on updates through June 2008.  The overall Bering Sea score is 68.5 of a possible maximum score of 88.  The BSAI groundfish score is 48.5 of a maximum possible 52 and BSAI king and tanner crabs score 20 of a possible score of 36.   The Gulf of Alaska groundfish score is 42 of a maximum possible 48.  The sablefish, which are managed as a BSAI/GOA complex, score is 4.  Since the inception of the FSSI index in 2005, scores expressed as a proportion of the total possible scores have been above 0.88 and 0.93 for GOA and BSAI groundfish, respectively, and 0.5 or higher for BSAI king and tanner crabs.

Factors Causing Trends:  Groundfish FSSI scores are high because it is thought that they are conservatively managed.  No BSAI or GOA groundfish stock or stock complex is overfished and no BSAI or GOA groundfish stock or stock complex is being subjected to overfishing.  Halibut is a major stock (but a non-FSSI stock, since it is jointly managed by PFMC and NPFMC) that is not subject to overfishing, is not approaching an overfished condition, and is not considered overfished.  The groundfish stocks that had low scores in the BSAI include rougheye rockfish (1.5).   The reasons for this low score are:  it is undefined whether this stock is overfished and unknown if it is approaching an overfished condition.  The stocks that scored low in the GOA are shortspine thornyhead rockfish (indicator species for thornyhead rockfish complex) and yelloweye rockfish (indicator species for demersal shelf rockfish complex), which both scored 1.5.  The reasons for these low scores are:  it is undefined whether these species are overfished and unknown if they are approaching an overfished condition.   One BS crab stock is considered overfished:  Pribilof Island blue king crab.  Three stocks of crabs are under continuing rebuilding plans:  BS snow crab, Pribilof Island blue king crab, and St. Matthew Island blue king crab.  The EBS Tanner crab stock is considered rebuilt.

Implications:  The majority of Alaska groundfish fisheries appear to be sustainably managed. 

4.  Number of endangered or threatened species

With contributions from Shannon Fitzgerald, Lowell Fritz, Kathy Kuletz, Marcia Muto, Elizabeth Sinclair, and Ward Testa, NFMS

Index:  Another measure of diversity in ecosystems in the number of species that are listed as threatened or endangered through the Endangered Species Act (ESA).  The list of threatened and endangered species below was reported on the U.S. Fish and Wildlife service (http://ecos.fws.gov/tess_public//pub/stateListingAndOccurrence.jsp?state=AK, August 22, 2008) and on the NOAA Fisheries Office of Protected Resources (http://www.nmfs.noaa.gov/pr/species/mammals/, August 22, 2008).  To have a proactive approach to the conservation of species, we also list species of concern, which are those species about which NOAA's National Marine Fisheries Service (NMFS) has some concerns regarding status and threats, but for which insufficient information is available to indicate a need to list the species under the Endangered Species Act (ESA).  Depleted stocks are those listed under the Marine Mammal Protection Act.  Some species that may or may not be listed here have been officially proposed as either threatened or endangered in a Federal Register notice after the completion of a status review and consideration of other protective conservation measures (e.g., Cook Inlet beluga whales).  Additionally, bearded, ribbon, ringed, and spotted seals are candidate species (i.e., being considered for listing as endangered or threatened under the ESA).  Conservation status of seabirds are taken from the U.S. Fish & Wildlife Service (USFWS) Migratory Bird Management  Nongame Program Alaska seabird information series (http://alaska.fws.gov/mbsp/mbm/seabirds/pdf/asis_complete.pdf; Denlinger 2006).

Status and Trends:  There are 9 species listed as endangered and 5 species that are listed as threatened in Alaska.  Three marine mammal species are considered depleted and three species of birds are considered species of concern.  The USFWS considers three seabird species as highly imperiled in Alaska:  black-footed albatross, red-legged kittiwakes, and Ancient murrelets.  Also, the USFWS considers seven seabird species in Alaska of high concern:  Laysan albatross, pelagic cormorants, red-faced cormorants, Arctic terns, marbled murrelets, Kittlitz’s murrelets, and Cassin’s auklets.  Ten seabird species in Alaska are of moderate concern:  Northern fulmars, Leach’s storm-petrels, black-legged kittiwakes, Aleutian terns, black guillemot, pigeon guillemot, Least auklets, whiskered auklets, crested auklets, and horned puffins.  Low to moderate concern was identified for parasitic jaegers and herring gulls in Alaska.  Low concern was identified for fork-tailed storm-petrels, Pomarine jaegers, Sabine’s gulls, common murres, Parakeet auklets, and Rhinoceros auklets in Alaska.  Fourteen other seabird species in Alaska are not of concern or do not have a conservation status.  Two endangered fish species that migrate to Alaskan waters include Lower Columbia River chinook salmon and upper Willamette River chinook salmon.

