Micronekton community structure in the epipelagic zone of the northern California Current upwelling system

Progress in Oceanography 80 (2009) 74–92

Contents lists available at ScienceDirect

Progress in Oceanography journal homepage: www.elsevier.com/locate/pocean

Micronekton community structure in the epipelagic zone of the northern California Current upwelling system A. Jason Phillips a,*, Richard D. Brodeur b, Andrey V. Suntsov b,1 a b

Cooperative Institute for Marine Resources Studies, Oregon State University, 2030 SE Marine Science Dr., Newport, OR 97365, USA National Marine Fisheries Service, Northwest Fisheries Science Center, 2030 SE Marine Science Dr., Newport, OR 97365, USA

a r t i c l e

i n f o

Article history: Received 17 September 2008 Received in revised form 12 December 2008 Accepted 15 December 2008 Available online 25 December 2008 Keywords: Pacific Ocean Marine micronekton Lanternfish Community composition Species diversity California Current

a b s t r a c t Spatial and temporal variability in the micronekton community and in oceanographic conditions were evaluated from nighttime midwater trawl samples collected between Heceta Head, Oregon (44.0°N) and Willapa Bay, Washington (46.6°N). Collections from 13 cruises (176 trawls) from 2004 to 2006 yielded over 17,000,000 micronekton individuals (350,000 excluding euphausiids), representing 76 taxa and 43 families. The community was numerically dominated by euphausiids, followed in decreasing order by midwater shrimp (Sergestes similis), lanternfishes (Myctophidae), late larval/juvenile rockfishes (Sebastes spp.), age-0 Pacific hake (Merluccius productus), and pelagic squid (Abraliopsis felis). We used cluster analysis, ordinations, multi-response permutation procedures (MRPP), and indicator species analysis (ISA) to examine community structure of the 28 dominant taxa. Ordination and cluster results indicated that distance from shore and sea-floor depth best characterized habitats used by different assemblages of the micronekton community. Temperature and salinity at various depths influenced community structure to a lesser extent, along with Ekman transport. MRPP and ISA results indicated that nearly all dominant taxa were associated with cross-shelf gradients. Based upon a comparison between historical samples collected in 1976 and 1981 and comparable trawls from this survey, distinct decadal differences among micronektonic fish assemblages were observed, including more juvenile flatfishes and rockfishes but a lower diversity of mesopelagic fishes, which may be related to interdecadal environmental changes between the two time periods. This study represents the first examination of the relationships between both vertebrate and invertebrate members of the epipelagic nekton community. Ó 2008 Elsevier Ltd. All rights reserved.

1. Introduction The California Current, which forms the eastern limb of the North Pacific subtropical gyre, is one of the major eastern boundary currents in the world. It supports highly productive pelagic ecosystems via upwelling-related enrichment (Wooster and Reid, 1963; Barber and Smith, 1981; Bakun, 1993). The pelagic ecosystem of the northern California Current (NCC) is subject to significant seasonal and decadal variability (Peterson and Schwing, 2003; Brodeur et al., 2005a; Legaard and Thomas, 2006) and is heavily influenced by oceanic and terrestrial effects (Hickey and Banas, 2003). Similar to other upwelling regions, the NCC is characterized by high biomasses of pelagic fish stocks such as sardine, anchovy, hake, and mackerel (Brodeur et al., 2003, 2005b).

* Corresponding author. Present address: 104 COAS Administration Building, Corvallis, OR 97331, USA. Tel.: +1 541 231 5021; fax: +1 541 867 0389. E-mail address: [email protected] (A. Jason Phillips). 1 Present address: Scripps Institution of Oceanography, University of California San Diego, 9500 Gilman Dr., La Jolla, CA 92093, USA 0079-6611/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved. doi:10.1016/j.pocean.2008.12.001

