Identifying Key Habitats for Juvenile Salmon in the Fox River Flats Estuary
Final Project Report T-30-3P17
Prepared for the State and Tribal Wildlife Grant Fund
by
Coowe Moss Walker1 Charles Simenstad2 Tammy Hoem Neher3 Steven J. Baird1 Jasmine Maurer1 Elizabeth Sosik2
1Kachemak Bay Research Reserve, 95 Sterling Highway, Suite 2, Homer, AK 99603, 2University of Washington, School of Aquatic and Fishery Sciences, Wetland Ecosystem, Wetland Ecosystems Team, 3NOAA Kasitsna Bay Lab, 2181 Kachemak Drive, Homer, AK 99603
Completed September 2013 Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 1
Identifying Key Habitats for Juvenile Salmon in the Fox River Flats Estuary
Final Report
Study History We studied juvenile salmon use of estuarine habitats in the Fox River estuary at the head of Kachemak Bay, Alaska. The Fox River estuary is located within the boundaries of the Kachemak Bay National Estuarine Research Reserve, and is identified as a Critical Habitat Area by the state of Alaska. Very little was known about salmon use in the estuary prior to our initial field investigations, which began in 2009. In the first year of research, we demonstrated broad spatial and temporal patterns in fish use of the various estuarine habitats. Subsequent field studies in 2010 and 2011 focused on refining understanding of select habitats that were identified as being heavily used by juvenile salmon. Emphasis was placed on understanding juvenile salmon residency in estuarine habitats along a gradient of physical characteristics, and in particular on patterns in diet and growth in the predominant juvenile salmon species; Coho (Onchorhynchus kisutch), and Sockeye (O. nerka) salmon. This work was funded through a State and Tribal Wildlife Grant from the US Fish and Wildlife Service to the Kachemak Bay Research Reserve, and was conducted in collaboration with the University of Washington, and the University of Alaska, Fairbanks.
Project Abstract This project offers the first scientific research of fish communities in the Fox River Flats Critical Habitat Area, located in Kachemak Bay, southern Cook Inlet, Alaska. Fish assemblages were assessed in a variety of habitats initially, with research becoming focused on juvenile salmon in four tidal channel habitats spanning an estuarine gradient from oligohaline near the top of the estuary to euryhaline near the mouth. Water temperature, salinity and depth in these channels were driven by varying mixing of glacial melt water from the Fox River and tidal influence. Over 14,000 fish were sampled in the four tidal channels over the duration of the project. Juvenile Coho and Sockeye salmon were the most abundant species in the channels, with both species peaking in July, with densities of 30 Coho/m and 18 sockeye/m. Both species were present from at least late April through early October. Diet samples were analyzed from 113 Coho and 104 Sockeye. Juvenile Coho and Sockeye salmon fed on 83 different prey taxa overall, with Sockeye having somewhat more compressed diets than Coho salmon. Seasonal bioenergetic models show high rates of growth in August (0.15-0.19 g g-1 d-1), likely fueled by high consumption rates and high energy density of prey. The prolonged presence and active feeding, substantiates that the Fox
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 2 River estuary provides is providing beneficial rearing and outmigration staging habitats for juvenile Coho and Sockeye salmon.
Key Words Kenai Peninsula, Alaska, glacial landscape, estuary, salmon, insect prey, diet, growth, residency, bioenergetics, saltmarsh mapping, Coho salmon, Sockeye salmon
Project Data Description of data- Data were collected in the field for water level, temperature, barometric pressure, fish species, fish counts, fish size fish diets, invertebrates, and vegetation. Fish were identified in the field and laboratory, with digital images taken as reference. Macroinvertebrate identifications were completed at the University of Washington, Wetland Ecosystem Team Lab in Seattle, WA, where a voucher collection is housed. Plant analyses were conducted by the Kachemak Bay Research Reserve staff, Homer Alaska. Format - All data were entered as Excel spreadsheets and Access database. Custodian – contact Coowe Walker, Kachemak Bay Research Reserve, 95 Sterling Highway, Suite 2, Homer, AK 99603.
Citation Walker CM, Simenstad CA, Hoem Neher T, Baird SJ, Maurer J, and Sosik E. 2013. Identifying Key Habitats for Juvenile Salmon in the Fox River Flats Estuary. State Wildlife Grant Project T-10-3 P17 Final Report.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 3 Table of Contents, including Lists of Tables and Figures
Introduction ...... 6 Objectives ...... 12 Overview of study area ...... 12 Methods ...... 15 Site Selection ...... 15 Physical Habitat Sampling ...... 17 Fish Abundance Sampling ...... 19 Laboratory Analysis; fish condition, weight and age ...... 20 Outmigration Timing ...... 21 Diet Composition ...... 22 Prey Availability ...... 22 Diel Consumption Rate ...... 24 Bioenergetics estimation of potential growth ...... 24 Salt Marsh Mapping ...... 25
Results ...... 26 Physical Habitat ...... 27 Fish Community and Abundance Data ...... 32 Juvenile Salmon Abundance ...... 33 Fish Growth and Habitat Use ...... 39 Vegetation Composition ...... 40 Prey Availability ...... 45 Fish Diet Composition ...... 51 Diel Consumption Rate ...... 63 Bioenergetics estimate of growth ...... 67
Discussion ...... 70 Future Studies ...... 72 Acknowledgements ...... 72 Literature Cited ...... 73
Figures Figure 1.View from the bluff, looking down to Fox River CHA ...... 8 Figure 2. Aerial image of Fox River Flats, including CHA boundary ...... 13 Figure 3. Sampling locations for 2009...... 16 Figure 4. Sampling locations for 2010 and 2011 ...... 17 Figure 5 Photo: collecting physical habitat data ...... 18 Figure 6. Photo: sampling with fyke nets ...... 19 Figure 7. Photo: sampling with pole seines ...... 20 Figure 8.Photo: Bismark brown dyed fish ...... 21 Figure 9. Photo: antenna array ...... 22 Figure 10. Photo: insect fall-out traps ...... 23 Figure 11. Salinity and temperature point data ...... 27 Figure 12. Seasonal salinity data ...... 28 Figure13. Interannual seasonal variability across channels ...... 29 Figure 14. 2011 seasonal temperature and depth for 4 channel ...... 30 Figure 15. 2009 fish communities in ten habitats ...... 31 Figure 16. Fish community composition in channel TS00 ...... 33 Figure 17. Fish community composition in channel TS01 ...... 34
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 4 Figure 18. Fish community composition in channel TS02 ...... 34 Figure 19. Timing of juvenile salmon abundance in TS01 for 2010 and 2011...... 35 Figure 20. Patterns of inundation and fish abundance for four channels ...... 37 Figure 21. Coho and Sockeye age classes by date ...... 38 Figure 22. Coho residency and growth determined from recaptures ...... 39 Figure 23. Salt marsh mapping of Fox River Flats ...... 40 Figure 24. Mapped vegetation communities adjacent to the four channels ...... 42 Figure 25. Percent cover of vegetation communities in plots ...... 43 Figure 26. Diversity and evenness of vegetation communities across channels ...... 44 Figure 27. NMDS plots of all vegetation communities ...... 45 Figure 28. Insect fall out trap seasonal abundance for each channel ...... 46 Figure 29. Insect fall-out trap (IFT) seasonal percent composition ...... 47 Figure 30. General abundance of IFT insect taxa summed ...... 48 Figure 31. NMDS plot of temporal associations of IFT insect compositions ...... 48 Figure 32. September 22-23, 2011 IFT insect abundance of general taxa ...... 49 Figure 33. September 22-23, 2011 IFT insect composition ...... 49 Figure 34. September 22-23, 2011 IFT insect abundance - specific ...... 50 Figure 35. September 22-23, 2011 IFT insect composition- specific ...... 51 Figure 36. Total summer hours of inundation in each of the four channels ...... 53 Figure 37. IRI diagram of Coho salmon diet composition...... 54 Figure 38. IRI diagram of Coho salmon diet composition-dominant prey ...... 55 Figure 39. NMDS diagram of juvenile salmon diet compositions ...... 56 Figure 40. IRI diagrams of juvenile salmon diet composition, TS00 ...... 56 Figure 41. IRI diagrams of juvenile salmon diet composition, TS01 ...... 57 Figure 42. IRI diagrams of juvenile salmon diet composition, TS02 ...... 58 Figure 43. IRI diagrams of juvenile salmon diet composition, TS03 ...... 59 Figure 44. Multivariate (NMDS) analysis of juvenile Coho salmon diet ...... 60 Figure 45. Multivariate (NMDS) analysis of juvenile Sockeye salmon diet ...... 60 Figure 46. NMDS analysis of juvenile Coho salmon diet across channels ...... 61 Figure 47. NMDS analysis of juvenile Sockeye salmon diet across channels ...... 62 Figure 48. Canonical Correspondence Analysis (CCA) ...... 63 Figure 49. Prey energy value of diet during growth rate experiments ...... 64 Figure 50. Consumption rate experiment prey densities ...... 64 Figure 51. Diel feeding chronology May Coho salmon ...... 65 Figure 52. Diel feeding chronology July Coho salmon ...... 65 Figure 53. Diel feeding chronology August Coho salmon ...... 66 Figure 54. Diel feeding chronology August Sockeye salmon ...... 66 Figure 55. Growth potential of juvenile Coho and Sockeye, May-Aug ...... 69
Tables Table 1. .Juvenile salmon estuarine residency documented in literature ...... 11 Table 2. Schedule of diel consumption rate sampling in TS02...... 24 Table 3.Hours of inundation for each channel ...... 52 Table 4. Estimated consumption rate values ...... 67 Table 5. Input parameters for bioenergetics model ...... 68
Appedix A: Vegetation communities mapped adjacent to tidal channels. Appendix B: 2010 diet samples for juvenile Coho and Sockeye salmon collected in channels.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 5 Introduction
Estuaries are a defining ecotone between river and ocean environments, and have been shown to provide vital habitat for juvenile salmon. For Pacific salmon, estuaries have long been known as transitional staging areas between freshwater and marine environments, but their potential as rearing habitat has evolved over recent decades (Healey 1982; Simenstad et al. 1982; Quinn 2005; Levings and Boullion 2008; Koski 2009). In Alaska, salmon are held in high esteem as a local and exportable food source with high economic value. However, there are few substantive studies on the relationships between habitat features and juvenile salmon life history and ecology in Alaska. Conservation and management of salmon amid rapid global change requires an understanding of the spatial and environmental factors that drive species use of different habitats (Sowa 2007; Milly 2008). To help address this knowledge gap, we conducted a three year investigation on juvenile salmon use of estuarine habitats in the glacially derived Fox River estuary, located in Southcentral Alaska, spanning an area of approximately 28.7 square kilometers at the head of Kachemak Bay, southern Cook Inlet. This estuary is largely within the Fox River Flats Critical Habitat Area (FRF CHA), which was legislatively designated in 1993 for the purpose of protecting fish and wildlife while providing for public access and use (Alaska Department of Fish and Game 1993). The Fox River CHA is also within the boundaries of the Kachemak Bay National Estuarine Research Reserve (KBNERR), which is dedicated to conducting estuarine research, gathering and making available information necessary for improved understanding and management of estuarine areas.
Despite the CHA designation, very little scientific data has been available for the Fox River area, particularly for fish species. Local residents and visiting anglers have long recognized the Fox River for salmon fishing, and historical accounts indicated commercial salmon fishing in the nearshore waters at the mouth of the Fox River (KBRR/NOAA 2001). A few aerial counts of adult fish in Clearwater Slough, a tributary of the Fox River were obtained in the late 1970’s- mid 1980’s by the Alaska Department of Fish and Game (ADFG) (ADFG 1976; ADFG 1982; ADFG 1983; ADFG 1985). In 1985- 86, the US Fish and Wildlife Service (FWS) conducted the only scientific study of salmon use of the Fox River (Faurot and Palmer 1992). The FWS report identifies all five species of Pacific salmon as being present in the Fox River drainage, with Sockeye (Onchorynchus nerka) and Coho (O. kisutch) salmon being most abundant (Faurot and Palmer 1992). However, their work focuses primarily on the riverine ecosystems above the estuary. No studies of the estuarine environment, where the FRF CHA is designated, had been conducted, and this lack prompted the KBNERR to conduct the work we report here.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 6
Figure 1. View from the bluff looking down into the Fox River Flats Critical Habitat Area, located at the head of Kachemak Bay. (The Bradley Lake Hydroelectric building is located on the far side of the Bay, identifiable by the white roof near the marsh edge.)
Estuaries are generally recognized as important habitats for juvenile salmon, serving as transitional zones on seaward migrations, high value feeding areas, and for predator avoidance (Groot and Margolis 1991; Thorpe 1994; Quinn 2005). As ecotones between freshwater and marine habitats, estuaries encompass a wide range of juvenile salmon habitats influenced by tides and river flow to varying degrees. These transitional habitats play an important role, buffering against osmoregulatory and physiological stress in smolts prior to ocean entry (Healey 1982; McMahon and Holtby 1992; Miller and Sadro 2003; Beamish et al. 2004; Bottom et al. 2005b), and providing opportune foraging conditions and refuge from predation (Healy 1982; Quinn 2005).
All of the five Pacific salmon species spend variable time transitioning through and rearing in estuaries (Healy 1982; Simenstad et al 1982). Juvenile salmon using estuaries show patterns of increased growth relative to their siblings utilizing freshwater habitats upstream (Miller and Simenstad 1997; Beiber 2005; Bottom et al 2005b; Gray 2005). Generally, Chum (O. keta) and Chinook (O.tshawytscha), salmon have been considered to be the most estuarine dependent (Simenstad et al 1982; Levy and Northcote 1981;
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 7 Itwata and Komatsu 1984; Neilson et al. 1985; Groot and Margolis 1991; Miller and Simenstad 1997). Research on Chinook salmon has shown increased survival rates (Magnusson and Hillborn 2003) and life history variability (Levy and Northcote 1981; Bottom et al. 2005a; Campbell 2010; Volk et al. 2010) with increased estuarine habitat use.
The use of estuarine habitats by juvenile Sockeye and Coho salmon is less well understood. Several authors document Coho and Sockeye salmon movements downstream, but fewer analyze residence times, meaning how long individual fish stay in the estuary. Table 1 highlights studies of juvenile Chinook, Chum, Coho and Sockeye salmon residency and growth in estuaries. The few studies available indicate that both Coho and Sockeye salmon potentially spend considerable amounts of time rearing in estuaries (Table 1). Powers et al. (2007) analyzed 14 years of life history data from Sockeye salmon in the Copper and Bering River deltas in Alaska, and found Sockeye salmon fry from river spawning populations are more likely to go to sea as sub-yearlings, as compared to fry from lake spawning populations. They suggest very limited freshwater rearing times, in keeping with findings by Faurot et al. 1989 and Murphy et al. 1997. However, Heifetz et al. 1989 found much longer residence times for Sockeye salmon in the Situk River estuary Southeast Alaska -of up to three months. Studies of juvenile Coho salmon use of estuaries indicate variability in the timing and duration of rearing (Crone and Bond 1976; Thedinga et al. 1993; Cornwell et al. 2001; Miller and Sadro 2003). In the past, Coho salmon were considered to be relatively ‘ocean ready’ primarily moving through estuaries as yearling smolts on their way to the ocean, however several workers have noted the presence of Coho salmon ‘nomads’; juveniles that hatch and move downstream to the estuary and ocean in their first year (Miller and Sadro 2003; Quinn 2005; Koski 2009). There is uncertainty whether these nomads rear in estuaries, and then move on to ocean environments, or go back to freshwater to overwinter (Miller and Sadro 2003), however they are recognized as representing a viable life history strategy (Koski 2009).