Factors Causing Trends:  Exploitation in the early part of the 20^th century reduced populations of large whales, such as North Pacific right, blue, fine, sei, humpback, sperm whales and minke, and sea otters to the point of depletion.  Relatively recent surveys suggest that humpback, fin, and minke whales were abundant in old whaling grounds (Zerbini et al. 2004).  Currently, potential causes of declines in marine mammals include direct takes in fisheries, resource competition, indirect competition, and environmental change (see Steller sea lion section below).  Reduced polar bear numbers have been attributed to climate change and the loss of sea ice, representing a loss of habitat, in the Arctic.  Trends in seabird populations may be related to fishery mortality, climate variability, predation, nesting habitat destruction, prey availability, and/or food provisioning (see Seabirds, this report).  Bycatch of salmon in Alaska has the potential to affect the endangered lower Columbia River and upper Willamette River chinook salmon, but is closely monitored. 

5.  Steller sea lion non-pup counts and pup production

Contributed by Lowell Fritz and Elizabeth Sinclair, NMML

Indices:  The western stock, which occurs from 144°W (approximately at Cape Suckling, just east of Prince William Sound, Alaska) westward to Russia and Japan, was listed as “endangered” in June 1997 (62 Federal Register 24345, May 5, 1997).  The eastern stock, which occurs from Southeast Alaska southward to California, remained classified as threatened (since 1990).  To elucidate trends in Steller sea lion stocks, non-pup counts and pup production are two indices that are monitored.  Population assessment for Steller sea lions is currently achieved by aerial photographic surveys of non-pups (adults and juveniles at least 1 year-old) and pups, supplemented by on-land pup counts at selected rookeries each year.  Trends in the non-pup western stock in Alaska are monitored by surveys at groups of ‘trend sites’ (all rookeries and major haul-outs) that have been surveyed consistently since the mid-1970s (N=87 sites) or 1991 (N=161 sites).  To investigate spatial differences in population trends, counts at trend sites within sub-areas of Alaska are monitored.

Status and Trends:  NMFS estimated that the western Steller sea lion population increased approximately 11-12% from 2000 to 2004 (Fritz and Stinchcomb 2005).  Although counts at some trend sites are missing for both 2006 and 2007, available data indicate that the size of the adult and juvenile portion of the western Steller sea lion population throughout much of its range in Alaska has remained largely unchanged between 2004 (N=23,107) and 2007 (N=23,118).  This was the same general conclusion reached following the incomplete survey of 2006. However, there are significant regional differences in recent trends: increases between 2004 and 2007 in the eastern AI (E ALEU), western Gulf of Alaska (W GULF) and central GULF (C GULF) have largely been offset by decreases in the eastern-central AI (eastern C ALEU) and eastern GULF (E GULF).  Winship and Trites (2006) also noted that significant differences in regional trends could affect the species’ ability to occupy its present range in the future. 

Steller sea lion pup production at western stock trend rookeries in the Kenai to Kiska area (C GULF west through C ALEU) declined 40% in the 1990s.  However, from 2001 to 2005, there were small increases in pup numbers of 4% (+265 pups) at trend rookeries in the Kenai to Kiska area and 3% (+239 pups) across the range of the western stock in Alaska.  These recent trends in pup counts, while encouraging, were less than those observed in non-pup counts from 2000 to 2004, which increased 11-12% (Fritz and Stinchcomb 2005).  The ratio of pups to non-pups (at trend sites) has declined steadily since the early 1990s, and may reflect a decline in the reproductive rates of adult females (Holmes and York 2003, Holmes et al., in press).

Factors Causing Trends:

NMFS, along with its research partners in the North Pacific, is exploring several hypotheses to explain these trends, including climate or fisheries related changes in prey quality or quantity, and changes in the rate of predation by killer whales. 