Similar to many other areas of the ocean, a substantial part of pelagic biomass in the NCC is concentrated in what is known as micronekton: a diverse community of small but actively moving fishes, cephalopods, and crustaceans, all ranging in length from 2 to 10 cm (Blackburn, 1968; Brodeur and Yamamura, 2005; Brodeur et al., 2005b; Pereyra et al., 1969). This diverse assemblage is known to have significant influence on vertical transport of organic material from the surface to deeper waters through daily vertical migrations. Such transport constitutes a significant food source for larger nektonic organisms (Hidaka et al., 2001; Yamamura and Inada, 2001). The importance of small pelagic fishes in transferring energy to top predators is well-recognized in marine ecosystems (Pearcy, 1972b; Pereyra et al., 1969; Cury et al., 2000; Cartes et al., in press). However, the role of lanternfishes and other components of the micronekton assemblages in upwelling systems remains enigmatic. Micronektonic fishes in the NCC have been extensively studied in terms of species composition (Pearcy, 1964; Willis, 1984), crossshelf and vertical distribution (Pearcy and Laurs, 1966; Pearcy, 1976, 1983; Pearcy et al., 1977; Willis and Pearcy, 1982), acoustic patchiness and structure (Greenlaw and Pearcy, 1985; Kalish et al., 1986), distribution in relation to large-scale physical and chemical

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

75

variables (Aron, 1962), population structure (Willis and Pearcy, 1980), feeding habits (Tyler and Pearcy, 1975; Pearcy et al., 1979; Fisher and Pearcy, 1983), and growth and reproduction (Smoker and Pearcy, 1970). Among the micronekton, crustaceans have been studied less than fishes, but studies have been made of their vertical and geographical distribution (Pearcy and Forss, 1966, 1969; Pearcy, 1970, 1972a; Krygier and Pearcy, 1981) and diets (Renfro and Pearcy, 1966; Nishida et al., 1988). Similarly, relatively little is known of micronektonic cephalopods, with the exception of studies on composition and distribution (Pearcy, 1965; Jefferts, 1982) and acoustic assessment (Jefferts et al., 1987). Much of the present knowledge on micronekton for the North Pacific is summarized by Brodeur and Yamaura (2005). Other than a limited study of micronekton collected during a single cruise in Astoria Canyon (Bosley et al., 2004), there has been no comprehensive examination of micronekton community structure. In addition to the potential advantage of the community-oriented approach to understanding the general functioning, persistence, and variability of micronekton, changes in assemblages of these communities over time can be useful proxies for documenting environmental variability or major regime shifts (Watanabe and Kawaguchi, 2003). The objectives of the present study were to assess spatial and temporal variations in composition, concentration, and other characteristics of the near-surface micronekton community off the coast of Oregon and Washington, and relate them to environmental conditions in the region during the main upwelling season (April to September) from 2004 to 2006. We also compare these recent results to observations made several decades earlier during highly different environmental conditions.

At each station, a Nordic 264-rope trawl was towed for 15– 30 min with the headrope at a target depth of 30 m, with two exceptions. For the first two cruises in 2004, we used a modified Cobb trawl with a mouth opening one-third the size of the Nordic rope trawl (see Phillips et al., 2007). The Nordic 264-rope trawl had an effective fishing mouth 12 m high and 28 m wide (336 m2) based on net mensuration estimates and variable mesh sizes (162.6 cm at mouth to 8.9 cm at cod end) (Emmett et al., 2004), with a 6.1-m long, 3-mm stretched knotless web liner in the cod end. All trawls were conducted at night except for two day tows identified in Table 1. After removing all fishes and invertebrates >10 cm in length, the catch was subsampled as follows: samples with a remaining volume of unsorted catch 60.25 m3 were frozen in their entirety, while samples with a remaining volume of unsorted catch >0.25 m3 were subsampled in the amount of 0.25 m3 or 20% of the entire sample (whichever was larger). The retained unsorted catches were frozen at sea and later thawed and sorted in the lab. Species densities were determined by multiplying the distance towed (as determined by a flowmeter) by the mouth opening of the net. For more detailed collection methodology, see Phillips et al. (2007). Individuals captured were identified to the lowest possible taxonomic level (mostly species). However, due to the large number of individuals collected and ambiguous meristics, some of the taxa were identified to a higher taxonomic level than species (Table 2). Two specific examples are: late-larval/juvenile rockfishes (Sebastes spp.) and Euphausiidae (family level only). At each station, physical data were collected using a Sea-Bird SBE25 CTD cast from the surface to 100 m in depth. Ekman transport vectors were expressed as monthly averages at a 1  1 cell resolution and were obtained from the Pacific Fisheries Environmental Laboratory website (http://www.pfeg.noaa.gov).