Estuarine environments are transitional areas between the marine and freshwater environments. They are variable and complex, ranging from upper, primarily freshwater regions with limited tidal influence, to a lower mouth that is primarily saltwater (Kaiser et al. 2005). Seasonal changes in river discharge, interacting with tidal regimes, introduces variability in freshwater and allochthonous material (such as high sediment loads or large wood debris) inputs that influence salinity gradients and thermal regimes within these zones (Mann and Lazier 2006). In many places, estuary habitats have been severely changed by anthropogenic modification, such as water withdrawals for irrigation and hydroelectric projects that alter sediment, allochthonous inputs, nutrient content, and natural freshwater flow regimes (Bottom et al.2005). These alterations can have a profound impact on species that are sensitive to changing thermal
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 8 and salinity gradients. In Alaska, anthropogenic impacts to estuaries are generally less; however, regional shifts in temperature and precipitation levels due to climate change may be amplified in northern latitudes (Hinzman et al. 2005; Bryant 2009).
We conducted field investigations in the Fox River estuary from 2009 through 2010 and 2011, with the purpose of understanding how juvenile salmon are using the estuarine habitats there, and focusing particular attention on potential connections between surrounding wetlands and salmon production. In this estuary, changes in river level from glacial melt-off coupled with the huge tidal range (8m) create spatially and temporally dynamic habitats. In comparison with most estuaries in the contiguous 48 states, the Fox River estuary has endured relatively few human impacts. However the area has been used as part of a cattle grazing lease since the 1940’s and all-terrain vehicles (ATVs) and other vehicle traffic in the estuarine intertidal wetlands is increasing along with the human population of the area. These impacts occur on top of long term changes from glacial loss due to climate warming (Baird 2009), and land level changes due to glacial rebound and sea level rise (Freymueller et al. 2008). Prior to our investigations, there was no documented information about juvenile salmon use of this, or any other estuary, in the Kenai Peninsula-Cook Inlet area. Our research, embodied in this report, is intended to provide baseline information for use by resource managers and the general public in efforts towards conservation and stewardship of the Fox River Flats Critical Habitat Area.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 9 Table 1. Estuarine residency documented in the scientific literature for juvenile Chinook, Coho, Chum and Sockeye salmon.
Juvenile Chinook, Coho, Chum and Sockeye Estuarine Residency (R) and Growth (G) R in days, except where noted/ G in mm d-1 Chinook Chum Sockeye (R/G) Coho (R/G) (R/G) (R/G) Location Reference document - - -/20.6-27.7mg/d - Big Creek, WA Bell 2001 - 40-60/ .46-.55 - - Campbell River, BC Levings et al 1986 - - -/.12-.13 - Carnation Creek, BC Tschaplinski 1987 - 1-7/-.05mm d-1 - - Chehalis River, WA Miller and Simenstad 1997 - 1-10 hours/- - - Columbia R Sather et al 2009 - 33 /- - - Columbia R. Bottom et al 2009
18.5 -28d/- Columbia R. Bottom et al 2005a - 6-83/.29-.54 - - Coos Bay, OR Fisher and Pearcy 1990 30/- - - - Copper R. AK Powers et al 2007 - 15.5 /- - - Duwamish River, WA Bostick 1955 - Weitkamp and Schadt 14-24/- - 7/- Duwamish R. WA 1982 - 30/ - - 11/- Fraser R. BC Levy and Northcote 1981 - 30/.39-.56 - - Fraser River, B.C. Bottom et al 2005 - 6-49/- - - Kachemak Bay, AK Hoem-Neher et al 2013 - 25/1.32 - - Nanaimo River, B.C. Healey 1980 - - - 5-16/- Nearts Bay, OR Pearcy et al 1989 - -/.62 - - Nitnat Estuary, BC Bottom et al 2005 - 1-43/.37 - - Puyallup River, WA Schreffler et al 1990 - 40 d/.53-1.01 - - Sacramento River, CA Bottom et al 2005 - 35/ .0218g/g//d - - Salmon River, OR Bottom et al 2005 20.24-31.25/ - .0178-.0231g/g/d - - Salmon River, OR Gray 2005 - 4.9 hours/- - - Salmon River, OR Hering et al 2010 - -/.006-.058 - - Salmon River, OR Bieber 2005 - 34.2/- - - Skagit R., Washington Beamer and Larson 2004 - Sinclair Inlet, Puget 6-59/- - - Sound, WA Fresh et al 2006 - - 90-120/- Situk R. AK Heifetz et al 1989 - 90/1.4-3.0 - - Sixes River, OR Neilson et al 1985
/0.27-0.77 Sixes River, OR Riemers 1983 - 10/ - - - Squamish Estuary, BC Ryall and Levings 1987 - - 18 -64/- - Winchester Creek, OR Miller and Sadro 2003
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 10 Project Objectives Objective 1: Map tidal channels and marsh habitats.
Objective 2: Collect discharge, salinity and temperature in each tributary as well as the main stem of the Fox River. Objective 3: Identify temporal variability in juvenile salmon use of different intertidal and fresh water habitats.
Objective 4: Index the relative abundance of juvenile salmon and monitor their growth in the tributaries sampled.
Objective 5: Examine residency time of juvenile salmonids in the tributary habitats
Objective 6: Describe and compare the relative availability of prey organisms for juvenile salmonids in the tributaries sampled.
Objective 7: Describe diet composition of juvenile salmonids in the tributaries sampled.
Objective 8: Compare seasonal growth potential of juvenile Coho and Sockeye salmon in one channel using bioenergetic modeling. Objective 9: Develop a communication plan to outreach the results from the project. Objective 10: Synthesize existing data from each year’s work into one comprehensive report, documenting details of juvenile salmon use of the Fox River estuary as rearing and outmigration habitat. Objective 11: Deliver products following the communication plan developed in 2011; including preparation of at least 2 manuscripts to be published in peer reviewed scientific journals, at least one oral and/or poster presentation at a national science conference; presentations to natural resource management agencies and local stakeholders invested in the Fox River estuary. Objective 12: Outline potential future directions, options and recommendations for research in the Fox River estuary, and/or other estuaries of the Kachemak Bay region
Study Area
The Fox River Flats is at the seaward end of a flat, steep walled, poorly drained valley that is bound by the Kenai Mountains to the southeast, and by a broad, rolling upland area, the Caribou Hills, to the northwest (KBRR/NOAA 2001). The Fox River Flats encompasses an area of approximately 28.7 square km, and 6.4 km wide in the vicinity of the river mouth, and extending 6 km from the rivers emergence into the delta to the river mouth at the Bay (Figure 2).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 11
Figure 2. The Fox River Flats estuary, located at the head of Kachemak Bay, encompasses the river deltas two glacially-derived rivers; Fox River, and Sheep Creek. This high latitude northern estuary (latitude 59°) is fed by glacial rivers, and has an extreme (8 m) tidal range. Much of the Fox River Flats is designated as a State Critical Habitat Area (outlined in red).
Regional climate patterns are derived from a combination of the continental climate (less rain, warmer summers, colder winters) that prevails to the north, and the pronounced maritime (more rain, more snow) climate that prevails to the south (KBRR/NOAA 2001). Total precipitation from 1932–2005 averages 63 cm annually (rain plus water equivalent of snow); approximately 13–18 cm of which are snow (recorded at the Homer Airport, which is the closest weather station to the Fox River Flats). Average maximum temperature occurs in July at 16.1°C, and the minimum occurs in January at -8.5°C (Western Region Climate Center, http://www.wrcc.dri.edu/cgi-bin/cliMAIN.pl?akhome).
Underlying the Fox River Flats valley are unconsolidated fluvial and glacial deposits. The entire Kachemak Bay area experienced subsidence on the order of 1.22 m as a result of the 1964 earthquake. However, a rich supply of sediments is provided to the surface of the Flats by shifting glacial rivers that
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 12 deposit a range of sediments, from silts to coarse gravels, from the mountains, so it is likely that this area will quickly aggrade back to pre-earthquake levels (Batten et al. 1978). Soils in the Fox River Flats are predominantly hydric (KBRR/NOAA 2001)
Vegetation of the Fox River Flats area is dominated by broad bands of tidal wetland plant communities, with embedded localized bands and patches in response to variation in tidal influence, drainage patterns, elevation, aspect, and substrate texture. The Fox River Flats CHA is identified as providing important habitat for birds, especially migrating waterfowl and shorebirds, and is recognized by the Western Hemisphere Shorebird Reserve Network as a ‘Site of International Importance”. Marine mammals, notably harbor seals and beluga whales are known to frequent the Fox River Flats. Terrestrial mammals known to use the area include moose, black and brown bears, coyotes, wolves, foxes, lynx, wolverine, as well as numerous small mammals (snowshoe hares, voles, shrews) (KBRR/NOAA 2001).
Historically, the Fox River Flats area has been used by several groups, including ranchers, hunters, and equestrian, and ATV enthusiasts. Access is predominantly via land, through a switchback trail, authorized by the State for pedestrian, equestrian, and ATV access only. This switchback trail is used as a primary access point for the Russian villages of Kachemak Selo, Voznesenka, and Razdolna, and villagers continually improve the trail for unauthorized vehicular access. There are also several private parcels that are being developed at the head of the Flats. Currently, there is pressure from the Russian community of Razdolna to construct a road down the bluff for easier access to their community. This would also provide another access point to Fox River Flats. Recreational ATV use on the Flats has become more prevalent over the past several years. Hunters, especially duck hunters, access the Flats seasonally by ATV, as well as the residents of the local villages and adjacent landowners. ATV tours across the Flats have recently started and are becoming increasingly popular. As the population of Homer and the surrounding area increases, use of the Flats will also increase.
Cattle have grazed on Fox River Flats since the gold rush of 1886. Cattlemen have been leasing this land from the State since 1961. The current 25-year lease ends in 2020. Several studies were conducted prior to the most recent grazing lease renewal to investigate the effects of grazing on plant communities. The studies included exclusion experiments, assessment of cattle utilization, evaluation of plant annual production, ecological site mapping (basic plant communities and soils), and visual reconnaissance assessments by several biologists. The results of these studies indicate that cattle prefer to graze the intertidal sedge communities and to use upland areas for loafing. Studies through the Natural Resources
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 13 Conservation Service have shown the Flats to have a carrying capacity of 800 head of cattle, although only 300 range the area currently (HSWCD 2010).
The Fox River Flats area is primarily managed by the Alaska Department of Fish and Game through the Kachemak Bay and Fox River Flats Critical Habitat Areas Management Plan that was adopted in 1993 to provide guidance to the Alaska Department of Fish and Game and other agencies involved in managing the CHAs. The Plan is the result of an extensive public planning process and the goals and policies have been adopted into state regulation (ADFG 1993).
Methods
Site Selection Since there were no previous studies of juvenile salmon use of the Fox River Flats, in the first year of the project, we attempted to identify sites that represented habitat types within zones of zero, low, medium, and high salinity. Our goal was to have sites that were accessible and able to be effectively sampled for juvenile fish repeatedly from May through September. We selected ten sites in four representative habitat types for repetitive sampling through spring, summer and fall: two sites represented tidal guts that emptied directly into the bay (non-river channels); four sites were in overflow side channels that filled as glacial melt increased river flows; two sites were in the main channel of the river; and two sites were in tidal channels adjacent to the main river channel (Figure 3).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 14
Figure 3. Aerial image of the Fox River estuary showing the sampling locations for 2009. NR= non-river; SCS=side channel; MS= main river channel; TCS= tidal channels.
In years two and three of the project, we focused efforts on four tidal channels that represented habitats along a salinity gradient from the head of the estuary, where the river emerges from the forest, to the river mouth at Kachemak Bay (Figure 4).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 15
Figure 4. Aerial image of the Fox River estuary showing sampling locations for 2010 and 2011, highlighted in boxes. The tidal channel sites spanned a gradient from predominantly freshwater influence (TS00, red box) located near the river’s emergence onto the delta, to predominantly tidal influence (TS03, green box) close to the mouth of the river at the Bay.
Physical Habitat Sampling In 2009, we collected point data for temperature and salinity at each sampling location and event. In 2010 and 2011, we sampled temperature and water level continuously, as well as taking point measurements for temperature, salinity and water level at each location during each sampling event.
Temperature and depth were measured and recorded using Solinst TM 3001 level loggers (Solinst Canada Ltd., Ontario, Canada) calibrated with a Solinst TM 3000 barologger set onsite. Level loggers were set at 15 min recording intervals and placed in 5 cm wide by 25 cm long plastic PVC housings attached to steel fence posts driven approximately 25 cm into the substrate. Fence posts were located five meters upstream from the channel mouth in each of the channels sampled, and one logger was placed along the margin of the river channel. In addition, measurements were taken for each sampling event at a cross-section downstream of the fence. Thalweg depth, conductivity (direct and standardized for temperature), salinity (measured as salt concentration, parts per thousand, ppt), and temperature (with probe held just below the surface, in mid-water column, and at the channel bottom) were measured using a YSI Model 30.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 16
Photo by KBRR Figure 5. Collecting habitat data (water depth, temperature, salinities) in one of the tidal channels (TS02).
For this report, habitat data were summarized for analyses as continuous water level depths for each estuary channel and the mainstem river. Continuous temperature data were summarized as daily averages. Point measurements of salinity, collected at each sampling event , were combined and expressed as monthly mean, minimum, and maximum recordings. Stationary logger data were summarized as daily mean values. Point measurements of salinity collected at each sampling event were combined and expressed as average, minimum, and maximum recordings for each sampling event. Additional statistical analyses of habitat data were included in companion studies (Hoem Neher 2013a; Hoem Neher 2013b).
Data were compared spatially using the channel locations along the tidal inundation zone from most upstream (TS00) to most downstream sampling site (TS03). We compared environmental conditions (temperature, depth, distance from low tide line, salinity) with patterns of salmon abundance for each channel to determine if environmental conditions related to patterns in fish abundance, body condition, and size.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 17 Fish Abundance Sampling
In 2009, we used two capture methods for juvenile salmon. In the non-river tidal guts, we used fyke nets, as these channels dewatered with each tide. Fyke nets were placed in the channel at full tide, and fish were collected after the tide had recessed (Figure 6). In the main stem, side channel and tidal channel habitats, we used pole seines (Figure 7). In 2010 and 2011, sites were sampled in each of four channels over consecutive days, twice per month from early May to late September. In 2010, fish abundance was sampled once per month, and the second sampling of each month was conducted to assess potential growth. For the abundance estimates, we used multiple-pass depletion methods (Hayes et al. 2007), validated to determine if they reflected actual fish abundances (see below). A 20 m length of channel was measured from the stationary logger location parallel to the channel upstream. The start and end points of each sampling unit were then obstructed with block nets (2.2 m x 6.1 m, 0.31 cm mesh) secured along the sides and bottom with stakes to prevent fish escape. Pole seines (2.2 m x 6.1 m, 0.31 cm mesh) were used to sample the site, pulled three times in the downstream direction (Figure 7). Fish from each haul were placed in separate, 19 L aerated tubs filled with water from the channel. All fish captured were identified to species and counted.
Photos by KBRR
Figure 6. Fyke nets were used to sample non-river tidal guts that dewatered with each tide. Nets were set at high tide (left). Fish that had moved into the channel and were upstream of the net were captured as the channel dewatered (right).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 18
Photo by KBRR
Figure 7. Sample reaches were blocked with nets and then multiple passes were conducted with pole seines to sample tidal channels to the main river.
Removal estimates of abundances with 95% confidence intervals were generated for each salmon species using depletion techniques for a closed population (Hayes et al. 2007). Removal estimates may be negatively biased due to declining sampling efficiency among depletion passes, and this bias can be affected by habitat conditions within sites (Rosenberger and Dunham 2005). To determine how well removal abundance estimates and total catch reflected actual fish numbers, we used mark-recapture sampling techniques as baseline measures of fish abundance once per sampling event within a single channel. Mark-recapture abundance estimates were calculated using single marking and single recapture estimates for a closed population following Hayes (2007). Fish were captured using the same methods described for depletion (three hauls of the seine net). They were then batch dyed each month with Bismarck brown mixed in concentrations of 21 mg/L (Gaines and Martin 2004; Figure 8).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 19
Photo by KBRR
Figure 8. Juvenile salmon were marked with a temporary dye, Bismark Brown, in a mark-recapture validation of our sampling techniques.