There is both direct and indirect overlap in the species and size of primary prey consumed by marine mammals and targeted in commercial fisheries.  For example, adult and juvenile walleye pollock are both consumed by adult and juvenile Steller sea lions (Merrick and Calkins 1996, Sinclair and Zeppelin 2002, Zeppelin et al. 2004).  The hypothesis is that either direct or indirect competition for food with commercial fisheries may limit the ability of apex predators to obtain sufficient prey for growth, reproduction, and survival (NRC 1996).  In the case of Steller sea lions, direct competition with fisheries may occur for walleye pollock, Atka mackerel, salmon, and Pacific cod (Calkins and Pitcher 1982, Sinclair and Zeppelin 2002, Zeppelin et al. 2004).  Competition may also exist where marine mammal foraging areas and commercial fishing zones overlap.  More difficult to identify are the indirect effects of competition between marine mammals and fisheries for prey resources.   Such interactions may limit foraging success through localized depletion (Lowe and Fritz 1996), destabilization of prey assemblages (Freon et al. 1992, Nunnallee 1991, Laevastu and Favorite 1988), or disturbance of the predator itself. 

There is considerable uncertainty on how and to what degree environmental factors, such as the 1976/77 regime shift (Benson and Trites 2000), may have affected both fish and marine mammal populations.  Some authors suggest that the regime shift changed the composition of the fish community resulting in reduction of prey diversity in marine mammal diets (Sinclair 1988, Sinclair et al. 1994, Piatt and Anderson 1996, Merrick and Calkins 1996), while others caution against making conclusions about long-term trends in Steller sea lion diets based on small samples collected prior to 1975 (Fritz and Hinckley 2005).  Shima et al. (2000) hypothesized that the larger size and restricted foraging habitat of Steller sea lions, especially for juveniles that forage mostly in the upper water column close to land, may make them more vulnerable than other pinnipeds to changes in prey availability, and spatial and temporal changes in prey, especially during the critical winter time period.  Determining the individual magnitudes of impacts that fisheries and climate changes have had on localized prey availability for foraging marine mammals is difficult; however, this interaction warrants research consideration and may require large-scale experimentation, as proposed by the National Research Council (NRC 2003) and the Steller Sea Lion Recovery Team (NMFS 2006), to unravel. 

6.  Northern fur seal pup production

Contributed by Lowell Fritz, NMML

Index:  Northern fur seals were listed as depleted under the 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 1993).  Fisheries regulations were implemented in 1994 (50 CFR 679.22(a) (6)) to create a Pribilof Islands Area Habitat Conservation Zone, in part, to protect the 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 1993).  The population size and trends of northern fur seals on the Pribilof Islands are estimated by NMFS biennially using a mark-recapture method (shear-sampling) on pups of the year. 

Status and Trends:  NMFS estimated that 127,008 pups were born on the Pribilof Islands in 2006: 109,937 (SE = 1,521) pups were born on St. Paul Island and 17,070 (SE = 144) pups were born on St. George Island. Pup production on St Paul Island has been declining since the mid-1990s (Towell et al. 2006), and was 43% less in 2006 than in 1994.  Pup production on St George was relatively stable between 2002 and 2006, but declined 23% between 1994 and 2006.  Estimated pup production on both Pribilof Islands in 2006 was similar to the level observed in 1916; however the population trend at the beginning of the 20th century was much different than at beginning of the 21st.  In 1916, the northern fur seal population was increasing at approximately 8% per year following the cessation of extensive pelagic sealing, while currently (1998 through 2006), pup production on both Pribilof Islands is estimated to be decreasing at approximately 6% per year.  The trend in pup production on Bogoslof Island in the 1990s has been opposite those observed on the Pribilofs.  Pup production increased at approximately 20% per year on Bogoslof Island between 1995 and 2007. 

Factors Causing Trends:  The increase in pup production rate on Bogoslof Island is faster than what could be expected from a completely closed population of fur seals, indicating that at least some of it is due to females moving from the Pribilof Islands (presumably) to Bogoslof to give birth and breed.  However, declines observed on the Pribilof Islands are much greater than the increase in numbers on Bogoslof, indicating that the decline on the Pribilofs cannot be due entirely to emigration.  Differences in trends between the predominately shelf-foraging Pribilof fur seals and the predominately pelagic-foraging Bogoslof fur seals are unlikely related to large-scale spatio-temporal changes in the North Pacific Ocean (e.g., regime shifts, Pacific Decadal Oscillation), since these populations are almost entirely sympatric.