2. Methods

2.2. Data analysis

2.1. Sampling procedures

Shannon–Wiener diversity (H0 ) (MacArthur and MacArthur, 1961) and Pielou’s index of evenness (J0 ) (Pielou, 1969) were calculated from a matrix of (64) micronekton taxa (Table 2) grouped by cruise. Cephalopoda, Oegopsida, Gonatidae, Euphausiids (not quantifiable), Teleostei, Osmeridae, Bathylagidae, Paraplepididae, Myctophidae, Gadiformes, Scorpaenidae, and Citharichthys spp. were excluded because they were most likely unidentified individuals of lower taxonomic categories already included. Using 65 micronekton, taxa-area curves were generated to assess adequacy of sample size (McCune and Grace, 2002). These determined 76 taxa for a first-order jackknife estimate, and 80 taxa for a second-order jackknife estimate (±1 SD; Fig. 2). Based on jackknife estimates, which tended to be positively biased with large sample sizes, it appeared that adequate sample sizes for this survey were obtained. Several multivariate analyses were used to investigate the community structure, including (1) Nonmetric multidimensional scaling (NMS) ordinations, (2) one-way and two-way cluster analyses, (3) multi-response permutation procedures (MRPP), and (4) indicator species analysis (ISA). Analyses were conducted using PC-ORD (version 5) software (McCune and Mefford, 1999).

In 2004, the Northwest Fisheries Science Center (NWFSC) Fish Ecology Division initiated a Stock Assessment Improvement Program (SAIP) to survey juvenile fishes off central Oregon and Washington. Juvenile fishes and other micronekton were sampled with midwater trawls from summer to fall 2004–2006. Several stations were sampled along four transects: Heceta Head (HH) (44.00°N), Newport (NH) (44.65°N), and the Columbia River (CR) (46.16°N) off Oregon; and Willapa Bay off Washington (WB) (46.67°N) (Fig. 1). Stations ranged from approximately 20 to 100 km offshore along each transect (Table 1). 48 N

Willapa Bay, Washington (46.67°) Columbia River, Oregon (46.16°)

46 N

2.3. Nonmetric multidimensional scaling (NMS) analysis Newport, Oregon (44.65°) Heceta Head, Oregon (44.00°)

44 N

SAIP survey OSU survey 126 W

124 W

Fig. 1. Location of the sampling stations from the NOAA Fisheries, NWFSC SAIP survey from 2004 to 2006 and Oregon State University experimental midwater trawls from July 1976 and September 1981. The solid line indicates the 200 m isobath.

For multivariate analyses, only taxa that occurred in at least 10% of trawls taken over bottom depths of either <200 m or >200 m (total 28 taxa) for all years combined were used (Table 3). NMS was conducted using Bray–Curtis similarity matrices with a grouped-average linkage strategy (see Clarke and Ainsworth, 1993 and McCune and Grace, 2002 for a detailed description of NMS analysis). The taxon data were fourth-root transformed to deemphasize the dominant taxa. We experimented with several other transformation (e.g., loge-transformed and presence/absence) and subsets

76

Table 1 Stations occupied during each cruise. (+) indicates positive tow. All tows were conducted at night unless otherwise noted. Station

Sea floor depth (m)

Dist from shore (km)

Lat. (°N)

Long. (oW)

2004 June 30– July 1

2005 September 31– October 4

+

+

+

+

+

+

+

+

+

+

November 5–11

HH05 HH11 HH15 HH20 HH25 HH29 HH35 HH37 HH45 HH46 HH55 HH65 HH75

77 115 129 150 117 100 519 950 1650 1600 3000 3000 3000

9.26 21.1 27.78 37.4 46.3 53 64.82 68.9 83.34 91.9 101.86 120.38 138.9

44.00 44.00 44.00 44.00 44.00 44.00 44.00 44.00 44.00 44.00 44.00 44.00 44.00

124.26 124.40 124.49 124.61 124.72 124.80 124.96 125.00 125.19 125.29 125.42 125.66 125.89