All captured salmon were placed in containers of dye solution with portable aerators for 50 minutes. Water temperature was checked for increases that could cause thermal stress to the fish at 20-minute intervals during dying. Salmon were then released into the enclosed transect and allowed to acclimate and disperse randomly within the channel for 1 to 3 hours. After recovery, the channel was resampled using the same effort (multiple pass seining), noting recapture of marked and unmarked individuals.
Laboratory analysis: fish condition, weight and age
The first 50 juvenile salmon captured from each seine haul of each species were anesthetized in 70 mg/L methane tricane sulfonate, MS-222 (Bailey et al. 1998; Chittenden et al. 2008) and measured for fork length to the nearest 1.0 mm. Up to three Coho or Sockeye salmon (not to exceed 10% of the total catch), distributed among three size classes (small, medium, and large), were randomly selected and euthanized at each site using 140 mg/L MS-222 (maximum 24 individuals each month). These fish were labeled and frozen for laboratory analysis to determine condition, weights, and age.
We also collected data from fish captured for use in a companion study (Hoem Neher et al 2013a and Hoem Neher et al 2013b). This included using water weight, wet weight, and Fulton’s condition factor (K=W*L-3 *100,000, where W= laboratory weight [g] and L= laboratory length [mm]) for metrics of condition (Jonas et al. 1996; Pope and Kruse 2007). Specimens were measured for fork length (±1 mm), then blotted and weighed to determine wet weight (± 0.01 g). Samples were placed in a 65-70 °C drying
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 20 oven for three days, weighed, and returned to the oven for 24 hours to be dried and re-weighed. Samples were considered dried when a minimal weight change was detected between consecutive daily weights (Jonas et al. 1996). Water weight was determined by subtracting the oven dried sample weight from the wet weight (Jonas et al. 1996; Sutton et al. 2000). For the companion study, we also removed and prepared sagittal otoliths from a subset of captured and sacrificed fish. Otoliths were aged after preparation for microstructure and microchemistry analysis by counting the winter annuli characterized by large, translucent rings composed of numerous, relatively small incremental growth bands (Campana and Neilson 1985). Ages generated from otolith analysis were used to validate size-at-age inferred from length frequency histograms.
For fish retained for laboratory analyses, our protocol was to sample evenly across age classes; as a result, the composition of the laboratory fish sample did not correspond to catch composition. Age class composition of the total catch was inferred via length-frequency histograms and validated with otolith age for each sampling event. Detailed analyses and results are presented in Hoem Neher et al 2013b.
Outmigration Timing
In 2011, an antenna array capable of detecting and logging PIT tags was installed in one of the tributary channels (TS01). One antenna was installed near the mouth of the channel, and another was installed approximately 10 m upstream (Figure 9). At each sampling event, we marked all juvenile Coho salmon that were > 60 mm with passive integrated transponder (PIT) tags. Juvenile Coho salmon were marked in channel TS00 as well, which was located farther up the main channel for the purpose of potentially tracking fish that may move from channel to channel. The number of fish tagged depended on the number of captured fish that were large enough.
To assess potential fish mortality due to PIT tagging, during each sampling event, we did not PIT tag five salmon, and we held these fish along with five PIT tagged fish in a covered aerated tub (screened holes to allow water flow) within the channel overnight, and assessed them the following day for survival.
At the outset of the project, we attempted to PIT tag juvenile Sockeye salmon. However, unlike the Coho, the Sockeye did not tolerate tagging well, and we ceased tagging after a few attempts for that species.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 21
Photo by KBRR Figure 9. An antenna array capable of detecting PIT tagged fish was installed near the mouth of TS01. The wooden platforms held batteries and solar panels which powered the antennas. This image is taken looking downstream towards the main channel of the Fox River.
The lower antenna was out of commission for much of the season due to heavy loads of silts, sands and gravels that the river system deposited at the mouth of the tributary. The upper antenna was operable for a longer period, however it too became inoperable due to cows rubbing against the wooden platforms and disrupting the power system.
Diet Composition
In 2010 and 2011, three juvenile salmon of each species and size class found (not to exceed 10% of the catch) were euthanized at each 20-m reach, during each sampling event using 140 mg/L MS-222. These fish were given unique identifiers to site, date, and size, photographed with their label, and were placed on ice until transported to the laboratory. In the lab, the whole fish were weighed (after being blotted dry) and the stomachs were removed from the abdominal cavity, preserved in 10% buffered formalin, and retained for later analysis. Stomach contents of the juvenile salmon retained from each channel were analyzed to determine the diet composition and their relative consumption rate. Prey organisms were identified and counted under a dissecting microscope to class, family, or order (depending on the physical state of the organism), and weighed (blotted dry). Fullness of stomachs was ranked by stomach fullness index, using a qualitative ranking ranging on a scale from 1 (empty) to 6 (distended). The total weight of
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 22 the stomach contents was obtained by adding the total weight of each prey taxa to the weight (by difference from the total fish weight) of the unidentified material. We also assigned a qualitative ranking, scaled from 1 (nothing could be identified) to 6 (100% identification), to the overall state of digestion of the stomach contents. The total weight of the stomach content was obtained after the blotting process to avoid damage to the insects prior to identification.
Prey Availability Insects potentially available as prey of juvenile salmon occupying the four tidal channel systems (TCS00- TCS03) were sampled using insect fallout traps (IFT). IFT are 0.24 m2 (50-cm x 35-cm x 14-cm) plastic bins supported on the bottom by a PVC platform and ‘coralled’ in place on the sides by PVC pipes and attached by monofilament line, such that they could float vertically with the tides. The tub was filled partially with water and biodegradable dish soap, which acts as a surfactant and prevents insects that have fallen into the bin from escaping. At the end of a 24-hr period, the contents of each IFT was sieved through a 106-µm sieve, washed and fixed in a 70% isopropanol solution. The taxa present were later identified in the laboratory under a dissecting microscope, according to a taxonomic key, usually to class, order or family level as feasible. The IFT were deployed in the vegetated marsh immediately adjacent to the tidal channel margins (Figure 10; Figure 23). This sampling technique has been shown to provide a reliable indicator of flying adult insects that are found in the diets of juvenile salmon from tidal marshes in other regions (e.g., Cordell et al. 1994; Gray et al. 2002; Lott 2004).
We focused processing and analysis of the IFT on two questions: (1) how well do the IFT sample prey of juvenile salmon in this emergent marsh system, and (2) how well do they display the variability and composition of the vegetation assemblages sampled in September?
Photo by KBRR Figure 10. Insect fallout traps were placed to accommodate the 8 m tidal range in Kachemak Bay (examples of low tide at left, high tide at right).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 23
Diel consumption rate sampling
Diel consumption rate sampling was conducted on four monthly occasions in 2011 to gather data for estimation of gastric evacuation and consumption rates (Table 2). Fish were sampled by seine approximately every three hours and processed as for diet analysis, described above. We used the Doble and Eggers (1978) method to estimate gastric evacuation, R, and the Eggers model (Eggers 1977) with a correction factor (Eggers 1979) to quantify daily ration (D; g food g-1 fish d-1): where 24 is the number of hours in a day, is the gastric evacuation rate, is the mean instantaneous ration of fish sampled regularly throughout a 24 h period, and and are the mean instantaneous rations of fish sampled at the beginning and end of the 24 h period.
For each diel consumption rate sampling, we estimated R by regressing the natural logarithm of the maximum mean weight of food in the stomach of fish captured during the night over the elapsed time to the minimum three hours:
R = (ln (Wtmin) / ln (Wtmax) * 3
where Wtmax is the maximum stomach contents weight and Wtmin is the minimum stomach contents weight three hours following. If no definable decline in mean stomach contents weight was observed, we concluded that some feeding continued during the night and thus gastric evacuation and consumption rate could not be effectively estimated (Table 2).
Table 2 . Schedule of diel consumption rate sampling in TS02, May-August 2011
Dates Time Time Number Number Number Effective: Effective: Start End Seine Coho Sockeye Coho Sockeye Samples 5/20-21/2011 0145 0445 10 21 23 Y N 6/22-23/2011 1740 0530 4 12 11 N N 7/20-21/2011 1720 1630 7 17 15 Y N 8/14-15/2011 1800 1700 8 20 19 Y Y
Bioenergetics estimation of potential growth We used the Wisconsin Fish Bioenergetics model (Hanson et al. 1997) to estimate variability in the growth potential of juvenile Coho and Sockeye salmon in tidal channel TS2 over the period May- September 2011 as a function of ambient temperature, fish size, consumption rate and diet composition. Bioenergetics models are mass balance equations where the energy stored as growth by an individual fish
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 24 can be estimated as the energy consumed minus the energy used by metabolism and lost as waste given species-specific physiological processes (Chipps and Wahl 2008): G = C - (R+S) - (F+U) where G is potential growth (g g-1 d-1), C is consumption rate (g d-1), metabolism is the cost of respiration (R) and specific dynamic action (metabolic cost of digestion, S) and wastes as egestion (F) and excretion (U).
The model can be used to simulate either consumption or growth and operates on a daily time step. In the case of juvenile Coho and Sockeye salmon in the Fox River delta, we used the data for consumption, diet energy density, fish weight, and water temperature as inputs to the model to solve for temperature- dependent growth potential. The Wisconsin Fish Bioenergetics model uses species-specific parameters derived from laboratory experiments to parameterize each of the physiological processes (Hansen et al. 1997). Consumption rates are input into the model as proportions of maximum theoretical consumption given fish weights and water temperatures (p-value; Hanson et al. 1997). We calculated p separately for Coho and Sockeye based upon species-specific parameters provided in Thornton and Lessem (1978), Beauchamp et al. (1989) and Stewart and Ibarra (1991) using empirical data for temperature, weight and daily ration (D). Estimated p values that sometimes exceeded 1.0 (100%) were set to 1.0.
Consumption rate was proportioned by the energy quality (energy density, J g-1 and indigestible fraction) of the prey consumed from the diel consumption sampling. We applied our empirical data, including measured daily ration and estimated p values, to focus effects on potential growth of differences in diet composition energy value.
Saltmarsh Mapping
Salt marsh plant communities were mapped using a simplified version of the methods used in the Lake Clark Coastal Marsh Mapping project (Tande 1996). Salt marsh plant communities surrounding the channels were delineated using high-resolution color aerial photography in ArcView GIS. Points were then placed within each polygon (generally multiple points within each polygon), and these were transferred to a hand-held GPS as waypoints. These waypoints were assessed in the field by establishing 1 meter square vegetation at each site when located. Within each plot, all plant species were recorded, and assigned percent cover values (to the nearest 5%, or trace for species that occurred in the plot in very low percent cover). Photos and notes were also taken, and polygon boundaries were adjusted when necessary.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 25
To arrive at plant community descriptions, percent cover data were reclassified into plant cover classes (Tande 1996, Mueller-Dombois and Ellenberg 1974). Field data sheets were then sorted into groups based on the single species with the highest percent cover. These groups were again sorted into associations based on other plant species with relatively high plant cover class values. Groups with similar plant species assemblages were then compared to determine whether it was appropriate to combine them. Finally, all photos from plots within each grouping were compared to ensure that they were consistent with the group. The final two steps were repeated until we were satisfied that no further combination was appropriate.
Percent cover for 14 dominant vegetation species and bare mud were plotted individually for each vegetation sample plot, tidal channel and date. In addition to basic areal cover, species richness (number of species), Shannon-Weiner diversity and evenness were also compared for the four tidal channels based on the 22-23 September (peak vegetation cover) sampling.
Results Physical Habitat In 2009, the pilot year of the study, we sampled a total of ten locations spread across four different habitat types. These habitats exhibited a wide range of physical characteristics that changed during our study. The river main channel sites were deep and fast with highly shifting sediments, low salinities and very cold temperatures. These conditions resulted when the glaciers were actively melting and contributing to stream flow. The river tidal channels were slightly warmer than the main channel, deep and had slower apparent stream flow. The overflow side channels tended to be very warm, with slow apparent velocities and relatively high salinities. Water level in these channels was dependent on main river levels, precipitation, runoff from the adjacent flats, and tidal levels. Once water entered the overflow channels it appeared to linger for long periods and increase in temperature and salinity. The non-river tidal channels were saline, deep (when flooded), and moderately warm (Figure 11).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 26
Figure 11. Salinity (ppt) and temperature (temp, °C) single point readings by site (TCS=tidal channels to the main stem; SCS= overflow side channels; MS=main; and NR=non-river associated salt marsh channels
Interpretation of these point data is restricted to a general understanding that the conditions reflected complex interactions of the daily tidal cycle, the annual glacial melt that drives river flow, and the geomorphology of the area. Viewing the data over the sampling season, May through September, the influence of the glacial melt can be observed (Figure 12). The overflow side channels, sampling station SCS04 (upstream) and SCS03 (downstream), had consistently higher salinities than other channels. This side channel receives water primarily during higher tides, and the water does not readily drain, becoming warmer and more saline, except during the peak river discharge in July, when the channel becomes primarily fresh. The other side channel, sampling stations SCS01 (upstream) and SCS02 (downstream) received overflow from the river much earlier in the season, and had less marine influence.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 27
Figure 12. 2009 Salinity (PPT) summarized by habitat type across the sampling season. Sampling locations shown above, and point data presented below. MS= mainstem; NR= non-river tidal guts; SCS= overflow side channels; TCS=tidal channels.
In 2010 and 2011, we focused efforts on four tributary habitats (TS00, TS01, TS02, TS03) to the main channel because these were the sites where juvenile salmon were most prevalent in 2009. These channels represented a tidal gradient, with TS03 being the closest to the Bay, and the most tidally influenced, and TS00 being farthest from the Bay, and primarily oligohaline. These channels were sampled for salinity through point measurements at each sampling event; and for temperature and depth through continuous loggers. We observed substantial variation in physical attributes (temperature, salinity, depth, and velocity) within and between the four channels throughout the sampling period related to tidal inundation, the rate of glacial melt, and the proximity of the sites to the main river.
Figure 13 shows the salinity profile point data for the four channels. TS00 is predominantly fresh water throughout the season, with no stratification. TS01 and TS02 both exhibit considerable stratification, with highest salinities near the bottom of the channels, and freshwater near the top of the channel, except in late July to August of both 2010 and 2011 when stratification diminishes and both channels are predominantly freshwater, reflecting glacial melt increases. TS02 had the highest observed range in salinity, ranging from over 14ppt at the channel bottom to <2ppt near the surface. TS03, the tributary closest to the Bay, showed predominantly freshwater influence, however this was an artifact of sample timing as we always sampled this channel at low tide.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 28
Figure 13. Interannual and seasonal variability represented in salinity (ppt) in the four channels in 2010 (blue dots), and 2011 (green dots).
Figures 14-17 show seasonal variability in temperature and depth for the four channels. Three of the channels; TS01, TS02 and TS03, were sampled in both 2010 and 2011. TS00 was sampled in an exploratory manner in 2010, and on a regular basis in 2011. This is the shallowest channel, only becoming full when the river flow increases from the glacial melt as summer temperatures increase, and the influence of the tidal cycle is present, though subtle. Temperatures in TS00 drop corresponding to the cool temperatures of the glacial melt water (Figure 14). Farther down in the
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 29 estuary, the influence of the tidal cycle becomes more pronounced, with monthly pulses in depth and temperatures in both TS01 and TS02. TS01, which is farther upstream than TS02, showed a stronger influence of the glacial river melt, with decreasing temperatures in July when the river was fullest. Both TS01 and TS02 provided continuously available habitats for fish throughout the sampling season (May through October), with depths greater than 1m, and temperatures consistently over 6°C. In TS01, the daily temperatures mean decreases in July, corresponding to glacial river flow, and then increases in August. TS02 maintains a higher daily mean temperature throughout the summer (Figure 14). TS03, the channel closest to the Bay, is most strongly influenced by the tidal cycle. In this channel, habitat availability changes on a daily scale, with mean depth rarely exceeding 1m, and temperatures consistently over 6°C (Figure 14).