There is both direct and indirect overlap in the species and size of primary prey consumed by marine mammals and targeted in commercial fisheries (see Steller sea lions, above).  The hypothesis is that either direct or indirect competition for food with commercial fisheries may limit the ability of apex predators to obtain sufficient prey for growth, reproduction, and survival (NRC 1996).  In the case of northern fur seals, direct competition with fisheries may occur for walleye pollock and salmon (Kajimura 1984, Perez and Bigg 1986, Lowry 1982, Sinclair et al. 1994, 1996).  Competition may also exist where marine mammal foraging areas and commercial fishing zones overlap.  Female northern fur seals from the Pribilof Islands forage extensively at distances greater than 81 nm (150 km) from rookeries (Robson 2001), placing them within range of commercial groundfish vessels fishing for walleye pollock on the eastern Bering Sea shelf during the summer and fall. 

Gaps in diversity knowledge:

Information or indicators that would improve our understanding of diversity in Alaska marine ecosystems includes:

1.  an index of guild diversity

2.  trophic level of ecosystem

3.  better understanding of diversity indices and what causes trends

4.  ratio of target to nontarget fish catches

G.
Issue:  Climate
Driver:  Concern about climate change
Pressure:  Change in atmospheric forcing (resulting in changes in the ocean temperature, currents, ice extent, etc)
Status/Impacts Indices:

1.  North Pacific climate and SST indices

Contributed by Nick Bond (UW/JISAO), and Jim Overland (NOAA/PMEL)

Indices:  To examine potential effects of climate on groundfish distribution, recruitment and survival, indices of climate conditions are assessed.  Four indices of climate conditions that influence the north Pacific are:  the NINO3.4 index to characterize the state of the El Nino/Southern Oscillation (ENSO) phenomenon, the Pacific Decadal Oscillation (PDO) index (the leading mode of North Pacific sea surface temperature (SST) variability), and two atmospheric indices, the North Pacific index (NPI) and Arctic Oscillation (AO).  The NPI is one of several measures used to characterize the strength of the Aleutian low.  The AO signifies the strength of the polar vortex, with positive values signifying anomalously low pressure over the Arctic and high pressure over the Pacific and Atlantic at a latitude of roughly 45˚ N, and hence anomalously westerly winds across the northern portion of the Pacific and Alaska.  These indices, along with measures of sea surface temperature (SST) and sea level pressure (SLP) provide information on the climate conditions in the north Pacific.

Status and Trends:  The North Pacific atmosphere-ocean system from fall 2007 through summer 2008 featured relatively cool sea surface temperature (SST) along its northern flank along a band extending from the Bering Sea through the Gulf of Alaska to off the coast of California.  These SST anomalies were associated with a sea-level pressure (SLP) pattern that promoted enhanced westerly winds across most of the northern portion of the basin during fall through spring.  The SLP anomaly pattern itself is consistent with the remote forcing from the tropical Pacific.  In particular, a La Nina developed in late 2007, as signified by a negative sense for the NINO3.4 index.  Two other climate indices commonly used to represent this system, the Pacific Decadal Oscillation for the ocean, and the North Pacific index (NPI) for the atmosphere, were negative and positive, respectively, for most of the last year.  The Arctic Oscillation (AO) was also largely positive during the winter of 2008. 

Factors Causing Trends:  Large-scale atmospheric forcing causes the trends observed in these indices of climate conditions.

Implications:  Near-neutral ENSO conditions became established in the summer of 2008, and given the expectation that these conditions would persist into spring 2009, implies relatively low predictability for the North Pacific climate system in the upcoming 6-9 months.

2.  Combined standardized indices of groundfish recruitment and survival

Contributed by Franz Mueter, University of Alaska

Index:  Decadal scale varability in climate may affect groundfish survival and recruitment (Hollowed et al. 2001).   Indices of recruitment and survival rate (adjusted for spawner abundance) across the major commercial groundfish species in the Eastern Bering Sea / Aleutian Islands (BSAI, 11 stocks) and Gulf of Alaska (GOA, 11 stocks) provide an index that can be examined for decadal-scale variability. Time series of recruitment and spawning biomass for demersal fish stocks were obtained from the 2007 SAFE reports to update results of Mueter et al (2007).  Only recruitment estimates for age classes that are largely or fully recruited to the fishery were included. Survival rate (SR) indices for each stock were computed as residuals from a spawner-recruit model.  Each time series of log-transformed recruitment (logR) or SR indices was standardized to have a mean of 0 and a standard deviation of 1 (hence giving equal weight to each stock in the combined index, see below).  A combined standardized index of recruitment (CSI_R) and survival (CSI_SR) was computed by simply averaging indices within a given year across stocks. Uncertainty in the stock-specific estimates of logR and SR indices was not accounted for; therefore the most recent estimates of the combined indices should be interpreted with caution.