NH05 NH15 NH25 NH35 NH45 NH55 NH65 NH75 NH85

60 87 297 435 700 2889 2880 2950 3150

9.26 27.4 46.3 64.6 83.34 101.86 120.38 138.9 157.42

44.65 44.65 44.65 44.65 44.65 44.65 44.65 44.65 44.65

124.18 124.41 124.65 124.88 125.12 125.37 125.60 125.83 126.03

WB09 WB10 WB19 WB20 WB30 WB40 WB50

55 66 110 113 293 910 1020

19.3 21.7 34.1 39.1 55.6 71.1 86.5

46.67 46.67 46.67 46.67 46.67 46.67 46.67

124.30 124.31 124.50 124.52 124.78 124.98 125.18

+

CR10 CR20 CR30 CR40 CR50

81 135 732 853 1423

21.7 39.1 56.5 75.4 95.2

46.16 46.16 46.16 46.16 46.16

124.22 124.45 124.68 124.92 125.18

+ + + + +

June 7–11

+ +

+

+

+

+

+

July 10–14

August 15–19

+

+

+

+

+

September 19–23

October 19–23

May 15–16

June 15–17

August 7–11

September 24–28

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+

+ + + + + +

+ + + +

+ + + + +

+ + + Day+ Day+

+ +

+ + + + + + + +

+ + + + + + +

+ + + +

+ + + +

+ + + + +

+ + + +

+

+

+

+ + +

+

+ + + +

+

+

+

+ + +

+ +

+ + + +

+ + + +

+ + + +

+ + + + +

+ + + + +

+ + + +

+ + +

+ + + +

+ + + +

+ + +

+ + + +

+ + + +

+ + + + +

+ + + + +

+ + + + +

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

August 4–8

2006

Table 2 Phylogenetic listing of micronekton taxa captured in Stock Assessment Improvement Program (SAIP) Larval/Juvenile Survey trawls for each year. Concentration indices were calculated by numerical composition for quantifiable micronekton taxa with the exception that Euphausiidae were not quantifiable, but an index was calculated and included sincee they are most abundant taxa. Index numbers are standardized to no. per 106 m3. F.O.; indicates frequency of occurrence. (j) indicates late larvae and young-of-the-year juveniles; (a) indicates adult stage. Class/family

Cephalopoda Loliginidae Oegopsida Enoploteuthidae Gonatidae

Bathylagidae Stomiidae Paralepidae Myctophidae

Gadiformes Gadidae Merluccidae Ophidiidae Zoarcidae Trachipteridae Scorpaenidae

Hexagrammidae

Squid and octopuses (j) California market squid (a) Squid (not Loliginidae) (j) N/A (j,a) Armhook squid (j) (j) Boreopacific armhook squid (j,a) (j) Clawed armhook squid (j,a) (a) Octopus squid (a) Boreal clubhook squid (j,a) N/A (j,a) Octopus (j) Krill (a) Midwater shrimp (a) Shrimp (j) Pink shrimp (j,a) Bay shrimp (a) Bony fish (j) Leaflike eel (l) Pacific herring (j,a) Northern anchovy (j,a) Smelt (j) Night smelt (a) Eulachon (j,a) Whitebait smelt (a) Deepsea smelt (j,a) Eared blacksmelt (j,a) Longfin drangonfish (a) Pacific blackdragon (a) Barracudina (j) Slender barracudina (j,a) Lanternfish (j) Dogtooth lampfish (a) California headlightfish (j,a) (j) Pinpoint lampfish (j,a) Broadfin lampfish (j,a) Northern lampfish (j,a) California lanternfish (a) Blue lanternfish (j,a) Cod and hake (j) Pacific tomcod (j) Pacific hake (j) Brotula & cusk eel (j) Eelpout (j) King-of-the-salmon (j) Rockfish & Thornyhead (j) Rockfish (j) Thornyhead (j) Lingcod (j)