Figure 14. 2011 seasonal variability in temperature and depth for all four channels.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 30 Fish Community and Abundance Data In 2009, a total of 4,432 fish were captured in all ten sample locations. In the main stem, non-river tidal guts, and side channel habitats, salmonids were not common, and the fish communities were dominated by sculpins (Cottus spp.), starry flounders (Platichthys stellatus), and especially sticklebacks, both nine- spined (Pungitius pungitius) and three-spined (Gasterosteus aculeatus). The river tidal channels however, contained high densities of juvenile Coho and Sockeye salmon (Figure 15). In 2009, field identification between juvenile Sockeye and chum salmon was difficult. Genetic analysis revealed that all but one of the samples submitted for analysis were Sockeye salmon so it is most likely that the fish identified as chum salmon in the field (Figure 15), were in fact Sockeye salmon.
1000
900
800 CK 700 PK 600 DV
500 CO
400 CM SO 300 SM 200
100
0 MS01 MS02 NR01 NR02 SCS01 SCS02 SCS03 SCS04 TCS01 TCS02
Figure 15. 2009 fish community composition in 10 habitats sampled. CK=Chinook, PK=pink, DV= Dolly Varden, CO=Coho, CM=chum, SO=Sockeye, SM=SFL=Starry flounder, SCU=sculpin spp., EU-eulachon, SSC=Staghorn sculpin, 9S=nine-spine stickleback, 3S=3 spine stickleback
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 31 Juvenile Salmon Abundance Data Juvenile Coho and Sockeye salmon were found in the river tidal channels as soon as we were able to access the sites in spring (late April to early May) through fall (late September). Fish abundances for all channels were quantified using total catch and multiple pass depletion estimates (removal). For a subset of channels, these numbers were compared to mark-recapture (m-r) estimates (used as a baseline measure with the exception of the July sampling event, during which one block net failed) to determine which technique (total catch or removal) most consistently corresponded to baseline measures of fish abundance. Both of these metrics had a high degree of correspondence to m-r estimates (R2 values = 0.73, 0.85 for total catch and removal estimates respectively). Both estimates were lower than the baseline m-r value (73% and 78% of m-r estimate on average for total catch and removal estimates, respectively), but were consistently so. We did not have sufficient sample sizes to examine correlates of bias (such as differences in channel size, depth, and individual sampling technique). We therefore used the uncorrected total catch for relative fish numbers with standardized effort for description and analysis. Using uncorrected total catch for relative fish numbers does not account for differences in catchability (q) related to changes in environmental conditions, however we used standardized methods of three pass removal in blocked channels at each event throughout the season yielding a relative catch per unit effort for each sampling event. Because our goal was not to estimate total abundance, but rather to compare catch with environmental conditions within each channel, we feel this measure allows for accurate comparisons across space and time.
TS00, the channel farthest from the Bay, with the least amount of tidal influence was sampled opportunistically in 2010, meaning that we sampled this habitat once the river water levels were enough to inundate it. In 2011, we scheduled regular, bi-weekly sampling in TS00. Over 3000 fish were captured in TS00 over the two years, predominantly juvenile Coho and Sockeye salmon, and also a number of Dolly Varden (Salvelinus malma) (Figure 16).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 32
Figure 16. Fish community composition in the uppermost channel (TS00) in 2010 (top) and 2011 (bottom).
Both TS01 and TS02, were sampled in all three years (2009-2011) of the project: 6429 fish were captured in TS01, and 4653 fish were captured in. Both of these channels were dominated by juvenile Coho and Sockeye salmon, although the proportion of each varied between years (Figs. 17 and 18).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 33
Figure 17. Fish community composition in TS01 in 2009, 2010 and 2011.
Figure 18. Fish community composition in TS02 in 2009, 2010 and 2011.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 34
In 2010, we sampled fish abundance monthly; and we sampled a subset of fish for growth measurements once a month. This sample design resulted in bi-monthly sampling, however we only estimated abundance once a month. In 2011, we sampled abundance bi-monthly. Due to the different sampling periods, it is difficult to compare timing and abundance between years because the peak in abundance appears to coincide with mid to late July, and we only sampled in early July in 2010 (Figure 19). In 2011, overall abundance of Coho and Sockeye salmon peaked in the three upper channels in July. In TS03, our sampling was always near low tide due to access constraints. As this channel dewaters almost completely during the tidal cycle, we are able to use our data to determine species presence, but not patterns of abundance.
Figure 19. Timing of juvenile salmon abundance in channel, TS01 in 2010 and 2011. In 2010, abundance data was collected on a monthly basis; whereas in 2011, this data was collected bi- monthly. Monthly sampling appears to have missed the peaks in abundance.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 35 Patterns in abundance in all four channels throughout 2011 are shown in Figure 20. Patterns of inundation in the uppermost channel, TS00, are dominated by rising river levels due to glacial melt; this channel is inundated only when the river rises high enough to fill it, making it available habitat for a much briefer time than any of the other channels. Juvenile salmon used this channel when the depth reached at least 0.5m. The two channels in the middle of of the gradient, TS01 and TS02, are inundated by the river even before glacial melt, making them available throughout the season. These channels have the highest overall abundances of juvenile salmon, and were used throughout the sampling period, from May through September. TS03, the channel closest to the bay, and most influenced by the tides, was inundated throughout the sampling period, but only exceeded depths greater than 0.5m on a regular basis when the river levels rose in mid July. This channel had the lowest overall abundances of fish, and fish were present only when the depth was at least 0.5m (Figure 20). The fact that this channel has a high amount of variability in temperature and depth on a daily basis may be an important habitat feature for fish adjusting to nearshore conditions.
Coho salmon of three age classes (age 0, age 1 and age 2) were found in all channels throughout the sampling period. Age 1 Sockeye salmon were present early in the season, but were absent later, indicating outmigration. Age 0 Sockeye salmon were present throughout the sampling period.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 36
Figure 20. Patterns of inundation and fish abundance in all four tidal channels sampled in 2011.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 37
Figure 21. Juvenile Coho salmon (top) and Sockeye salmon (bottom) size class histograms present in the four tidal channels from May-Sep 2010. Age classes were determined through otolith analysis.
Fish Growth and Habitat Use
In 2011, we PIT tagged 215 Coho salmon in TS00 and TS01. Although the PIT tag reading antenna did not operate as intended, we were able to recapture 70 tagged fish (some of them multiple times) during our regular bi-weekly sampling events. Many of the recaptures were present in the channel habitats for over 60 days, and through these recaptures, we were able obtain measures of individual growth for tagged fish. Fish that were present earlier in the estuary, as determined by date of PIT tag, appeared to grow more than those that entered later (Figure 22). We used bioenergetic models to explore seasonal growth patterns more fully.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 38
Figure 22. Residency of juvenile Coho as determined from recapturing PIT tagged fish (left); and growth of individual PIT tagged fish obtained by repeated measures of PIT tagged fish (right).
Vegetation Composition Salt marsh vegetation in the Fox River estuary was mapped at a coarse community scale on the west side of the river, and at a fine scale around the four tributaries on the east side of the river which were intensively sampled for juvenile salmon (Figure 23).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 39
Figure 23. Salt marsh plant community mapping of the Fox River Flats. The west side of the river was mapped at a coarse scale; and mapping around the four tidal channels on the east side was completed at a fine scale.
Sixteen vegetation communities were identified throughout the channels (Appendix A). Vegetation at TS00 is diverse and heterogeneously distributed among nine to ten qualitatively delineated assemblages (Figure 24a). Argentina ededii, Carex lynbyei and C. ramenskii tend to provide the greatest cover (accounting for >55% in aggregate) but eight other species contribute to the remainder (Figure 25a). TS01 vegetation corresponds more with two prominent assemblages aligned with the tidal channel, resulting in dominance of percent cover by either mud and Plantago maritima, C. lymbyei or by C. ramenskii, but the overall compositions more balanced among eight species (Figure 24b; Figure 25b). TS02 is similarly stratified by two dominant vegetation assemblages along the channel (Figure 24c), reflected in the prominence of C. lyngbyei in at least eight of the 13 plots, and >30% overall. Poa eminens (20%), mud (17%), and C. ramenskii (11%) were the other major contributers (Figure 25c). The lowest, most saline tidal channel system, TS03, was characterized by three prominent assemblages (Figs. 24d and 25d): (1) C. ramenskii between 50-95% coverage and balance mud; (2) 25-65% mud integrated with
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 40 Puccinellia nutkaensis, P. phryganodes, and Triglochin palustris; and, (3) plots composed almost exclusively of C. lyngbyei.
Based on the aggregate composition at each tidal channel, the evenness and diversity of vegetation and other cover (mud) classes decreases across the down-estuary gradient from TS00 to TS03 (Figure 26) wherein the lowest salinity end of the gradient has the most diverse and evenly represented vegetation while the marine end of the spectrum is represented by fewer and less evenly distributed vegetation, and almost 30% unvegetated. Some vegetation classes appear to reach maximum representation in the middle of the gradient, e.g., P. maritima and P. nutkaensis in TS01 and C. lyngbyei, P. eminens and Argentina egedii in TS02.
Mud, Carex ramenskii and C. lyngbei are the three main drivers of site distribution analyzed by NMDS (Figure 27). The closer the sites are to each other, the more similar they are. The lengths of the vector arrows reflect the strength of correlation. Correlations of about .50 or higher are considered significant. Sites further along the arrow tend to have more of that species. Arrows 180° from each other are negatively correlated relative to each other. Arrows 90° from each other have no correlation to each other. Cluster rings show sites that are 50% and 60% similar to each other.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 41
Figure 24. Mapped vegetation communities for the 4 sampled channels. IFT locations are shown as yellow points. Red points represent vegetation plots. See Appendix A for plant community legend.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 42
Figure 25. Percent cover of vegetation species in plots at each channel.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 43
Figure 26. Diversity and evenness of vegetation communities across all four channels. (See legend of Fig 25 for species names.)
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 44
Figure 27. NMDS plot of all vegetation plots; vectors indicate vegetation taxa that are responsible for arranging the plots in multidimensional space. Cluster rings show sites that are 50% and 60% similar to each other.
Prey Availability
Insect taxa captured in the IFT’s in 2011 illustrated consistently higher densities in July and general decline until September (Figure 29). Except for the abundance of Collembola (springtails: primarily Sminthuridae, ‘globular springtails’ and Entomobryiidae, ‘slender springtails’) in June (30-75% abundance), Diptera (true flies, mosquitoes, gnats, midges: primarily Chironomidae, ‘nonbiting midges’) typically dominated from July (60-80%) through September (21-55%) commensurate to an increase in the representation of Hymenoptera (sawflies, wasps, bees and ants: primarily Mymaridae, or ‘fairyflies’ or ‘fairy wasps’), from 8-13% in July to 24-46% in September. Hemiptera (true bugs: primarily Cicadellidae, or ‘leafhoppers’) also tend to increase progressively over time, from 5-8% in June to 23- 28% in September.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 45
Collembola (adult) Chironomid (adult) Chrinomid (juv) Mymaridae (adult) Cicadellidae (adult)
Figure 28. Insect fall-out trap (IFT) seasonal abundance for each tributary. Images shown are for most common insect groups found in IFT.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 46
Figure 29 Insect fall-out trap (IFT) seasonal percent composition.
Based on the sum of all IFT at each site across all four months (Figure 30) broad insect diversity and evenness generally increase along the gradient of the river reach, reaching maxima in TS02 and TS03, while species richness is highest in TS01. Multivariate analysis of the overall composition indicates that six major taxa drive the differences among the four sites: (1) Diptera and Hemiptera numerically prevalent at TS01 and Collembola at TS02; (2) Collembola (Sminthuridae) gradually decreasing from TS00 to TS03; and, (3) Acari (mites, ticks) and other Arachnida (e.g., spiders) and Hymenoptera remaining relatively stable across the estuarine gradient but become increasingly abundant as overall abundance decreases toward TS03 (Figure 31).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 47
Figure 30 General abundance of IFT insect taxa summed over all replicates and sampling dates.
Figure 31 NMDS plot of temporal associations of IFT insect compositions.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 48
Further examination of the IFT samples from the September 22-23, 2011 sampling allows relating the vegetation and fish diet compositions. Overall insect abundance was appreciably less in late September, compared to June-August, with the exception of Collembola at TS00, and to a lesser extent in TS02 (Figures 32 and 33).
Figure 32 September 22-23, 2011 IFT insect abundance of general taxa.
Figure 33 September 22-23, 2011 IFT insect composition. Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 49
Relative insect composition at each of the sites suggests that although Hemiptera and Arachnida don’t actually increase numerically across sites, they do become relatively more important as overall insect abundance decreases. At the insect family level, the shift from Sminthuridae to Entomobryiidae down estuary, between TS00 and TS02 is evident (Figs. 34 and 35). Sminthuridae virtually disappear down estuary between TS00 and TS01, replaced by more diverse Acari, Chironomidae, Entomobryiidae, and Ephydridae (‘shore flies’) at TS01. Springtails (but this time Entomobryidae) again dominate the IFT composition at TS02 but then diversify again at TS03 with predominantly Acari, Ephydridae, Mymaridae and Saldidae (‘shore bugs’).
Figure 34. September 22-23, 2011 IFT insect abundance of at specific family level.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 50
Figure 35 September 22-23, 2011 IFT insect composition at specific family level.
Both insects and vegetation appear to follow somewhat of a bi-modal pattern in occurrence. We examined cumulative hours of inundation for each tributary over the five month sampling period to see if apparent differences in inundation could be a contributing factor (Table 2 and Figure 36). TS01 is the largest of the tributaries, with the largest contributing drainage area, and experiences the highest amount of inundation close to the channel, although there is a general trend of higher inundation of the marsh surface follows the river gradient, with the lower levels in TS00 and highest amounts of inundation in TS03.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 51 Table 3. Hours of inundation for the entire drainage of each of the four tributaries. The “Area” field shows the area of each drainage, and the “Mean” is the mean hours of inundation for all cells in the drainage.
Site Area(m) Min Max Range Mean Std Sum TS00 567 0.36 15.3 14.9 1.04 1.2 588.06 TS01 77465 0.36 67.1 66.8 1.68 4.3 129997.00 TS02 43612 1.82 239.3 237.5 15.53 21.1 677161.00 TS03 42841 2.92 410.0 407.1 29.56 34.3 1266520.00
Fish diet composition
Juvenile Sockeye and Coho salmon occupying the tidal channels of the Fox River delta draw primarily on aquatic and terrestrial insects as the source of prey. Preliminary diet composition of juvenile Sockeye and Coho salmon sampled from TS01 in May-July 2010 indicated that these insects likely originated from both the channel and adjacent emergent marsh environments but also likely drift from more terrestrial sources (Appendix B). Chironomid flies in the process of pupating to adults were the most consistent prey in terms of both numerical composition and frequency of occurrence (and often gravimetric composition and %Total IRI) of both species. Other insects, such as Cicadellidae (leafhopper), Dolichopodidae (long- legged fly), Saldidae (shore bug), Corixidae (boat bug), Araneae (spider), Curculionidae (weevil, snout- beetle) and Elateridae (click beetle) often dominated the gravimetric composition but were less numerous or frequently represented in their diets overall. In a few cases (e.g., O. nerka in TS01 on 21 June) fish larvae contributed to high numerical and gravimetric composition but just occurred in one (of five) fish.