Status and Trends:  The CSI_R and CSI_SR suggest that survival and recruitment of demersal species in the GoA and BSAI followed a similar pattern with below-average survival / recruitments during the early 1990s (GoA) or most of the 1990s (BSAI) and above-average inidces across stocks in the late 1990s / early 2000s. Because estimates at the end of the series were based on only a few stocks and are highly uncertain, we show the index through 2004 only, the last year for which data for at least 6 stocks was available in each region. There is strong indication for above-average survival and recruitment in the GoA from 1994-2000 (with the exception of 1996, which had a very low indices) and below-average survival / recruitment since 2001. From 2001 to 2004, 9 out of 11 or 8 out of 10 stocks have had below average-CSI_SR and CSI_R indices in the GoA. In the Bering Sea, recruitment estimates were available for fewer stocks, but there was no strong indication of below average recruitment across multiple stocks until 2004, when 6 of 6 stocks had below average recruitment and 5 out of 6 stocks had below-average stock-recruit indices. Therefore there was no evidence that the conditions that led to a series of below-average recruitments in Pacific cod and walleye pollock in the Bering Sea affected other species in the same way. Besides pollock and cod only flathead sole and atka mackerel had more than one year of below-average recruitment in the period 2001-2004.

Factors Causing Trends:  Trends in recruitment are a function of both spawner biomass and environmental variability. Trends in survival rate indices, which are adjusted for differences in spawner biomass, are presumably driven by environmental variability but are even more uncertain than recruitment trends. Typically, spawner biomass accounted for only a small proportion of the overall variability in estimated recruitment. The observed patterns in recruitment and survival suggest decadal-scale variations in overall groundfish productivity in the Gulf of Alaska and Bering Sea that are moderately correlated between the two regions (CSI_R: r = 0.42; CSI_SR: r = 0.47). These variations in productivity are correlated with and may in part be driven by variations in large-scale climate patterns such as the PDO or more regional measures such as ocean temperatures. The Nov-Mar PDO index for the preceding winter was positively correlated with all of the indices, but none of the correlations were significant at the 95% level.

3.  Ice indices

Contributed by Muyin Wang, Carol Ladd, Jim Overland, Phyllis Stabeno, Nick Bond, and Sigrid Salo, PMEL/NOAA

Indices:  Sea ice extent and time of retreat in the Bering Sea, which are determined by large-scale climate factors, determine the size and location of the cold pool (water <2°C; see Volume of cold pool, below) in the Bering Sea as well as the timing and extent of the spring bloom. It is valuable to examine several indices to understand trends in ice. Two indices are the ice retreat index, which is the number of days that ice remains in a 2° by 2° box surrounding Mooring 2 in the southeastern Bering Sea, and the number of days past March 15 that ice is present in the same 2° by 2° box surrounding Mooring 2. 

Status and Trends:  The year 2008 was a third sequential year with cold temperatures and extensive springtime sea ice cover.  The Bering Sea contrasted with much of the larger Arctic which had  extreme summer minimum sea ice extents in 2007 and 2008 ( 39 % below climatology) and positive autumn 2007 surface temperature anomalies north of Bering Strait of greater than 5°C.  These three recent cold years in the eastern Bering Sea followed a sequence of warm years earlier in the century.

Factors Causing Trends:  Bering Sea climate conditions are primarily controlled by local processes through winter, spring and summer, and tend to be decoupled from the continued major sea ice loss and warming taking place throughout the greater Arctic regions.  Also, the eastern Bering Sea is characterized by large monthly, interannual, and multi-annual variability, driven by large scale climate patterns.  La Nina and a positive Arctic Oscillation (see North Pacific review) contributed to the cool pattern in 2008. 

Implications:  Despite continuing warming trends throughout the Arctic, Bering Sea climate will remain controlled by large multi-annual natural variability, relative to a small background trend due to an anthropogenic (global warming) contribution.  Over the next five years we should look for the next shift back toward warmer temperatures and less sea ice.

4.  Volume of cold pool

Contributed by  Jim Overland, Muyin Wang, Carol Ladd, Phyllis Stabeno, Nick Bond, and Sigrid Salo, PMEL/NOAA and Troy Buckley, Angie Greig, and Paul Spencer, NMFS

Index:  The Bering Sea cold pool, defined by temperatures < 2°C, influences the mid-water and near-bottom biological habitat, groundfish distribution, the overall thermal stratification, the timing of the spring phytoplankton bloom, and the mixing of nutrient-rich water from depth into the euphotic zone during summer.  It is hypothesized that the timing of the spring bloom, as influenced by the presence of ice and water temperature, influences secondary production and, hence, groundfish survival and recruitment (Oscillating Control Hypothesis; Hunt et al. 2002).  Warm conditions tend to favor pelagic over benthic components of the ecosystem (Hunt et al. 2002, Palmer 2003).  