Scientific name

Cephalopoda spp. Loligo opalescens Oegopsida Abraliopsis felis Gonatidae Gonatopsis spp. Gonatopsis borealis Gonatus spp. Gonatus onyx Berryteuthis spp. Octopoteuthis deletron Onychoteuthis borealijaponicus Chiroteuthis calyx Octopoda Euphausiidae Sergestes similis Caridea Pandalus jordani Crangon spp. Teleostei Thalassenchelys coheni Clupea pallasii Engraulis mordax Osmeridae Spirinchus starksi Thaleichthys pacificus Allosmerus elongatus Bathylagidae Lipolagus ochotensis Tactostoma macropus Idiacanthus antrostomus Paralepididae Lestidiops ringens Myctophidae Ceratoscopelus townsendi Diaphus theta Nannobrachium spp. Nannobrachium regale Nannobrachium ritteri Stenobrachius leucopsarus Symbolophorus californiensis Tarletonbeania crenularis Gadiformes Microgadus proximus Merluccius productus Ophidiidae Zoarcidae Trachipterus altivelis Scorpaenidae Sebastes spp. Sebastolobus spp. Ophiodon elongatus

2004 n = 50

2005 n = 81

2006 n = 49

All years n = 180

n/106 m3

% F.O.

n/106 m3

% F.O.

n/106 m3

% F.O.

n/106 m3

0.10 0.03 0.31 9.34 0.03 0.00 0.91 0.00 0.52 0.01 0.01 0.15 0.03 0.04 8900.00 13.08 0.02 0.00 0.00 0.47 0.002 0.00 4.47 0.02 0.00 0.07 <0.01 0.02 0.91 0.39 0.00 0.07 0.22 4.18 0.00 161.71 <0.01 0.10 0.17 7.86 0.90 34.23 0.70 0.02 41.96 0.00 0.00 0.00 0.01 15.38 0.01 0.00

26 10 32 64 4 0 36 0 30 4 4 28 10 18 90 58 2 0 0 18 2 0 10 2 0 4 4 2 14 12 0 4 26 36 0 64 2 16 8 58 38 70 4 6 16 0 0 4 2 64 2 0

0.10 0.03 0.86 10.02 0.04 <0.01 0.40 0.20 1.14 0.00 0.01 0.15 0.39 0.19 19277.00 102.20 0.04 0.36 0.04 0.01 0.003 0.01 2.16 0.09 0.20 0.04 1.13 0.00 0.27 0.01 0.00 0.00 0.05 0.38 <0.01 13.08 0.00 0.00 0.03 8.77 0.62 31.68 0.00 0.01 0.96 <0.01 0.00 0.02 0.00 9.04 0.00 0.00

9 4 41 54 4 1 23 6 27 0 2 17 30 26 95 46 4 7 7 1 2 1 25 4 1 6 7 0 12 1 0 0 16 19 1 51 0 0 4 44 21 52 0 1 16 1 0 7 0 75 0 0

0.00 0.02 0.31 16.65 0.01 0.02 0.25 0.01 0.36 0.00 0.16 0.48 0.22 0.34 21757.00 507.20 0.07 8.66 1.34 0.56 0.00 0.01 0.06 0.29 0.00 0.11 0.01 0.01 1.91 0.08 <0.01 0.00 0.65 0.10 0.00 8.24 0.13 0.01 0.06 19.62 0.06 25.77 0.00 0.00 36.80 0.00 0.04 0.01 0.00 67.97 0.00 <0.01

0 4 27 51 2 4 20 2 27 0 6 22 27 22 96 49 4 18 4 2 0 2 4 4 0 4 2 2 14 8 2 0 20 2 0 41 8 4 6 45 12 49 0 0 43 0 2 2 0 78 0 2

0.08 11 0.03 6 0.56 34 11.63 56 0.03 3 0.01 2 0.50 26 0.09 3 0.75 28 <0.01 1 0.05 4 0.24 22 0.25 23 0.19 23 17095.10 94 187.69 50 0.04 3 2.52 8 0.38 4 0.29 6 0.002 2 0.01 1 2.23 15 0.12 3 0.09 <1 0.06 5 0.51 5 0.01 1 0.89 13 0.14 6 0.00 <1 0.02 1 0.26 20 1.36 19 <0.01 <1 53.05 52 0.04 3 0.03 6 0.08 6 11.47 48 0.55 23 30.78 56 0.20 1 0.01 2 22.11 23 0.00 <1 0.01 <1 0.01 5 <0.01 <1 26.84 73 <0.01 <1 <0.01 <1 (continued on next page)

% F.O.