The availability of more sources of samples from all four tidal channels and four months enabled characterization of inter-specific, spatial and temporal variability in juvenile salmon diets in 2011. The diet spectrum (IRI diagram) of 112 juvenile Coho (27-107 mm FL; 0.17-12.68 g ww) indicated that chironomid larvae, pupa and adults, adult cicadellids, adult collumbolans, and harpacticoid copepods were numerically the most important prey, but the adult cicadellids, fish larvae and lepidopteran larvae contributed the most prey biomass (Figure 37). Approximately 83 different prey taxa were involved in the prey spectrum; however, less than 20 taxa occurred in more than 20% of the samples. On the basis of the frequency of occurrence, however, the flies, leafhoppers and springtail insects were the most common in the samples.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 52
Figure 36 Total summer (May-Sep) 2011 hours of inundation in each of the four tributaries by tide for each 1-m pixel. Drainage areas for each tidal channel are outlined in black. Areas with no data (white) were either always inundated, or never inundated. (The real high is much higher than 45 (>400).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 53
Figure 37 IRI diagram of Coho salmon diet composition for all fish captured in tidal channels of the Fox River delta in 2011
Among the prey constituting to ≥5% frequency of occurrence (Figure 38), insects constituted almost 90% of these most important prey taxa, complemented by only plant matter, and the harpacticoid and isopod crustaceans. The prey spectrum of the total 104 juvenile Sockeye salmon sampled in the delta in 2011 was somewhat more compressed, where only five prey taxa occurred in ≥20% of the samples, numerically dominated (>85%) by harpacticoid copepods but gravimetrically by adult cicadellids and chironomid larvae as well. Only 15 taxa occurred in ≥5% of the samples and all of these were insects except for harpacticoid and calanoid (Eurytemora sp.) copepods, and plant matter.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 54
Figure 38 IRI diagram of Coho salmon diet composition for all fish captured in tidal channels of the Fox River delta in 2011, where only prey contributing to ≥5% frequency of occurrence are included.
Multivariate (NMDS) analysis indicated that the numerical diet compositions of juvenile Coho and Sockeye salmon occupying the delta in 2011 were only marginally different (Figure 39). The primary distinguishing prey taxa were harpacticoids fed upon by Sockeye.
Based on diet prey spectra of taxa ≥5% frequency of occurrence, juvenile Coho and Sockeye salmon co- occurring in tidal channels TS00, TS01, TS02 and TS03 illustrated some differences in prey selection even though they were captured simultaneously (Figure 40). At the oligohaline, up-estuary channel TS00, juvenile Coho salmon on a diverse diet of 25 prey taxa, dominated numerically by carabid (ground) beetles, psyliids (jumping plant lice), saidids and coleopterans, and gravimetrically by isopods (but by only one fish) and the more abundant, common taxa. In contrast, juvenile Sockeye salmon fed commonly on chironomid larvae and other dipterans, harpacticoids, cicadellids and ostracods but the cicadellids leafhoppers, and simuliid and chironomid flies dominated the numerical composition and the cicadellids, ostracods and empidids (dance flies; in one fish) documented the prey biomass.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 55
Figure 39 NMDS diagram of juvenile Coho and Sockeye salmon diet compositions over all dates and sampling sites in 2011.
Figure 40 IRI diagrams of juvenile Coho and Sockeye salmon diet composition for all fish captured in tidal channel TS00 in the Fox River delta in 2011.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 56 Similar differences in prey composition occurred between juvenile Coho and Sockeye salmon in channel TS01 (Figure 41), where juvenile Coho salmon fed on a broader array (31) of taxa than Sockeye salmon (21). Coho fed predominantly on chironomid, dolichopodid and other dipteran flies, cicadellids, mites and wasps, but fish larvae dominated the overall prey biomass (by one fish). Conversely, Sockeye preyed on almost exclusively on harpacticoid copepods and chironomid larvae, pupae and adults.
Figure 41 IRI diagrams of Coho and Sockeye salmon diet composition for all fish captured in tidal channel TS01 in the Fox River delta in 2011.
The greater diversity between juvenile Coho salmon (24 taxa) and Sockeye salmon (17) held up somewhat in channel TS02 (Figure 42), even considering the high samples sizes (71-77; these summaries included the additional fish from the consumption rate experiments). While chironomids (including larvae) and aphids occurred prominently in the diets of both species, Sockeye salmon were distinguished by the extensive prominence (~68% frequency of occurrence; 87% numerical composition) of harpacticoid copepods in their diet, and a few Coho salmon that also fed on large harpacticoids.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 57
Figure 42 IRI diagrams of Coho and Sockeye salmon diet composition for all fish captured in tidal channel TS02 in the Fox River delta in 2011.
At the more marine end of the estuary gradient, the diet spectra of juvenile Coho and Sockeye salmon continue to be diverse but the diversity by species somewhat reversed (Figure 43). The representation of prey taxa were similar for both species, but juvenile Coho concentrated extensively (69% numerical and 61% gravimetric composition; 80% frequency of occurrence; 49% total IRI) on leafhoppers and to a lesser degree on dipteran fly larvae. Conversely, juvenile Sockeye fed on a diversity of aphids (21% numerical composition), cicadellids (49% of gravimetric composition; 15% numerical composition), chironomid larvae, pupae and adults, dance and other flies, and harpacticoid copepods (23% numerical composition).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 58
Figure 43 IRI diagrams of Coho and Sockeye salmon diet composition for all fish captured in tidal channel TS03 in the Fox River delta in 2011.
Multivariate (NMDS) analysis of the juvenile Coho salmon numerical diet composition grouped by tidal slough indicated that, although prey taxa were broadly represented in prey of fish in all the channels (especially TS02, where the samples contributed from the diel consumption sampling increased the sample size considerably), the dominance of several taxa distinguished the channels (Figure 44). Most notably, collembolans, chironomids and ceratopogonids tended to distinguish TS02 from the other three channels. The diet compositions of juvenile Sockeye salmon in the different channels shows somewhat more distinct differences among channels (however, the greater sample size for TS02 also dominates this analysis) and illustrates the dominant effect of harpacticoids as a discriminating factor in the diets (Figure 45). Ostracods appear to discriminate in TS00 and differences in the contribution of cicadellids and aphids highlight the diets of Sockeye salmon in TS03.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 59
Figure 44 Multivariate (NMDS) analysis of juvenile Coho salmon diet composition across the four tributaries. There does not appear to be clear groupings by site.
Figure 45 Multivariate (NMDS) analysis of juvenile Sockeye salmon diet composition across the four tributaries.
Multivariate analysis of the numerical composition of juvenile Coho salmon by sampling date illustrates some level of transition in dominant prey between May and August (Figure 46). The shift in prey composition appears to be due to changes in the contributions of collembolans early in the season, more contributions by cicadellids and ceratopogonids in mid-summer, and more cicadellids later in the summer. Seasonal changes in juvenile Sockeye salmon diet composition is somewhat swamped by variation in the contributions of harpacticoids over the five months but several other prey taxa account for
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 60 somewhat tighter groupings of the monthly diets (Figure 47). The dominance by harpacticoid copepods is most evident in May, but by June the contributions of chironomids is more influential, and aphids and cicadellids influence the composition more in late summer.
To investigate the possible association between the diet composition of juvenile Coho and Sockeye salmon and prey availability as represented by the IFT composition, we conducted Canonical Correspondence Analysis (CCA). We removed harpacticoids from the diet data because they would not show up in IFT. The CCA did not show the model to be significant for either Coho or Sockeye salmon diets. That is, the whole IFT dataset did not explain either diet dataset better than could be expected by chance. Only the abundance of coleopterans in the IFT explained diet distribution better than could be expected by chance (Figure 48).
Figure 46 Multivariate analysis of juvenile Coho salmon diet composition (2011) in the four tributaries. Coho diets, prey species are grouped into coarser taxa levels for better accuracy, though less resolution. Groupings indicate shifting of diets by date, driven by hemiptera and diptera
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 61
Figure 47 Multivariate analysis of juvenile Sockeye salmon diet composition (2011) in the four tributaries. Not all diet items could be identified down to the same level– thus some diptera species are identified simply as ‘diptera’, and some are identified down to finer taxa. Groupings indicate diet shifts by date, driven primarily by harpacticoids, cicadellids, chironomids, and aphids.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 62
Figure 48 Canonical Correspondence Analysis (CCA) of the association among IFT composition and diet composition of juvenile Coho salmon in the Fox River delta, May-August 2011. Dipteras were divided into finer taxa that appeared in >5% of the IFT; red labels are species that occur in diets; blue labels are species that occur in IFT; dots are Coho diets averaged by site and by date; diets that cluster around red species indicate that those diets are rich in that particular species; arrows and blue species that cluster near red species indicate IFT species that predominate when diets are rich in the red species; and, arrow length and distance from the center indicate degree of correlation with the diet data.
Diel Consumption Rate
The major objective of attempting diel consumption rate sampling at TS02 in each of the four different months was to incorporate emergent differences in diet composition (and thus prey energy density), consumption rate and temperature regime into a bioenergetic estimation of potential growth. Diet compositions of juvenile Coho and Sockeye salmon over the three diel sampling events in 2011 that could be used for consumption rates—May 20-21, July 20-21 and August 14-15—were appreciably different, most evident when examining the contributions of prey taxa >1% (Figure 49). Gravimetric composition used for estimating prey energy density of fish in May was concentrated in lepidopteran (moth) larvae, chironomid pupae and adults, and adult gammarid amphipods, while fish larvae, adult cicadellids and insect larvae dominated the diets of fish in July, and cicadellids completely dominated the diet in August (Figure 50).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 63
Figure 49 IRI diagram of juvenile Coho and Sockeye salmon during May 20-21, July 20-21 and August 14-15, 2001 diel consumption rate sampling at TS02. Estimated prey energy density (kj) is indicated by the red diamonds.
Figure 50 IRI diagram of juvenile Coho and Sockeye salmon diets for prey items greater than 1% during May 20-21, July 20-21 and August 14-15, 2001 diel consumption rate sampling at TS02. Estimated prey energy density (kj) is indicated by the red diamonds.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 64 The biomass of stomach contents and mean instantaneous ration of juvenile Coho salmon, for which gastric evacuation could be estimated, were comparable in May and July (maximum 0.15-0.14 g and 2.0- 2.3%, respectively) but somewhat higher (0.17 g, 5%)) in August (Figs. 51-54).
Figure 51 Diel feeding chronology of juvenile Coho salmon in tidal channel TS02 of the Fox River delta on May 20-21, 2011
Figure 52 Diel feeding chronology of juvenile Coho salmon in tidal channel TS02 of the Fox River delta on July 20-21, 2011
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 65
Figure 53 Diel feeding chronology of juvenile Coho salmon in tidal channel TS02 of the Fox River delta on August 14-15, 2011
Juvenile Sockeye salmon (that were consistently smaller than Coho through this sampling event) were also feeding intensely-->0.035 g mean stomach contents biomass and >4% (Figure 53).
Juvenile Sockeye, Oncorhynchus nerka 41-52 mmFL
Figure 54 Diel feeding chronology of juvenile Sockeye salmon in tidal channel TS02 of the Fox River delta on August 14-15, 2011
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 66 Where the mean stomach contents weight declined over a three-hour nocturnal time period, indicative of net digestion of prior meals, the corresponding gastric evacuation rates varied between 0.42 and 0.91 (mean = 0.52)) for juvenile Coho salmon and 0.35-0.37 for juvenile Sockeye salmon; gastric evacuation rates based on instantaneous ration ranged between 0.36 and 0.56 (mean = 0.43) for juvenile Coho salmon and 0.38 and 0.40 for juvenile Sockeye salmon (Table 3). Based on these results, the estimated daily rations (% body weight per day) for juvenile Coho varied between 14.2% in May, to 19.2% in July and 40.25% in August on the basis of stomach contents weight, or 8.7%, 11.6% and 18.2%, respectively, based on instantaneous ration (Table 3). These converted to estimates of a daily meal of 0.63 to 1.67 g d-1 for juvenile Coho salmon and 0.18 g d-1 for juvenile Sockeye salmon when based on stomach contents weight, or 0.39 to 0.93 g d-1 and 0.2 g d-1 for juvenile Coho and Sockeye salmon, respectively, when based on instantaneous ration.
Table 4 Estimated gastric evacuation (R), daily ration (D), and daily meal (F), based on mean stomach contents weight or instantaneous ration (IR) of juvenile Coho and Sockeye salmon during four diel consumption rate experiments in the Fox River delta, 2011.
-
%of %of
- - w*24
) )
R R
r) r)
w)
day day day
Avg Avg
Date
Daily Daily Daily
Eggers
g/day g/day g/day
Eggers
Doble Doble
Species
-
Fullness Fullness
body wt body wt body
Stomach Stomach
eaten per eaten per eaten (F=D*bw) (F=D*bw)
Daily Meal Meal Daily Meal Daily
over 24hrs over24hrs
(S/bw=g/gb Ration Ration
R (IR) Doble R(IR) Doble
(ContentWt
(D=S/bw*24 (D=S/b R (content) R (IR) R (content) R (IR) Coho 5/20-21/2011 0.9099 0.5559 0.0065 14.22% 8.69% 0.63 0.39 6/22-23/2011 0.4216 0.3863 7/20-21/2011 0.6598 0.4008 0.0121 19.15% 11.63% 0.89 0.54 8/14-15/2011 0.6431 0.3569 0.0261 40.25% 22.34% 1.67 0.93 Sockeye 7/20-21/2011 0.3798 0.4035 8/14-15/2011 0.3526 0.3843 0.0197 16.70% 18.20% 0.18 0.20
Bioenergetics Estimation of Potential Growth
Based on the above estimations of consumption rate and diet energy density, the potential growth estimated for juvenile Coho salmon ranged from 3.43% to 16.86% , and 18.42% for juvenile Sockeye salmon (Table 5). Based on absolute growth rate, the juvenile Coho salmon averaged between 0.016 to 0.046 g g-1 d-1 in May and July but increased to 0.17 to 0.19 g g-1 d-1 in August, and juvenile Sockeye salmon ranged between 0.176 and 0.191 g g-1 d-1 in August (Figure 54). To evaluate the change in modeled potential growth between May-July and August, we also modeled the difference in growth in May for juvenile Coho between 1 g and 10 g, which indicated proportionally little variation as a function of fish size (Figure 55). Although differences in our estimated daily ration and potential growth rates between May-July and August are somewhat dramatic, the combination of increased foraging rate and
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 67 significantly higher energy value of the prey in August appears to be a realistic representation of higher habitat value in late summer than during initial rearing by juvenile coho and sockeye.
Table 5 Input parameters and potential growth estimated from Wisconsin Fish Bioenergetics Model based on the result of three diel consumption rate sampling experiments in the Fox River delta, 2011; P:T, P:W, and P:D are the temperature, fish weight and daily ration for which P, the proportion of measured consumption of maximum potential consumption, is estimated; ED is total prey energy density based on diet composition and Growth T are the modeled temperatures; P estimated to be maximum for Coho salmon on 8/14-15.
1
-
th T th
P
(g)
ED
P:T
P:D
P:W
Mean
Range Range
mm d mm
Weight Weight
(mmFL) (mmFL)
Stomach Stomach Stomach
(gdamp) (gdamp)
(g)Range Grow
GROWTH GROWTH
Date/Time
Species and and Species
Contents Wt Wt Contents Wt Contents
MeanLength
Length Range Range Length Mean Weight Weight Mean Coho 5/20-21 53-97 75.43 0.546- 4.46 0-0.0391 0.0069 9.25 4.46 0.087 0.548 6118 5 0.016405 0745-0445 9.370 6 0.023229 7 0.030148 8 0.036409 9 0.041447 10 0.045027 11 0.047198 3.43%
7/20-21 33-99 67.12 0.260- 4.65 0.0003- 0.0121 10.24 4.65 0.116 0.686 4874 9 0.039966 1720-1630 12.433 0.0367 10 0.043471 11 0.045608 4.30%
8/14-15 38-92 66.75 0.546- 4.15 0.0029- 0.085 9.63 4.15 0.403 1* 11859 8 0.150013 1800-1700 9.370 0.4196 9 0.170106 10 0.185536 16.86% Sockeye 8/14-15 43-54 47.74 0.662- 1.09 0.0017- 0.019 9.63 1.09 0.182 0.726 10620 8 0.176045 1800-1700 1.775 0.0372 9 0.182242 10 0.187246 11 0.19113 18.42%
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 68
Figure 55 Growth potential of juvenile Coho and Sockeye salmon in channel TS02 of the Fox River delta in May-August 2011; close symbols are calculated for mean fish weight and diet composition under mean ambient temperature; lines are estimated for juvenile Coho salmon under the range of individual fish weights and temperatures occurring during May.
