Sara Miller Presentation to the PCCRC board January 21st , 2010 Physical Mechanisms for Variation in Pink Salmon Survival in the Coastal Gulf of Alaska and Prince William Sound Environment Sara Miller Presentation to the PCCRC board January 21st , 2010

Outline Optimal Stability ‘Window’ Hypothesis Part I: Gulf of Alaska Methods Results Part II: Prince William Sound Conclusions Future Projects

Optimal Stability ‘Window’ Hypothesis Introduction Optimal Stability ‘Window’ Hypothesis Zooplankton stocks, recruitment PRIMARY PRODUCTION Water Column Stability (Strength of the Aleutian Low) (Gargett 1997)

Introduction Strong Aleutian Low NOAA-CIRES/Climate Diagnostics Center

Optimal Stability ‘Window’ Hypothesis Introduction Optimal Stability ‘Window’ Hypothesis Zooplankton stocks, recruitment PRIMARY PRODUCTION Water Column Stability (Strength of the Aleutian Low) (Gargett 1997)

Introduction Previous tests of the OSH (Gulf of Alaska) Support link between primary production and GOA stability (satellite data) Increased stability linked to increased magnitude (increased chl-a and longer bloom duration) and earlier onset of the spring bloom (Henson 2007) July Chlorophyll Concentration (Henson 2007)

Introduction Previous tests of the OSH (Eastern Bering Sea) As stability past the ‘optimal’ level, post-bloom production on the middle shelf , large grazers can not be supported reorganization of trophic levels YOY pollock from the middle shelf showed a dietary shift large small copepods in 2004 relative to 1999 Accompanied by a 3-fold increase in stability in 2004 relative to 1999 (Coyle et al. 2008) 2004 Based on Mooring data (M2) provided by Phyllis Stabeno, PMEL, NOAA Stability index (J/m3) 1999 Day of Year

Part I: GOA

Hypotheses Hypothesis 1: Fish condition and growth have a positive relationship with water column stability in the GOA. Hypothesis 2: Survival is positively related to fish condition, fish growth, and stability in the GOA. Condition or Growth Stability Survival Condition or Growth or Stability

Methods Hatcheries Samples GAK 1 Mooring Summer (2000-2004) www.ims.uaf.edu/gak1/ Hatcheries Samples Summer (2000-2004)

Methods Fish Variables Condition Index Standardized Growth Exponent n=376 hauls Standardized Growth Exponent n=284 hauls Haul used as the sampling unit -year, watermass, origin

Alaska Coastal Current Methods Identification of Watermasses: Summer Shelf Watermass ACC Watermass Transect stations separated into two watermasses (ACC and Shelf) based on salinity at surface ACC (salinity<30 at 2m) Shelf (salinity ≥ 31.5 at 2m) Outer Shelf ~150km ~40 km Alaskan Stream Alaska Coastal Current (After Weingartner 2007)

Methods Stability index calculated using Simpson et al. 1977 Work required to redistribute the mass during complete mixing (stratification) Calculated down to 100 m at 1 m intervals

ACC Watermass (Inshore) Methods Gulf of Alaska Stability (Jm^-3) ACC Watermass (Inshore) Day of Year (Summer)

ACC Watermass (Inshore) Shelf Watermass (Offshore) Methods Gulf of Alaska Stability (Jm^-3) ACC Watermass (Inshore) Shelf Watermass (Offshore) Day of Year (Summer)

ACC Watermass (Inshore) Shelf Watermass (Offshore) Methods Gulf of Alaska GAK 1 Mooring Stability (Jm^-3) ACC Watermass (Inshore) Shelf Watermass (Offshore) Day of Year (Summer)

Methods Year Year GAK 1 Mooring ACC Watermass Shelf Watermass Summer Stability Residuals Summer Stability Residuals Year Year

Results *p=0.01, n.s. =0.26 Survival (%) n.s. n.s. *ACC Watermass Shelf Watermass *p=0.01, =0.26 n.s. Summer Condition Survival (%) ACC Watermass Shelf Watermass n.s. n.s. Summer Growth

Results AFK Hatchery Only *ACC Watermass Summer Stability Residuals Mooring Survival Growth Survival (%) Summer Stability Residuals p=0.02, =0.71 p=0.09, Spearman’s ρ=0.62 Summer Stability Residuals Year

Discussion ACC Alaskan Stream

Part II: PWS

Hypotheses Hypothesis 2: Condition and growth of salmon originating from hatcheries within the estuary should have a positive relationship with PWS stability during the spring and summer. Survival of estuary salmon should be positively related to fish condition, fish growth, and PWS stability. Condition or Growth Stability Survival Condition or Growth or Stability

Background Mid-Summer Summer (1997-2005) (1998, 2001-2004) Spring (1998-2004) Early Period Later Period Summer (1997-2005)

Methods Fish Variables Condition Index Standardized Growth Exponent n=694 hauls Standardized Growth Exponent n=539 hauls Haul used as the sampling unit -year, origin