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

Octopoteuthidae Onychoteuthidae Chiroteuthidae Octopoda Euphausiidae Sergestidae Caridea Pandalidae Crangonidae Teleostei Family uncertian Clupeidae Engraulididae Osmeridae

Common name

77

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92 70

1.4

50

1.0

Species

30

Distance

0.2

0 4 4 6 0 0 14 0 2 22 2 0 39 35 6 4 16 4 6 6 4 0 0 35 0.00 0.02 0.01 0.06 0.00 0.00 0.13 0.00 <0.01 0.42 0.01 0.00 0.88 0.34 0.41 0.02 0.21 0.01 0.03 0.20 0.02 0.00 0.00 0.16 0 1 5 17 2 0 1 0 0 47 2 5 46 40 0 0 7 0 10 1 0 1 0 25 Pleuronectidae

Stichaeidae Ammodytidae Gobiidae Centrolophidae Paralichthyidae

Sablefish Sculpin (j) Poacher (j) Snailfish (j) Jack mackerel (j) Ronquils (j) Northern ronquil (j) Mosshead warbonnet (j) Pacific sand lance (j) Blackeye goby (j) Medusafish (a) Sanddab (j) Pacific sanddab (j) Speckled sanddab (j) Arrowtooth flounder (j) Petrale sole (j) Slender sole (j) Flathead sole (j) Dover sole (j) Butter sole (j) English sole (j) Curlfin turbot (j) Sand sole (j) Rex sole (j) Anoplopomatidae Cottidae Agonidae Liparididae Carangidae Bathymasteridae

Anoplopoma fimbria Cottidae Agonidae Liparidae Trachurus symmetricus Bathymasteridae Ronquilus jordani Chirolophis nugator Ammodytes hexapterus Rhinogobiops nicholsii Icichthys lockingtoni Citharichthys spp. Citharichthys sordidus Citharichthys stigmaeus Reinhardtius stomias Eopsetta jordani Lyopsetta exilis Hippoglossoides elassodon Microstomus pacificus Isopsetta isolepis Parophrys vetulus Pleuronichthys decurrens Psettichthys melanostictus Glyptocephalus zachirus

0.01 0.03 0.00 0.01 0.01 0.01 0.00 0.44 0.00 0.08 0.00 0.17 1.40 0.62 0.00 0.00 0.00 0.00 0.01 0.15 0.00 0.00 0.01 0.14

2 10 0 6 4 4 0 2 0 18 0 14 58 46 0 0 2 0 6 4 0 2 2 12

0.00 <0.01 0.02 0.05 <0.01 0.00 0.01 0.00 0.00 0.36 0.01 0.11 0.69 0.45 0.00 0.00 0.01 0.00 0.03 <0.01 0.00 <0.01 0.00 0.11

2006 n = 49

n/106 m3 % F.O.

2005 n = 81

% F.O. 2004 n = 50

n/106 m3

Common name Class/family

Scientific name

-0.2

-10

n/106 m3

% F.O.

10

Table 2 (continued)

0.6

Mean distance

Mean no. of taxa

<1 4 3 11 2 1 4 <1 <1 32 2 6 47 40 2 1 8 1 8 3 1 1 <1 24 <0.01 0.01 0.01 0.04 <0.01 1.11 0.04 0.12 <0.01 0.30 0.01 0.10 0.94 0.47 0.11 0.01 0.07 <0.01 0.03 0.10 <0.01 <0.01 <0.01 0.13

n/106 m3

All years n = 180

% F.O.

78

0

50

100

150

200

Number of hauls

Fig. 2. Summary area curve (dark line) for micronekton captured in the SAIP survey between 2004 and 2006. This analysis included the 65 of 76 total micronekton taxa (Cephalopoda, Oegopsida, Gonatidae, Teleostei, Osmeridae, Bathylagidae, Paraplepididae, Myctophidae, Gadiformes, Scorpaenidae, and Citharichthys spp. were excluded because they were most likely unidentified individuals of lower taxonomic categories already included), with a first-order jacknife estimate of 76 and a second-order jacknife estimate of 80. Dashed lines represent ±1 standard deviation. The distance curve (light line) describes the average Sorensen distance between the subsamples and the whole sample, as a function of subsample size. The area curve underestimates the number of species since some taxa (e.g. Sebastes spp.) are represented by an unknown number of species.