Status and Trends:  In the summers of 2006-2008, the extent of the cold pool increased from low values observed during 2000-2005.  The volume of the cold pool, which includes midwater layers, also increased in 2006.  The center of the cold pool is located further to the southeast during the cold years (Spencer, in press). 

Factors Causing Trends:  Sea ice extent and time of retreat (see Ice indices, above), which are determined by large-scale climate factors, determine the size and location of the cold pool in the Bering Sea. 

Implications:  Changes in the cold pool could affect the summer distribution of groundfish.  For example, subarctic and arctic species that moved further north in warm years (Mueter and Litzow 2008) could move south.  Changes in the cold pool could also affect the distribution and feeding migration of walleye pollock, because they tend to avoid the cold pool (Francis and Bailey 1983) and their feeding migration is delayed in colder years (Kotwicki et al. 2005).  Also, flathead sole and rock sole, which tend to be distributed further northwest in warm years relative to cold years (Spencer in press), could move further south.  The cold pool can also affect the spatial overlap between predators and prey, such as predatory Pacific cod and juvenile snow crab, thereby affecting predation mortality (Livingston 1989).  These effects in combination with others, such as changes in stratification, production, and community dynamics, however, are largely unknown.

5.  Summer zooplankton biomass

Contributed by Jeff Napp, NMFS, and Atsushi Yamaguchi, Hokkaido University, Japan

Index:  Summer zooplankton biomass data are collected in the eastern Bering Sea by the Hokkaido University research vessel T/S Oshoru Maru.  The time series (up to 1998) was re-analyzed by Hunt et al. (2002) and Napp et al. (2002) who examined the data by oceanographic domain.  The data continues to be collected annually. 

Status and Trends:  Up to 1998 there were no discernable trends in biomass anomalies in the time series for any of the four geographic domains (Napp et al. 2002).  However, the updated time series depicts a strong decrease in biomass during 2000-2004.  There was a strong decrease in biomass 2000 to 2004 or 2005 depending on the region.  The biomass now appears to be increasing, although the number of observations in some of the regions is very low. What is remarkable is that the trends appear to occur in all four domains although the initiation or time of the end of a trend may be slightly different.

Factors Causing Trends:  Part of the decrease in biomass over the middle shelf was most likely due to recent decreases in the abundance of Calanus marshallae, the only “large” copepod found in that area (Hunt et al. 2008).  It is not clear what might be the cause of declines in other regions. 

Implications:  It is possible the increased biomass of zooplankton in recent years could positively affect the growth and, hence, survival and recruitment of planktivorous fish.  

Gaps in climate-related knowledge:

Information or indicators that would improve our understanding of climate-related knowledge in Alaska marine ecosystems includes:

1.  knowledge of the effects of increased climate variation on ecosystem components

2.  indicators of ocean acidification and its effect on shell-building animals and their predators

3.  indicators of harmful algal blooms and their effects on ecosystem components.

Table 2.  Species and stock composition of guilds in the eastern Bering Sea guild analysis, and percent biomass according to 2007 surveys/stock assessment biomass estimates.