Discussion
Our work substantiates that there are large numbers of both Coho and Sockeye salmon juveniles using estuarine habitats of the Fox River for extended periods of time (more than six months of the year -April through September). We have shown, through repeated capture of PIT tagged individuals, that at least some of the juvenile Coho salmon in the Fox River system are remaining in the same estuarine habitats for over 60 days, and are feeding and growing in these habitats. The majority of the stomach contents obtained from both Coho and Sockeye salmon were undigested, indicating fish had recently fed in the marsh channel habitats, and providing further supporting evidence that juvenile fish are not just transitioning through, but are remaining in these estuarine habitats long enough to feed.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 69 The presence of large numbers of age 0 Coho and Sockeye salmon could be an indicator of a lack of suitable rearing habitat upstream, as is thought to be the case for Sockeye salmon in the Taku River, another glacially derived system in southeast Alaska (Eiler et al. 1988). A companion study to the work reported here (Hoem Neher et al 2013a; Hoem Neher et al 2013a), compared juvenile Coho salmon use between the glacially-derived Fox River and the snowmelt driven Anchor River, both of which are located in Kachemak Bay. The use of estuary habitats in these two systems was examined by using analysis of microchemistry and microstructure of sagittal otoliths of juvenile Coho salmon from these two systems. Coho Salmon captured in the glacial estuary had greater variability in body length and condition, and younger age-classes predominated the catch compared with the nearby snowmelt-fed, smaller estuary. Estuary-rearing fish in the glacial Fox River arrived later and remained longer during the summer than did fish using the snowmelt estuary (39 versus 24 d of summer growth). Finally, definitive patterns of overwintering in estuarine and near shore environments were observed in both estuaries (Hoem-Neher 2013a). Our sampling was limited by access due to ice flows, and the soonest we were able to begin sampling was late April, at which point there were considerable numbers of juvenile salmon present. Similarly, at the close of our sampling at the end of September (due to funding constraints), many juvenile salmon were still present, potentially fish that would overwinter.
The seasonally changing combination of river flow and tides creates dynamic habitats, with changing water depths, salinities and temperatures. River water is cold, from the glacier melt, whereas tidal water is relatively warm. In each tidal channel, we observed varying amounts of stratification, with cold, glacially derived water near the surface and warmer, saline water near the bottom. The tidal channel at the upper ecotone of the estuary, where freshwater influence is greatest (TS00) was inundated for three months (June through August) whereas the two tidal channels in the mid ecotone of the estuary (TS01 and TS02) were inundated for six months (April through September). TS03, which is farthest downstream, closest to the river mouth, and primarily influenced by tides, was intermittently available, with the greatest variability in depth and temperature; quite frequently water was less than 6 cm in depth. Water depth is often associated with juvenile fish use of estuarine channels (Miller and Simenstad 1997; Webster et al. 2001; Hering et al. 2010). In our study, the two most intermittently available habitats showed the greatest variation in fish size and relative abundance whereas fish using channels that remained consistently flooded showed less variation.
While others have used a bioenergetic approach to understand comparable growth rates in different habitats (e.g. Beauchamps et al. 1989; Levings and Boullion 2008), our study is one of the first to show seasonal variability in growth rates within a single habitat. We showed that growth rates in the Fox River
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 70 estuary for both Coho and Sockeye salmon changed seasonally, peaking in August with rates over, >0.18 mm d-1. Given that mean temperatures only ranged between 9.25°F and 10.24°F, the considerable increase in growth in August was apparently the cumulative effect of increased consumption rate (8.7% and 11.6% in May and July, and 22.3% in August, based on instantaneous ration rates) and prey energy density (6118 j and 4874 j in May and July, respectively, vs. 11859j in August).
The lack of amphipods, mysids and other benthic/epibenthic crustaceans in the diets juvenile Sockeye and Coho salmon is unusual (Miller and Simenstad 1997). The impressive taxa diversity of the insects from the Fox River IFT’s (approximately 83 different prey taxa were involved in the prey spectrum) may reflect the higher elevation marsh habitat with less frequent inundation. This may be a result of the dense clay-like glacial sediments that compose the estuary tributary channels. Sedges (Carex sp.) lining and overhanging the channels appear to be the primary habitat for potential prey, providing the basis for a detritus and sediment microalgae driven trophic structure (Siebert et al. 1977; Simenstad et al.1990).
Although Alaskan salmon stocks have been healthy, the effects of human uses on land and in the ocean, combined with changing climate create uncertainty for the future (Schindler et al 2008; Bottom et al 2009; Augerot and Smith 2010). There is no historical count data for the Fox River salmon runs, however returning adult counts for the Anchor River, the closest river system for which there is data, show returning Coho salmon numbers are far below the seven year averages (ADFG 2011). In the companion study comparing estuary habitat use by juvenile Coho salmon in the Anchor River and Fox Rivers, large numbers of fish were found in both systems, although the timing of estuary occupation was different, presumably driven by differences in water levels and temperatures between the snowmelt and rain supported Anchor River and the glacial melt water driven hydrology of the Fox River (Hoem Neher et al 2013a). Clearly, understanding the role that estuarine habitats play in the rearing of juvenile salmon in diverse estuarine settings such as these will be important for future management and conservation decisions. Our work provides the local community, managers, and regulators with detailed baseline information on juvenile salmon use of Fox River estuary. Anthropogenic modifications to this (and other) estuary ecosystem will likely be amplified by regional shifts in temperature and precipitation levels that also alter freshwater discharge regimes, and changes in tidal influence associated with glacial rebound and sea-level rise (Hinzman et al. 2005). Providing salmon populations with as much resilience potential as possible will require protecting these estuarine habitats from impacts to the extent possible.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 71 Future Studies
Continued research will build understanding of juvenile salmon use of estuarine rearing habitats, contributing to local and regional efforts to sustain salmon populations into the future. We suggest that future efforts focus on one or more of the following directions:
1. Comparisons with other estuary structures that vary in tidal mixing, thermal regimes, allochthonous inputs, and salmon populations, e.g., gradient between the Fox River and Anchor River. 2. Detailed residency studies for juvenile Coho and Sockeye salmon in the most stable estuary habitats. 3. Investigate a broader range of middle ecotone channels throughout the Fox River Flats area. 4. Expand temporal investigations to determine the full range of times that juvenile salmon are present in the estuary. 5. Conduct investigations of juvenile salmon residency and feeding in the nearshore environment at the mouth of the Fox River, including their distribution and ecology as they move from the Fox River delta through Kachemak Bay and out into nearshore Cook Inlet. 6. Investigate rearing habitats in the tidal freshwater environments upstream of the estuary.
Acknowledgments The Fox River delta is a dynamic place- filled with changing winds, water, mud, fish, insects, mammals and birds. We are grateful for the opportunity to get to know it so well, and to the many people who assisted with this project. Mark Marrette, resident cowboy of the Flats, provided local knowledge on access, bear protection, and rescued our river boat and IFT’s on several occasions. Mossy Kilcher graciously allowed us to store gear on her property adjacent to the Flats. Michelle Gutsch interned with us in 2009, worked hard, played hard, and got about as muddy as a person could get. Chuck Owens interned in 2010- coming from urban Philadelphia to Alaska– working hard, and catching his first fish ever. Jason Neher was our field technician in 2010. Trevon Cornwell, from Oregon Department of Fish and Wildlife, showed us how to use fyke nets, and happily endured a week long adventure that included a major wild fire, loss of electricity and guitar playing. The University of Washington, Wetland Ecosystem Team’s, Lia Stamatiou, patiently guided our attempts to sample prey, and even more patiently identified them. Geoff Coble tried his best to ward off cows and sediment to keep the antennas operational. Amanda Rosenberger was a generous colleague-without her, we would not have had a river boat. Adam Craig, ADFG Sportfish Division biometrician provided valuable advice and comments on yearly planning
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 72 efforts. Ted Otis, fisheries biologist with ADFG Commercial Fisheries, assisted with juvenile salmon identification. Ginny Litchfield, ADFG Habitat Division, furnished permits. Staff of the Kachemak Bay Research Reserve provided the backbone of administrative support that is necessary for any successful project-our sincere thanks to Kim Cooney and Amy Alderfer for their efforts. Thanks to KBRR staff Angie Doroff, Ori Badajos, Joel Markis and Carmen Field for their help in the field. We enjoyed the help, good company and hard work of numerous volunteers including Janet Fink, Kristin Berger, Becky Shaftel, Emily Kizzia, Liza Walker, Sammy Walker, Daisy Walker and Dennis Whigham. Finally, we’d like to acknowledge Russ Walkers generosity in lending a hand, tools, equipment, good humor and timeless sense of style.
Literature Cited
Alaska Department of Fish and Game (ADFG) 1977. A fish and wildlife resource inventory of the Cook Inlet-Kodiak areas. Vol. 2, Juneau, AK. ADFG. 1982. 1982 Kachemak Bay subsistence gillnet fishery. Division of Commercial Fisheries, Lower Cook Inlet Data Report Number 83-1. Homer, AK. ADFG. 1983. Lower Cook Inlet Data Report Number 83-1. Division of Commercial Fisheries, Homer, AK ADFG. 1985. Aerial counts of southern district-lower Cook Inlet area- Clearwater Slough and Fox River tributaries. Unpublished report. Division of Commercial Fisheries, Homer, AK ADFG. 1993. Fox River Flats Critical Habitat Area Plan. Division of Habitat and Restoration and Wildlife Conservation. Anchorage, AK. ADFG. 2011. Anchor River Coho salmon run. Alaska Outdoor Journal salmon run charts. Available from http://www.alaskaoutdoorjournal.com/Sonar/anchorcoho.html Augerot,X, Smith CL. 2010. Comparative resilience in five North Pacific regional salmon fisheries. Ecology and Society 15(2): 3. [online] URL: http://www.ecologyandsociety.org/vol15/iss2/art3/ Bailey RE, Irvine JR, Dalziel FC, Nelson TC. 1998. Evaluations of visible implant fluorescent tags for marking coho salmon smolts. North American Journal of Fisheries Management 18:191-196. Batten AR, Murphy S, Murray DF. 1978. Definition of coastal wetlands by floristic criteria. USEPA no. 804965-01. US Environmental Protection Agency, Corvalis Environmental Research Laboratory, Corvalis, OR. 490 p. Beauchamp DA, Stewart DJ, Thomas GL. 1989. Corroboration of a bioenergetics model for Sockeye salmon. Transaction of the American Fisheries Society. 118: 597-607. Beamer EM, Larsen K. 2004. The importance of Skagit delta habitat on the growth ofwild ocean-type Chinook in Skagit Bay: implications for delta restoration. Skagit River System Cooperative research report. Available at www.skagitcoop.org. Beamish RJ, Mahnken C, Neville CM. 2004. Evidence that reduced early marine growth is associated with lowermarine survival of Coho Salmon. Transactions of the American Fisheries Society 133:26–33. Bieber, A. 2005. Vairability in Juvenile Chinook Foraging and Growth Potential in Oregon Estuaries: Implications for Habitat Restoration. MS Thesis. University of Washington School of Aquatic and Fishery Sciences, Seattle. Bottom, D.L., K.K. Jones, T.J. Cornwell, A. Gray and S.A. Simenstad. 2005a. Patterns of Chinook salmon migration and residency in the Salmon River estuary (Oregon). Estuarine, Coastal and Shelf Science 64; 79-83.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 73 Bostick WE. 1955. Seattles aquatic environments: Duwamish estuary. Bottom DL, Simenstad CA, Burke J, Baptista AM, Jay DA, Jones KK, Casillas E, Schiewe MH. 2005. Salmon at river's end: The role of the estuary in the decline and recovery of Columbia River salmon. U.S. Dept. of Commerce, NOAA Tech. Memo., NMFS-NWFSC-68, 246 p. Bottom DL, Jones KK, Cornwell TJ, Gray A, Simenstad CA. 2005b. Patterns of Chinook salmon migration and residency in the Salmon River Estuary (Oregon). Estuarine Coastal and Shelf Science 64:79-93. Bottom DL, Jones KK, Simenstad CA, Smith CL. 2009. Reconnecting social and ecological resilience in salmon ecosystems. Ecology and Society 14(1): 5. [online] URL: http://www.ecologyandsociety.org/vol14/iss1/art5/ Campbell LA. 2010. Life histories of juvenile Chinook Salmon (Oncorhynchus tshawytscha) in the Columbia River estuary as inferred from scale and otolith microchemistry. Master’s thesis. Oregon State University, Corvallis. Campana SE, Neilson JD. 1985. Microstructure of fish otoliths. Canadian Journal of Fisheries and Aquatic Sciences 42:1014–1032. Chittenden CM, Sura S, Butterworth KG, Cubitt KF, Plantalech Manel-a N, Balfry S, Oakland F, McKinley RS.. 2008. Riverine, estuarine and marine migratory behaviour and physiology of wild and hatchery-reared coho salmon Oncorhynchus kisutch (Walbaum) smolts descending the Campbell River, BC, Canada. Journal of Fish Biology 72:614-628. Cordell JR, Tear L.M, Simenstad CA,,Wenger SM, Hood WG. 1994. Duwamish River Coastal America restoration and reference sites: results and Recommendations from year-one pilot and monitoring studies. FRI-UW-9416, Fisheries Research Institute, University of Washington, Seattle. Doble BD, Eggers DM. 1978. Diel feeding chronology, rate of gastric evacuation, daily ration and prey selectivity in Lake Washington juvenile sockeye salmon (Oncorhynchus nerka). Transactions of the American Fisheries Society 107: 36-44. Eggers DM. 1977. The nature of prey selection by planktivorous fish. Ecology 58: 46-59. Eggers DM. 1979. Comments on some recent methods for estimating food consumption by fish. J Fish Res Bd Can 36: 1018-1019 Eiler JH, Nelson BD, Bradshaw RP, Greiner JR, Lorenz JM. 1988. Distribution, stock composition, and location and habitat type of spawning areas used by sockeye salmon on the Taku River. Northwest and Alaska Fisheries Center Report 88-24. National Marine Fisheries Service. US Dept. Commerce. Faurot MW, Palmer DE. 1992. Survey of the fishery resources in the Fox River watershed, Alaska, 1985- 1986. U.S. Fish and Wildlife Service, Alaska Fisheries and Technical Report Number 18, Kenai, Alaska. Fisher JP, Pearcy WG. 1990. Distribution and residence times of juvenile fall and spring Chinook salmon in Coos Bay, Oregon. Fisheries Bulletin 88:55-58. Fresh KL, Small, DJ, Kim H, Waldbillig C, Mizell M, Carr, MI, Stamatiou L. 2006 Juvenile salmon use of Sinclair Inlet, Washington in 2001 and 2002. Washington Department of Fish and Wildlife Technical Report No. FPT 05-08 Olympia, WA. Gaines PC , Martin C D. 2004. Feasibility of dual-darking age-0 Chinook salmon for mark–recapture Studies. North American Journal of Fisheries Management 24:1456-1459. Gray A., Simenstad SA, Bottom DL, Cornwell TJ. 2002. Contrasting functional performance of juvenile salmon habitat in recovering wetlands of the Salmon River estuary, Oregon, U.S.A. Restoration Ecology 10 (3); 514-526. Groot C, Margolis L. 1991. Pacific salmon life histories. University of British Columbia Press, Vancouver. Hanson PC, Johnson TB, Schindler DE, Kitchell JF. 1997. Fish Bioenergetics 3.0. Technical Report WISCU-T-97-001. University of Wisconsin Sea Grant Institute, Madison, WI.