(e.g., year, transect) of the data, and all yielded similar results. The fourth-root transformation was selected over loge-transformation because it produces a less severe transformation on the high concentration values than a loge-transformation, and also allowed for the influence of variable concentrations not available in presence/absence transformations (Clarke and Warwick, 2001). This transformation essentially put taxa from different taxonomic groupings on the same scale, and we believe it reduced the bias of different catchability coefficients for the various types of micronekton. The environmental data set was incomplete, with six sea surface temperature and salinity readings missing at 5, 30, and 50 m. These were estimated from adjacent stations in six instances. In some cases, environmental variables were not collected for an entire cruise and could not be estimated, so either the variables or the entire cruise was excluded from analysis. Ordinations were then calculated with 26 variables included and 29% of hauls removed. Preliminary NMS results determined that the 5, 30, and 50 m fluorescence and turbidity variables were weakly correlated (R2 < 0.15) with the ordination, and they were removed. This reduction in variables allowed 87% of the hauls to be included (one cruise was removed due to missing density values). However, another NMS was calculated containing 87% of the trawls, and results indicated that salinity and density were highly correlated, varying no more than 0.03R2. These density variables were removed, which allowed 96% of the trawls to be included in the analysis. Two trawls made during daylight hours, a trawl that hit the sea floor bottom, and four trawls which were empty for the dominant taxa were removed from the analysis. We applied a relativization by maximum to the environmental data to equally weight the different variables. Due to their ubiquitous presence, euphausiids were treated as a biotic environmental variable for the NMS analysis. Variables included in the final analysis are listed in Table 4. For the NMS analysis the following procedure was employed: (1) a random starting configuration was chosen, (2) 250 runs were made for the Monte Carlo test, (3) a 6-axis solution was calculated, then subsequently one axis was reduced and the solution was recalculated for the 5-1 axes solutions, (4) up to 500 iterations were allowed to calculate a stress stability of < 0.000001 over the last 10 iterations, (5) a three-dimensional solution with a mod-

79

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

Table 3 Dominant micronekton taxa captured in at least 10% of trawls <200 m and 10% of trawls >200 m. Taxa are arranged into three distinct assemblages based on indicator species analysis (ISA) and separated by a solid line; (1) offshore, (2) nearshore, and (3) mixed determined by species in group 1 or 2 with p-values <0.02. IV = indicator value, which ranges from 0 to 100 (100 = perfect indication); (N) concentration standardized to no. per 106 m3; F.O. = frequency of occurrence as a percentage; j = late larva and young-of-the-year stage; a = adult stage. Scientific name

Common name

Group

IV

p Value

>200 m (n = 110)

<200 m (n = 66)

N

F.O.

N

F.O.

Tarletonbeania crenularis Diaphus theta Stenobrachius leucopsarus Abraliopsis felis Sebastes spp. Sergestes similis Gonatus onyx Gonatopsis borealis Citharichthys sordidus Symbolophorus californiensis Lestidiops ringens Chiroteuthis calyx Citharichthys stigmaeus Onychoteuthis borealijaponicus Engraulis mordax Lipolagus ochotensis Microstomus pacificus Tactostoma macropus

Blue lanternfish California headlightfish Northern lampfish N/A Rockfish (j) Midwater shrimp Clawed armhook squid Boreopacific armhook squid Pacific sanddab (j) California lanternfish (a) Slender barracudina N/A Speckled sanddab (j) Boreal clubhook squid Northern anchovy Eared blacksmelt Dover sole (j) Longfin drangonfish

2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2

89.3 81.4 79.1 75.5 68.0 66.2 45.5 42.7 39.4 38.2 32.7 32.3 31.1 27.7 22.2 21.8 10.8 10.0

<0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 <0.001 0.016 <0.001 0.016 <0.001 0.012 0.013

50.36 86.78 18.77 18.99 43.63 306.87 1.23 0.82 1.29 0.89 0.43 0.40 0.67 0.38 6.59 1.46 0.04 0.22