Guild	Species	Percent of 2007 biomass				Percent of 2007 biomass
Apex predators	P. Cod	30.1%		Pelagic foragers	W. Pollock	60.0%
	Arrowtooth	28.9%			W. Pollock_Juv	16.8%
	Grenadiers	12.6%			Myctophidae	3.5%
	Alaska skate	9.5%			Misc. fish shallow	3.0%
	Lg. Sculpins	6.7%			Herring	2.8%
	P. Halibut	3.9%			Squids	2.4%
	Gr. Turbot	2.5%			Fin Whales	1.9%
	Other skates	1.3%			Sandlance	1.9%
	Kamchatka fl.	1.2%			Eulachon	1.6%
	Sleeper shark	0.9%			Oth. managed forage	0.8%
	N. Fur Seal	0.4%			Scyphozoid Jellies	0.8%
	Wintering seals	0.4%			Herring_Juv	0.7%
	Minke whales	0.3%			Bathylagidae	0.7%
	Sablefish	0.3%			Capelin	0.6%
	Sperm and Beaked Whales	0.2%			Atka mackerel	0.5%
	Resident seals	0.2%			POP	0.5%
	Belugas	0.2%			Oth. pelagic smelt	0.5%
	Murres	0.1%			Salmon returning	0.3%
	Misc. fish deep	0.1%			Atka mackerel_Juv	0.2%
	Porpoises	0.0%			Northern Rock	0.1%
	Rougheye Rock	0.0%			Salmon outgoing	0.1%
	Steller Sea Lion	0.0%			Humpbacks	0.1%
	Resident Killers	0.0%			Other Sebastes	0.0%
	Sea Otters	0.0%			Bowhead Whales	0.0%
	Kittiwakes	0.0%			Sei whales	0.0%
	Fulmars	0.0%			Gr. Turbot_Juv	0.0%
	Puffins	0.0%			Sablefish_Juv	0.0%
	Shearwater	0.0%			Right whales	0.0%
	Kamchatka fl._Juv	0.0%			Auklets	0.0%
	N. Fur Seal_Juv	0.0%			Sharpchin Rock	0.0%
	Cormorants	0.0%			Dusky Rock	0.0%
	Transient Killers	0.0%		Pelagic production	Pelagic Detritus	97.8%
	Gulls	0.0%			Pelagic microbes	1.3%
	Albatross Jaeger	0.0%			Sm Phytoplankton	0.8%
	Steller Sea Lion_Juv	0.0%			Lg Phytoplankton	0.1%
	Storm Petrels	0.0%			Macroalgae	0.0%
Benthic foragers	YF. Sole	27.9%		Shrimp	Pandalidae	83.5%
	N. Rock sole	24.0%			NP shrimp	16.5%
	AK Plaice	21.9%		Structural epifauna	Urochordata	55.0%
	FH. Sole	11.7%			Hydroids	15.8%
	Other sculpins	4.5%			Sea Pens	11.9%
	Misc. Flatfish	2.9%			Anemones	9.6%
	YF. Sole_Juv	1.6%			Sponges	7.3%
	P. Cod_Juv	1.3%			Corals	0.3%
	N. Rock sole_Juv	1.3%		Mesozooplankton	Euphausiids	79.5%
	FH. Sole_Juv	1.3%			Pelagic Amphipods	7.8%
	Walrus Bd Seals	0.8%			Mysids	6.3%
	Rex Sole	0.4%			Chaetognaths	3.1%
	Gray Whales	0.2%			Gelatinous filter feeders	2.2%
	Shortraker Rock	0.1%			Pteropods	1.0%
	Shortspine Thorns	0.0%			Fish Larvae	0.1%
	Greenlings	0.0%		Motile epifauna	Brittle stars	27.2%
	P. Halibut_Juv	0.0%			Urchins dollars cucumbers	19.7%
	Dover Sole	0.0%			Sea stars	16.3%
	Arrowtooth_Juv	0.0%			Eelpouts	10.0%
Benthic production	Benthic Detritus	99.7%			Hermit crabs	7.4%
	Benthic microbes	0.3%			Opilio	6.1%
Infauna	Bivalves	83.3%			Snails	4.5%
	Benthic Amphipods	6.0%			Misc. crabs	4.1%
	Misc. Crustacean	5.1%			Bairdi	2.6%
	Polychaetes	3.6%			King Crab	1.7%
	Misc. worms	1.9%			Octopi	0.4%
Discards and offal	Offal	82.9%		Copepods	Copepods	100.0%
	Discards	17.1%

Conclusions

Climate:  Monitoring climate variability is necessary to understanding changes that occur in the marine environment and may help predict potential effects on biota.  Near-neutral ENSO conditions became established in the summer of 2008 and these conditions are expected to persist into spring 2009, implying a low predictability for the North Pacific climate system in the upcoming 6-9 months.  Large scale climate factors resulted in relatively cool sea surface temperatures in the GOA and BS in the fall 2007 through spring 2008. These large-scale climate factors also determine the size and location of the cold pool in the Bering Sea.  In the summers of 2006-2008, the extent of the cold pool increased from low values observed during 2000-2005.  Changes in the cold pool size and location may affect the distribution of some fish species and may also affect stratification, production, and community dynamics in the Bering Sea.  Observed changes in the physical environment in the Bering Sea may be, in part, responsible for the increased zooplankton biomass observed in the last two or three years.  The increased zooplankton biomass may have positive effects on zooplanktivorous fish, such as juvenile walleye pollock, in the Bering Sea.  It is apparent that many components of the Alaskan ecosystems respond to variability in climate and ocean dynamics.  Predicting changes in biological components of the ecosystem to climate changes, however, will be difficult until the mechanisms that cause the changes are understood (Minobe 2000).   