Methods Stability index calculated using Simpson et al. 1977 Work required to redistribute the mass during complete mixing (stratification) Calculated down to 100 m at 1 m intervals

Methods Prince William Sound Mid-Summer Stability (Jm^-3) Spring Day of Year

Methods PWS Spring PWS Mid-Summer Stability Residuals Year

Results No clear relationship between condition (or growth) and survival. Prince William Sound n.s. n.s. Survival (%) Survival (%) Summer Growth Summer Condition

Results Spring Prince William Sound Summer p<0.01, p<0.001, Summer Condition p<0.001, =0.39 p<0.01, =0.45 Stability Residuals Stability Residuals

Results Relationship between the PWS spring and mid-summer stability and adult hatchery survivals the following year was significant and negative. *PWS Spring *PWS Mid-Summer p=0.04, =0.16 p=0.01, =0.30 Survival (%) Stability Residuals

Overall Conclusions Summer Condition Stability Spring

Overall Conclusions Summer Condition Stability SUMMER

Overall Conclusions SUMMER Summer Condition Stability Summer Growth

Future Projects June to October 1997-2004 June to October 1997-2004

Acknowledgements Milo Adkison, Lew Haldorson (UAF, co-authors) Steve Moffitt (ADFG, Cordova) Jamal Moss, Ed Farley (NOAA, OCC) Jack Piccolo (UAF, GLOBEC) Mike Malick (UAF) Jennifer Boldt (UAF, UW) Ken Coyle (UAF) Keith Criddle (UAF) Ned Cokelet (NOAA) Lisa Eisner (NOAA) Stephanie Henson (University of Maine) Richard Thomson (Institute of Ocean Sciences) Nandita Sarkar (NOAA) Dione Cuadra (ADFG) Funding: PCCRC, GLOBEC, NSF

References Coyle, K.O., A.I. Pinchuk, L.B. Eisner, and J.M. Napp. (2008). Zooplankton species composition, abundance, and biomass on the eastern Bering Sea shelf during summer: The potential role of water-column stability and nutrients structuring the zooplankton community. Deep Sea Research II. Gargett, A. (1997). The optimal stability 'window': a mechanism underlying decadal fluctuations in North Pacific salmon stocks? Fisheries Oceanography 6(2): 109-117. Henson, S.A. (2007). Water column stability and spring bloom dynamics in the Gulf of Alaska. Journal of Marine Research, 65(6):715-736. Simpson, J., D. Hughes, et al. (1977). The relation of seasonal stratification to tidal mixing on the continental shelf. Deep Sea Research 24 (Suppli.): 327–340. Weingartner, T.J. (2007). The physical environment of the Gulf of Alaska. Pg. 12-46 in R. B. Spies (editor) Long Term Ecological Change in the Northern Gulf of Alaska. Elsevier, Amsterdam.

EXTRA SLIDES

The hydrographic information was averaged over a two week time period because in the northern GOA, sigma t at 75 m has been found to have 14 to 18 day fluctuations from April to July (Niebauer et al., 1981), on average the spring bloom commences about 10 days after stability is established (Henson, 2007), which, after a short time lag, supports high densities of zooplankton, that provide food for juvenile salmon and salmon prey (Cooney and Coyle, 1988; Boldt and Haldorson, 2003).

Hatcheries release fry into PWS (After: Cooney 2007) Gulf of Alaska Late Summer Late April – Early June May-June July - Aug Summer First Winter (7-8 Months) Second Winter (11 Months) Hatcheries spawn eggs Enter GOA (Samples collected) Return to PWS Returns + Harvest /Releases=Survival % Hatcheries release fry into PWS Near Shore Areas

Methods Condition index: Using the log transformed standard allometric equation, the estimates for large fish condition were biased Suggests fish exhibiting size dependent growth LOWESS models were fit with a range of smoothness values (0.2 – 0.9) Residuals examined from each of the fits to determine the proper smoothness value Residuals (condition index) from the LOWESS model were calculated for each fish

Methods Methods Standardized Growth Exponent: Wc=weight at time of capture Wr=weight at time of release t=number of days between release and capture i=AFK, CCH, SGH, or WNH

Methods Release Weight (g) Julian Day (late April – early June)

Methods Summer Water Column Stability Calculation (Simpson et al. 1977) where h=water column depth z=vertical coordinate g=acceleration of gravity, ρ=density =work required to redistribute the mass during complete mixing (stratification) Calculated down to 100 m at 1 m intervals

Methods Onset of stratification (about May) on inner shelf (ACC watermass) depends primarily on three-dimensional processes, circulation and mixing, that are driven by winds and coastal freshwater run-off. Onset of stratification (after inner shelf) is mainly driven by one-dimensional vertical mixing and air-sea heat exchange on the outer shelf (shelf watermass) (Weingartner et al. 2005).

Background Survival (%) Year

Results

Results