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 74 Hayes DB, Bence JR, Kwak TJ, Thompson BE. 2007. Abundance, biomass, and production.in C. S. Guy and M. L. Brown, editors. Analysis and interpretation of freshwater fisheries data. American Fisheries Society, Bethesda, Maryland. Healey M C. 1982. Juvenile Pacific salmon in estuaries: the life support system. Pages 315–341 in V. S. Kennedy, editor. Estuarine comparisons. Academic Press, New York. Heifetz JS, Johnson SW, Koski KV, Murphy ML. 1989. Migration timing, size, and salinity tolerance of sea-type Sockeye salmon (Oncorhynchus nerka) in an Alaskan estuary. Canadian Journal of Fisheries and Aquatic Sciences. 46:633-637. Hinzman LD, Bettez ND. Bolton WR, Chapin FS, DyurgerovMB, Fastie CI, Griffith B, Hollister RD, Hope A, Huntington HP, Jensen AM,. Jia GJ, Jorgenson T, Kane DI, Klein DR, Kofinas G, Lynch AH, Lloyd AH, Mcguire AD, Nelson FE, Oechel WC, Osterkamp TE, Racine CH, Romanovsky VE, Stone RS , Stow DA, Sturm M, Tweedie CE, Vourlitis GI, Walker MD, Walker DA, Webber PJ, Welker JM, Winker KS, Yoshikawa K. 2005.Evidence and implications of recent climate change in northern Alaska and other arctic regions. Climate change 72: 251- 298. DOI: 10.1007/~10584-005-5352-2. Hering DK, Bottom DL, Prentice EF, Jones KK, Fleming IA. 2010. Tidal movements and residency of subyearling Chinook Salmon (Oncorhynchus tshawytscha) in an Oregon salt marsh channel. Canadian Journal of Fisheries and Aquatic Sciences 67:524–533. Hoem Neher TD, Rosenberger AE, Zimmerman CE, Walker CM, Baird SJ. 2013a. Estuarine Environments as Rearing Habitats for Juvenile Coho Salmon in Contrasting South-Central Alaska Watersheds, Transactions of the American Fisheries Society 142 (6) 1481-1494. Hoem Neher TD, Rosenberger AE, Zimmerman CE, Walker CM, Baird SJ. 2013b. Use of glacial river- fed channels by juvenile Coho salmon: transitional or rearing habitats? Environmental Biology of Fish. DOI 10.1007/s10641-013-0183-x. Homer Soil and Water District. 2010. Fox River Flats Grazing Management Plan. Community Resource Management Plan, Homer AK. Jonas JL, Kraft CE, Margenau TL. 1996. Assessment of seasonal changes in energy density and condition in age-0 and age-1 Muskellunge. Transactions of the American Fisheries Society 125:203–210. Kachemak Bay Research Reserve (KBRR) and National Oceanic and Atmospheric Administration (NOAA), Coastal Services Center (CSC). 2001. Kachemak Bay Ecological Characterization. CD- ROM. NOAA/CSC/20017-CD. Charleston, S.C.: NOAA Coastal Services Center. Koski KV. 2009. The fate of Coho Salmon nomads: the story of an estuarine rearing strategy promoting resilience. Ecology and Society [online serial] 14(1):article 4. Levings CD, McAllister CD, Cheng BD. 1986. Differential use of the Campbell River estuary, British Columbia, by wild and hatchery-reared juvenile Chinook salmon (Oncorhynchus tshawytscha). Canadian Journal of Fisheries and Aquatic Science 43: 1386-1397. Levings C, Boullion D. 2008. Scaling salmonid life-history by habitat area: a conceptual approach to estimating estuarine conservation needs. American Fisheries Symposium 49: 1597-1604. Levy DA, Northcote TG. 1982. Juvenile salmon residency in a marsh area of the Fraser River estuary. Can. J. Fish. Aquat. Sci. 39(2); 270-276. Lott MA. 2004. Habitat-Specific Feeding Ecology of Ocean-Type Juvenile Chinook Salmon in the Lower Columbia River Estuary. U.W. M.S. Thesis. McMahonTE, Holtby LB. 1992. Behaviour, habitat use, and movements of Coho salmon (Oncorhynchus kisutch) smolts during seaward migration. Canadian Journal of Fisheries and Aquatic Sciences 49:1478-1485. Miller BA, Sadro S. 2003. Residence time and seasonal movements of juvenile Coho Salmon in the ecotone and lower estuary of Winchester Creek, South Slough, Oregon. Transactions of the American Fisheries Society 132:546–559. Miller JA., Simenstad CA. 1997. A comparative assessment of a natural and created estuarine slough as rearing habitat for juvenile Chinook and Coho salmon. Estuaries 20:792–806
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 75 Mueller-Domboise D, Ellenberg H. 1974. The count-plot method and plotless sampling techniques in Aims and Methods in Vegetation Ecology, Mueller-Domboise D, Ellenberg H. (eds) John Wiley and Sons. New York. Pearcy WG, Wilson D, Chung AW, Chapman JW. 1989. Residence times, distribution, and production of juvenile chum salmon, Oncorhynchus keta, in Netarts Bay, Oregon. Fishery Bulletin. Pope KL, Kruse CG. 2007. Condition. Pages 423–471 in C. S. Guy and M. L. Brown, editors. Analysis and interpretation of freshwater fisheries data. American Fisheries Society, Bethesda, Maryland. Powers SP, Bishop MA, Moffitt S, Reeves GH. 2007. Variability in freshwater, estuarine, and marine residence of Sockeye salmon Oncorhynchus nerka within the Copper and Bering River deltas, Alaska. American Fisheries Society Symposium 54: 97-89. Quinn TP. 2005. The behavior and ecology of Pacific salmon and trout. 1st edition. University of Washington Press, Seattle, WA. Rosenberger AE, Dunham JB. 2005. Validation of abundance estimates from mark-recapture and removal techniques for rainbow trout captured by electrofishing in small streams. North American Journal of FisheriesManagement 25:1395–1410. Ryall R, LevingsCD. 1987. Juvenile salmon utilization of rejuvenated tidal channels in the Squamish estuary, British Columbia. Canadian Manuscript Report of Fisheries and Aquatic Science 1904. Sather NK , Dawley EM, Johnson GE , Zimmerman SA, Storch AJ, Borde AB, Teel DJ, Mallette C, Skalski JR , Farr R, Jones TA. 2009. Ecology of Juvenile Salmon in Shallow Tidal Freshwater Habitats in the Vicinity of the Sandy River Delta, Lower Columbia River, 2008. Prepared for Bonneville Power Administrationunder an agreement with the U.S. Department of Energy Contract DE-AC05-76RL01830 Schindler DE, Augerot X, Fleishman E, Mantua NJ, Riddell B, Ruckelhaus M, Seeb J, Webster M. 2008. Climate change, ecosystem impacts, and Management for Pacific Salmon. Fisheries 33:10, 502- 506. Shreffler DK, Simenstad CA, Thom RM. 1992. Foraging by juvenile salmon in a restored wetland. Estuaries 15:204–213. Simenstad CA, Fresh KL, Salo EO. 1982. The role of Puget Sound estuaries in the life history of Pacific salmon: An unappreciated function. Pages 343-364 in V.S. Kennedy, ed. Estuarine comparisons. Academic Press, New York. Sowa SP, Annis G, Morey ME, Diamond DD. 2007. A gap analysis and comprehensive conservation strategy for riverine ecosystems of Missouri. Ecological Monographs 77:301–334. Stewart DJ, Ibarra M. 1991. Predation and production by salmonine fishes in Lake Michigan, 1978- 1988. Canadian Journal of Fisheries and Aquatic Sciences 48: 909-922. Sutton SG, Bult TP, Haedrich RL. 2000. Relationships among fat weight, body weight, water weight, and condition factors in wild Atlantic Salmon parr. Transactions of the American Fisheries Society 129:527–538. Tande GF, Lipkin R, Duffy M. 1996. A floristic inventory of Fort Wainwright military installation, Alaska. Alaska Natural Heritage Program, University of Alaska, Anchorage. Thornton KW, Lessem AS. 1987. A temperature algorithm for modifying biological rates. Transactions of the American Fishery Society 107 (2):284-287. Tschaplinski PJ. 1982. Aspects of the population biology of estuarine-reared and stream-reared juvenile coho salmon in Carnation Creek: a summary of current research, p.289-307. In GF Hartman (ed), Proceedings of the Carnation Creek Workshop: a ten-year review, Malaspina College, Nanaimo, British Columbia. Thorpe JE. 1994. Salmonid fishes and the estuarine environment. Estuaries 17:76–93. Volk EC, Bottom DL, Jones KK, Simenstad CA. 2010. Reconstructing juvenile Chinook Salmon life history in the Salmon River estuary, Oregon, using otolith microchemistry andmicrostructure. Transactions of the American Fisheries Society 139:535–549.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 76 Webster SJ, Dill LM, Korstrom JS. 2007. The effects of depth and salinity on juvenile Chinook Salmon Oncorhynchus tshawytscha (Walbaum) habitat choice in an artificial estuary. Journal of Fish Biology 71:842–851. Weitkamp DE and Schadt TH. 1982. 1980 juvenile salmonid study. Report to Port of Seattle. Document No. 82-0415-012F. Parametrix Inc. Seattle, WA.
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 77
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 78 Appendix A Vegetation Communities Mapped Adjacent to tidal channels of the Fox River
Carex lyngbyei Community (code: Cl) Dense communities of Carex lyngbyei occur in the tidal gut channels, sometimes extending all the way to the bottom of the channel, more frequently occurring along the sides of the channels, and up to the top edge. The plant composition often appears to be pure Carex lyngbyei, but generally includes other species, especially Carex ramenskii and Triglochin maritima. In small areas, Carex ramenskii may take over as the dominant species. At the top of the channels, this community grades into levee communities, and may include other species such as Argentina egedii and Hordeum brachyantherum. Due to its location, this community was frequently inundated by tides, and partially inundated as river levels increased with glacial melt.
This community occurs in the channels of the three lower sites (TS01, TS02, and TS03), where it is partially protected from cattle grazing by growing on the steep, slippery channel sides. Insect fallout traps were generally placed in this community or the community immediately adjacent to it at these three sites.
Average % cover values (number of plots=10), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 60 Argentina egedii t Carex ramenskii 20 Puccinellia nutkaensis t Hordeum brachyantherum 5 Triglochin palustris t Triglochin maritima t Comarum palustre t Bare mud 10
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 79 Carex ramenskii Community (code: Cr) Nearly pure stands of Carex ramenskii occur on flats, generally away from the channel edge. The only other plants recorded (in a few plots) were small percentages of Carex lyngbyei, Triglochin maritima, and Puccinellia phryganodes. This community occurred around the two lowest sites (TS02 and TS03), and was likely inundated at least monthly by tides. It was grazed heavily by cattle, especially at the lowest site, which may have resulted in lower percent cover values than if it had been ungrazed.
Average % cover values (number of plots=10), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 80 Carex lyngbyei t Triglochin maritima t Puccinellia phryganodes t Bare mud 15
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 80
Plantago maritima - Puccinellia nutkaensis Community (code: Pm-Pn) A nearly equal mix of Plantago maritima and Puccinellia nutkaensis occurs along raised levees on the edges of major channels. It contains smaller percentages of a variety of species, as well as a fair amount of unvegetated mud. Interestingly, this community was found at the lowest site (TS03) and the third lowest site (TS01), but not at the site between these two.
Average % cover values (number of plots=7), rounded to the nearest 5% (less than 2.5% = trace[t]). Plantago maritima 25 Argentina egedii 5 Puccinellia nutkaensis 20 Puccinellia phryganodes 5 Triglochin maritima 5 Leymus mollis t Carex ramenskii 5 Poa eminens t Bare mud 35
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 81
Puccinellia - Triglochin maritima Community (Code: P-Tm) This community is somewhat sparsely vegetated, occurring on flats surrounding the channel. In seven vegetation plots analyzed in this community, only three plant species were found, and all three were present in six of those seven plots. None of the species grow very robustly, and plots contained fairly high percentages of bare mud. This may be partially due to trampling and grazing. This was the dominant community immediately adjacent to the channel at the lowest site (TS03), and was not found at the other sites.
Average % cover values (number of plots=7), rounded to the nearest 5% (less than 2.5% = trace[t]). Puccinellia nutkaensis 25 Puccinellia phryganodes 15 Triglochin maritima 25 Bare mud 40
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 82
Carex ramenskii - Poa Eminens - Argentina egedii Community (Code: Cr-Pe-Ae) This community covers flats slightly removed from the channel. Either Carex ramenskii or Poa eminens can be dominant over small areas. Plantago maritima is generally also present. Cattle grazing pressure appears to limit the size of the grass and sedge. This community covers an extensive flat on the north side of the 2nd-highest site (TS01).
Average % cover values (number of plots=3), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 40 Plantago maritima 10 Poa eminens 20 Ranunculus cymbalaria t Argentina egedii 15 Bare mud 20
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 83
Poa eminens - Argentina egedii - Carex Community (Code: Pe-Ae-C) This community consisted of a nearly equal mix of Poa eminens and sedges (Carex lyngbyei and Carex ramenskii), with Argentina egedii. Several other species also occurred. This was the dominant community adjacent to the channel at the 2nd-lowest site(TS02).
Average % cover values (number of plots=5), rounded to the nearest 5% (less than 2.5% = trace[t]). Poa eminens 30 Plantago maritima 5 Carex lyngbyei 20 Puccinellia nutkaensis 5 Argentina egedii 15 Triglochin maritima t Carex ramenskii 15 Ranunculus cymbalaria t Bare mud 20
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 84
Poa eminens - Plantago maritima - Argentina egedii Community (Code: Pe-Pm-Ae) Similar to the previous community, but with no sedge and a higher percentage of Plantago maritima. This community occurs on raised levees along the channel. Found only at the 2nd-lowest site (TS02).