Habitat:  It is difficult to assess the effects of fishing on habitat and HAPC biota.  Increased knowledge of habitat disturbance as a function of fishing intensity would improve our ability to assess this objective.  Also, it would be beneficial to have improved knowledge of the importance of HAPC biota as habitat for different species and life stages of fish, estimates of HAPC biota population abundance and distribution, particularly in areas currently untrawlable with standard survey gear, the relationship between physical factors such as sediment type, bathymetry, and oceanography and the abundance and distribution of HAPC biota, and an index that reflects the amount of fish habitat that is damaged by fishing gear

Diversity:  Measures of diversity are subject to bias and we do not know how much change in diversity is acceptable (Murawski 2000).  Furthermore, diversity may not be a sensitive indicator of fishing effects (Livingston et al. 1999, Jennings and Reynolds 2000).  We, therefore, attempted to look at a variety of indicators for the diversity objective.  In the GOA both average species diversity and local richness showed very similar trends, suggesting that relative species composition (evenness) was relatively stable.  In contrast, trends in species diversity in the EBS differed markedly from those in richness.  Changes in BS species richness have been attributed to changes in subarctic fish species distribution relative to the cold pool (Mueter & Litzow 2008).  BS species diversity has been low in recent years, suggesting that species remain patchily distributed such that a given haul may be dominated by one or a few species.  With regards to size diversity of fish in the Bering Sea, unlike other marine ecosystems, there has not been a linear decreasing trend in groundfish size or abundance during 1982-2006 (Boldt et al. in review).  No groundfish species is overfished or subject to overfishing; however, Pribilof Island blue king crab are considered overfished.  These indices, however, apply only to fish and invertebrate species.  There are eight endangered and five threatened marine mammal and seabird species in Alaska.  One of those endangered species is the western stock of Steller sea lions, of which, the adult females may be experiencing declines in reproductive rates since the early 1990s (Holmes and York 2003, Holmes et al., in press).  The number of northern fur seal pups born on the Pribilof Islands and Bogoslof Island show opposite trends, which can not be explained by immigration/emigration, or large-scale spatio-temporal environmental changes in the North Pacific Ocean.  Further research is needed to improve our understanding of diversity indices and what causes some of these trends.

Predator-prey relationships and energy flow:  Unlike other regions, such as the Northwest Atlantic, the FIB index and the trophic level of the catch in the EBS, AI, and GOA have been relatively constant and suggest an ecological balance in the catch patterns.  Further examination supports the idea that fishing-down the food web is not occurring in Alaska, and there does not appear to be a serial addition of lower-trophic-level fisheries in the BS or GOA.  Recent exploitation rates on biological guilds in the Bering Sea are within one standard deviation of long-term mean levels.  An exception was for the forage species of the Bering Sea (dominated by walleye pollock) which has relatively high exploitation rates 2005-2007 as the stock declined.  The 2008 and 2009-recommended catch levels are again within one standard deviation of the historical mean.  This is a more direct measure of catch with respect to food-web structure than are trophic level metrics. 

Gaps in knowledge:  There are gaps in understanding the system-level impacts of fishing and spatial/temporal effects of fishing on community structure and prey availability.  Validation and improvements in system-level predator/prey models and indicators are needed along with research and models focused on understanding spatial processes.  Improvements in the monitoring system should include better mapping of corals and other benthic organisms, development of a system for prioritizing non-target species bycatch information in groundfish fisheries, and identification of genetic subcomponents of stocks.  In the face of this uncertainty, additional protection of sensitive or rare ecosystem components such as corals or local spawning aggregations should be considered.   Improvements in understanding both the nature and direction of future climate variability and effects on biota are critical.  An indicator of secondary production or zooplankton availability would improve our understanding of marine ecosystem dynamics and in prediction of groundfish recruitment and survival.

Conclusions and future research needs:  No significant adverse impacts of fishing on the ecosystem relating to predator/prey interactions and energy flow/removal, diversity, or habitat are noted.  There are, however, several cases where those impacts are unknown because of incomplete information on population abundance of certain species such as forage fish or HAPC biota not well-sampled by surveys.  Identification of thresholds and limits through further analyses, research, and modeling is also needed to identify impacts.  Also, not included in this assessment was an objective that addressed socio-economic factors.  This is something that should be included in future drafts.