Average % cover values (number of plots=2), rounded to the nearest 5% (less than 2.5% = trace[t]). Poa eminens 40 Aregentina egedii 10 Plantago maritima 15 Bare mud 35
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 85
Carex lyngbyei - Argentina egedii Community (Code: Cl-Ae) The Poa eminens - Argentina egedii - Carex Community at the 2nd lowest site transitions into this community in slightly lower areas.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 70 Poa eminens t Aregentina egedii 5 Triglochin maritima t Bare mud 25
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 86
Plantago maritima - sedge - Poa eminens Community (Code: Pm-C-Pe) A single example of this community was found on the levee around a small side-gut feeding into the channel at the 2nd-lowest site (TS02).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Plantago maritima 20 Argentina egedii 5 Carex lyngbyei 10 Triglochin maritima t Carex ramenskii 10 Puccinellia nutkaensis t Poa eminens 10 Ranunculus cymbalaria t Bare mud 45
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 87
Leymus mollis - Poa eminens - Argentina egedii Community (Code: Lm-Pe-Ae) This Leymus mollis community was found on a few raised levees at the 2nd-lowest site (TS02).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Leymus mollis 20 Carex lyngbyei 15 Poa eminens 20 Plantago maritima 10 Argentina egedii 20 Bare mud 15
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 88
Carex ramenskii - Plantago maritima Community (Code: Cr-Pm) This community was found on a levee at the 2nd-highest site (TS01). The presence of Agrostis scabra indicates that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 30 Poa eminens 5 Plantago maritima 20 Argentina egedii 5 Hordeum brachyantherum 15 Agrostis scabra t Bare mud 25
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 89
Argentina egedii - Poa eminens Community (Code: Ae-Pe) This community was found on a levee at the 2nd-highest site (TS01). The presence of Agrostis scabra indicates that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Argentina egedii 40 Carex ramenskii 10 Poa eminens 20 Agrostis scabra 10 Plantago maritima 10 Puccinellia nutkaensis t Bare mud 10
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 90
Diverse Levee Community (Code: dl) This community was found on a levee along a small gut feeding into the channel at the 2nd-highest site (TS01). The presence of Agrostis scabra and Poa annua indicate that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Agrostis scabra 40 Carex lyngbyei 5 Poa eminens 20 Ranunculus cymbalaria t Hordeum brachyantherum 15 Poa annua t Argentina egedii 10 Lomatogonium rotatum t Bare mud 10
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 91
Carex lyngbyei - Eleocharis kamtschatica Community (Code: Cl-E) This community was found in a small gut feeding into the channel at the 2nd-highest site (TS01).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 50 Eleocharis kamtschatica 30 Bare mud 20
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 92
Hordeum brachyantherum - Agrostis scabra Community (Code: Hb-As) This community was found on a levee at the 2nd-highest site (TS01). The presence of Agrostis scabra indicates that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Hordeum brachyantherum 30 Puccinellia nutkaensis 10 Agrostis scabra 20 Ranunculus cymbalaria 5 Bare mud 35
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 93
Sedge - Argentina egedii - Ranunculus cymbalaria Community (Code: C-Ae-Rc) This community is primarily a mix of sedges and Argentina, with a good amount of Ranunculus cymbalaria. Either of the two common sedges can be the dominant species. In higher areas, a mix of less salt-tolerant grass species is found as well. This community was only found at the highest site (TS00). Higher areas within this community are almost never inundated by salt water, as is shown by the presence of such species as Poa annua and Trifolium repens.
Average % cover values (number of plots=3), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 25 Plantago maritima t Argentina egedii 20 Agrostis scabra t Carex lyngbyei 15 Hordeum brachyantherum t Ranunculus cymbalaria 15 Poa anua t Poa eminens t Triglochin palustris t Trifolium repens t Bare mud 30
Hordeum brachyantherum - Argentina egedii Community (Code: Hb-Ae) This community is rarely inundated by salt water. It was found on the levee on the north side of the highest channel (TS00).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Hordeum brachyantherum 60 Trifolium repens t Argentina egedii 15 Polygonum aviculare t Poa annua t Plantago maritima t Agrostis scabra t Lomatogonium rotatum t Poa eminens t Achillea millefolium t Bare mud 20
Puccinellia nutkaensis - Argentina egedii - Carex ramenskii Community (Code: Pn-Ae-Cr) This community includes a diverse mix of species, including some that are not highly salt-tolerant.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Puccinellia nutkaensis 40 Hordeum brachyantherum t Argentina egedii 30 Polygonum aviculare t Carex ramenskii 15 Plantago maritima t Carex lyngbyei t Lomatogonium rotatum t Bare mud 10
Puccinellia nutkaensis - Carex ramenskii Community (Code: Pn-Cr) This was the only community found at the highest site (TS00) that contained only species that are highly salt-tolerant.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 94 Puccinellia nutkaensis 50 Atriplex alaskensis t Carex ramenskii 10 Plantago maritima t Bare mud 40
Carex - grass Gut Community (Code: C-grass) This community was found in the channel at the highest site (TS00), and consisted primarily of sedges and grasses.
Average % cover values (number of plots=2), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 35 Trifolium repens t Carex ramenskii 30 Polygonum aviculare t Unidentified grass 10 Bare mud 20
Carex lyngbyei - Argentina egedii - Plantago maritima Community (Code: Cl-Ae-Pm) This community occurred on the north bank of the highest channel (TS00).
Average % cover values (number of plots=2), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 40 Hordeum brachyantherum t Argentina egedii 15 Trifolium repens t Plantago maritima 5 Lomatogonium rotatum t Agrostis scabra 5 Bare mud 35
Grass - Trifolium repens Community (Code: grass-Tr) A freshwater community occurring on a high levee to the south of the highest channel (TS00).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 95 Appendix B 2010 Diet Samples for Juvenile Coho and Sockeye Collected in tributaries of the Fox River
B1. IRI diagram of O. nerka diet composition at TS01 on 24 May 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 96
B. 2. IRI diagram of O. nerka diet composition at TS01 on 21 June 2010
B.3. IRI diagram of O. kisutch diet composition at TS01 on 3 May 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 97
B.4.IRI diagram of O. kisutch diet composition at TS01 on 24 May 2010
B.5. IRI diagram of O. kisutch diet composition at TS01 on 21 June 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 98
B.6. IRI diagram of O. kisutch diet composition at TS01 on 27 July 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 99
Appendix A Vegetation Communities Mapped Adjacent to tidal channels of the Fox River
Carex lyngbyei Community (code: Cl) Dense communities of Carex lyngbyei occur in the tidal gut channels, sometimes extending all the way to the bottom of the channel, more frequently occurring along the sides of the channels, and up to the top edge. The plant composition often appears to be pure Carex lyngbyei, but generally includes other species, especially Carex ramenskii and Triglochin maritima. In small areas, Carex ramenskii may take over as the dominant species. At the top of the channels, this community grades into levee communities, and may include other species such as Argentina egedii and Hordeum brachyantherum. Due to its location, this community was frequently inundated by tides, and partially inundated as river levels increased with glacial melt.
This community occurs in the channels of the three lower sites (TS01, TS02, and TS03), where it is partially protected from cattle grazing by growing on the steep, slippery channel sides. Insect fallout traps were generally placed in this community or the community immediately adjacent to it at these three sites.
Average % cover values (number of plots=10), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 60 Argentina egedii t Carex ramenskii 20 Puccinellia nutkaensis t Hordeum brachyantherum 5 Triglochin palustris t Triglochin maritima t Comarum palustre t Bare mud 10
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 100
Carex ramenskii Community (code: Cr) Nearly pure stands of Carex ramenskii occur on flats, generally away from the channel edge. The only other plants recorded (in a few plots) were small percentages of Carex lyngbyei, Triglochin maritima, and Puccinellia phryganodes. This community occurred around the two lowest sites (TS02 and TS03), and was likely inundated at least monthly by tides. It was grazed heavily by cattle, especially at the lowest site, which may have resulted in lower percent cover values than if it had been ungrazed.
Average % cover values (number of plots=10), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 80 Carex lyngbyei t Triglochin maritima t Puccinellia phryganodes t Bare mud 15
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 101
Plantago maritima - Puccinellia nutkaensis Community (code: Pm-Pn) A nearly equal mix of Plantago maritima and Puccinellia nutkaensis occurs along raised levees on the edges of major channels. It contains smaller percentages of a variety of species, as well as a fair amount of unvegetated mud. Interestingly, this community was found at the lowest site (TS03) and the third lowest site (TS01), but not at the site between these two.
Average % cover values (number of plots=7), rounded to the nearest 5% (less than 2.5% = trace[t]). Plantago maritima 25 Argentina egedii 5 Puccinellia nutkaensis 20 Puccinellia phryganodes 5 Triglochin maritima 5 Leymus mollis t Carex ramenskii 5 Poa eminens t Bare mud 35
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 102
Puccinellia - Triglochin maritima Community (Code: P-Tm) This community is somewhat sparsely vegetated, occurring on flats surrounding the channel. In seven vegetation plots analyzed in this community, only three plant species were found, and all three were present in six of those seven plots. None of the species grow very robustly, and plots contained fairly high percentages of bare mud. This may be partially due to trampling and grazing. This was the dominant community immediately adjacent to the channel at the lowest site (TS03), and was not found at the other sites.
Average % cover values (number of plots=7), rounded to the nearest 5% (less than 2.5% = trace[t]). Puccinellia nutkaensis 25 Puccinellia phryganodes 15 Triglochin maritima 25 Bare mud 40
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 103
Carex ramenskii - Poa Eminens - Argentina egedii Community (Code: Cr-Pe-Ae) This community covers flats slightly removed from the channel. Either Carex ramenskii or Poa eminens can be dominant over small areas. Plantago maritima is generally also present. Cattle grazing pressure appears to limit the size of the grass and sedge. This community covers an extensive flat on the north side of the 2nd-highest site (TS01).
Average % cover values (number of plots=3), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 40 Plantago maritima 10 Poa eminens 20 Ranunculus cymbalaria t Argentina egedii 15 Bare mud 20
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 104
Poa eminens - Argentina egedii - Carex Community (Code: Pe-Ae-C) This community consisted of a nearly equal mix of Poa eminens and sedges (Carex lyngbyei and Carex ramenskii), with Argentina egedii. Several other species also occurred. This was the dominant community adjacent to the channel at the 2nd-lowest site(TS02).
Average % cover values (number of plots=5), rounded to the nearest 5% (less than 2.5% = trace[t]). Poa eminens 30 Plantago maritima 5 Carex lyngbyei 20 Puccinellia nutkaensis 5 Argentina egedii 15 Triglochin maritima t Carex ramenskii 15 Ranunculus cymbalaria t Bare mud 20
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 105
Poa eminens - Plantago maritima - Argentina egedii Community (Code: Pe-Pm-Ae) Similar to the previous community, but with no sedge and a higher percentage of Plantago maritima. This community occurs on raised levees along the channel. Found only at the 2nd-lowest site (TS02).
Average % cover values (number of plots=2), rounded to the nearest 5% (less than 2.5% = trace[t]). Poa eminens 40 Aregentina egedii 10 Plantago maritima 15 Bare mud 35
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 106
Carex lyngbyei - Argentina egedii Community (Code: Cl-Ae) The Poa eminens - Argentina egedii - Carex Community at the 2nd lowest site transitions into this community in slightly lower areas.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 70 Poa eminens t Aregentina egedii 5 Triglochin maritima t Bare mud 25
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 107
Plantago maritima - sedge - Poa eminens Community (Code: Pm-C-Pe) A single example of this community was found on the levee around a small side-gut feeding into the channel at the 2nd-lowest site (TS02).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Plantago maritima 20 Argentina egedii 5 Carex lyngbyei 10 Triglochin maritima t Carex ramenskii 10 Puccinellia nutkaensis t Poa eminens 10 Ranunculus cymbalaria t Bare mud 45
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 108
Leymus mollis - Poa eminens - Argentina egedii Community (Code: Lm-Pe-Ae) This Leymus mollis community was found on a few raised levees at the 2nd-lowest site (TS02).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Leymus mollis 20 Carex lyngbyei 15 Poa eminens 20 Plantago maritima 10 Argentina egedii 20 Bare mud 15
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 109
Carex ramenskii - Plantago maritima Community (Code: Cr-Pm) This community was found on a levee at the 2nd-highest site (TS01). The presence of Agrostis scabra indicates that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 30 Poa eminens 5 Plantago maritima 20 Argentina egedii 5 Hordeum brachyantherum 15 Agrostis scabra t Bare mud 25
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 110
Argentina egedii - Poa eminens Community (Code: Ae-Pe) This community was found on a levee at the 2nd-highest site (TS01). The presence of Agrostis scabra indicates that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Argentina egedii 40 Carex ramenskii 10 Poa eminens 20 Agrostis scabra 10 Plantago maritima 10 Puccinellia nutkaensis t Bare mud 10
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 111
Diverse Levee Community (Code: dl) This community was found on a levee along a small gut feeding into the channel at the 2nd-highest site (TS01). The presence of Agrostis scabra and Poa annua indicate that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Agrostis scabra 40 Carex lyngbyei 5 Poa eminens 20 Ranunculus cymbalaria t Hordeum brachyantherum 15 Poa annua t Argentina egedii 10 Lomatogonium rotatum t Bare mud 10
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 112
Carex lyngbyei - Eleocharis kamtschatica Community (Code: Cl-E) This community was found in a small gut feeding into the channel at the 2nd-highest site (TS01).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 50 Eleocharis kamtschatica 30 Bare mud 20
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 113
Hordeum brachyantherum - Agrostis scabra Community (Code: Hb-As) This community was found on a levee at the 2nd-highest site (TS01). The presence of Agrostis scabra indicates that this community is rarely inundated by salt water.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Hordeum brachyantherum 30 Puccinellia nutkaensis 10 Agrostis scabra 20 Ranunculus cymbalaria 5 Bare mud 35
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 114
Sedge - Argentina egedii - Ranunculus cymbalaria Community (Code: C-Ae-Rc) This community is primarily a mix of sedges and Argentina, with a good amount of Ranunculus cymbalaria. Either of the two common sedges can be the dominant species. In higher areas, a mix of less salt-tolerant grass species is found as well. This community was only found at the highest site (TS00). Higher areas within this community are almost never inundated by salt water, as is shown by the presence of such species as Poa annua and Trifolium repens.
Average % cover values (number of plots=3), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex ramenskii 25 Plantago maritima t Argentina egedii 20 Agrostis scabra t Carex lyngbyei 15 Hordeum brachyantherum t Ranunculus cymbalaria 15 Poa anua t Poa eminens t Triglochin palustris t Trifolium repens t Bare mud 30
Hordeum brachyantherum - Argentina egedii Community (Code: Hb-Ae) This community is rarely inundated by salt water. It was found on the levee on the north side of the highest channel (TS00).
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Hordeum brachyantherum 60 Trifolium repens t Argentina egedii 15 Polygonum aviculare t Poa annua t Plantago maritima t Agrostis scabra t Lomatogonium rotatum t Poa eminens t Achillea millefolium t Bare mud 20
Puccinellia nutkaensis - Argentina egedii - Carex ramenskii Community (Code: Pn-Ae-Cr) This community includes a diverse mix of species, including some that are not highly salt-tolerant.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]). Puccinellia nutkaensis 40 Hordeum brachyantherum t Argentina egedii 30 Polygonum aviculare t Carex ramenskii 15 Plantago maritima t Carex lyngbyei t Lomatogonium rotatum t Bare mud 10
Puccinellia nutkaensis - Carex ramenskii Community (Code: Pn-Cr) This was the only community found at the highest site (TS00) that contained only species that are highly salt-tolerant.
% cover values from the single plot in this type, rounded to the nearest 5% (less than 2.5% = trace[t]).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 115 Puccinellia nutkaensis 50 Atriplex alaskensis t Carex ramenskii 10 Plantago maritima t Bare mud 40
Carex - grass Gut Community (Code: C-grass) This community was found in the channel at the highest site (TS00), and consisted primarily of sedges and grasses.
Average % cover values (number of plots=2), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 35 Trifolium repens t Carex ramenskii 30 Polygonum aviculare t Unidentified grass 10 Bare mud 20
Carex lyngbyei - Argentina egedii - Plantago maritima Community (Code: Cl-Ae-Pm) This community occurred on the north bank of the highest channel (TS00).
Average % cover values (number of plots=2), rounded to the nearest 5% (less than 2.5% = trace[t]). Carex lyngbyei 40 Hordeum brachyantherum t Argentina egedii 15 Trifolium repens t Plantago maritima 5 Lomatogonium rotatum t Agrostis scabra 5 Bare mud 35
Grass - Trifolium repens Community (Code: grass-Tr) A freshwater community occurring on a high levee to the south of the highest channel (TS00).
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 116 Appendix B 2010 Diet Samples for Juvenile Coho and Sockeye Collected in tributaries of the Fox River
B1. IRI diagram of O. nerka diet composition at TS01 on 24 May 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 117
B. 2. IRI diagram of O. nerka diet composition at TS01 on 21 June 2010
B.3. IRI diagram of O. kisutch diet composition at TS01 on 3 May 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 118
B.4.IRI diagram of O. kisutch diet composition at TS01 on 24 May 2010
B.5. IRI diagram of O. kisutch diet composition at TS01 on 21 June 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 119
B.6. IRI diagram of O. kisutch diet composition at TS01 on 27 July 2010
Walker et al. 2013 Fox River Estuary Juvenile Salmon Habitats Page 120