90 83 79 82 87 72 45 43 58 38 33 35 47 32 30 22 12 10

0.00 0.05 0.00 0.08 0.49 0.44 0.00 0.00 0.41 0.00 0.00 0.01 0.16 0.02 0.20 0.00 0.00 0.00

3 3 0 17 53 17 0 0 32 0 0 5 30 6 14 0 2 0

Merluccius productus Rhinogobiops nicholsii Glyptocephalus zachirus

Pacific hake (j) Blackeye goby (j) Rex sole (j)

3 (ISA 1) 3 (ISA 2) 3 (ISA 2)

20.7 20.4 18.2

0.037 0.309 0.110

22.51 0.41 0.17

17 35 29

22.78 0.14 0.08

35 29 17

Liparidae Pandalus jordani Lyopsetta exilis Allosmerus elongatus Thaleichthys pacificus Crangon spp. Loligo opalescens

Snailfish (j) Pink shrimp Slender sole (j) Whitebait smelt (a) Eulachon Bay shrimp California market squid

1 1 1 1 1 1 1

23.3 18.9 15.7 13.6 13.6 12.1 10.2

<0.001 <0.001 0.001 <0.001 <0.001 <0.001 0.006

0.01 0.01 0.02 0.00 0.00 0.00 0.01

3 2 3 0 0 0 2

0.10 6.86 0.15 1.40 0.17 1.04 0.05

26 20 18 14 14 12 12

erate stress of 15.8 was selected because adding further dimensions only slightly reduced the stress (four-dimension stress was 12.3), (6) the NMS plots were overlaid with the relativized variables, and (7) the ordination was rotated to maximize correlations with distance from shore.

expands the MRPP test with a description of how well each species separates among groups (McCune and Grace, 2002). Group assignment, indicator values (IV), and p-values were calculated for each taxon within each factor. For all factors a p-value cut off of <0.001 was selected to indicate statistical significance.

2.4. Cluster analysis

2.7. Historical midwater trawl comparisons

Hierarchical cluster analysis was performed on the same data set used for the NMS analysis (fourth-root transformed with excluded trawls) using a Sorensen (Bray–Curtis) distance measure and a flexible beta strategy (b = 0.25; McCune and Grace, 2002). Two-way cluster analyses with a cut off of <10% frequency of occurrence were performed for each year.

Researchers from Oregon State University (OSU) conducted experimental midwater trawls off Oregon in July 1976 and September 1981 (W.G. Pearcy, College of Oceanic and Atmospheric Sciences; OSU, unpublished data). A modified Cobb trawl (100 m2 mouth opening) with an opening/closing cod end was towed at approximately 2–2.5 knots and at various depths for approximately 30 min at each depth along the NH and HH (formerly Siuslaw) transects (Pearcy, 1983). Fish data were made available from trawls with depth ranges (<50 m) comparable to those of the SAIP survey; this yielded 12 samples along two transects corresponding to the NH and HH transects. Although a large amount of data was collected during these surveys (mostly with small obliquely fished nets), much of it could not be compared to the SAIP survey due to significantly different methodologies (see Section 4.1). We concluded that only data from the 12 OSU trawls could be compared with SAIP survey data. SAIP and OSU trawls that were comparable in location (NH and HH transects from 124.88°–125.89°W) and season (July–September) were selected, and 34 trawls from 2004 to 2006 were used for comparison. Although similar in methodology, sampling between the SAIP and OSU surveys differed enough so that a comparison required transformation. A fourth-root transformation was applied to raw catch data from the OSU trawls. This down-weighted the different fishing efforts but still allowed some density effect. A NMS was conducted for all trawls from both surveys that contained at least 10% of the fish taxa (n = 46) found among the 47 trawls included

2.5. Multi-response permutation procedures (MRPP) The MRPP, a nonparametric procedure (similar to a multivariate analysis of variance), was used to test the hypothesis of no difference between two or more groups with a Sorensen (Bray–Curtis) distance measure. The advantage of this analysis was it did not require a normal distribution or linear relationships (McCune and Grace, 2002). Factors and groups within those factors were defined as follows: (1) latitude separated into the four transects (HH, NH, CR, WB), (2) cross-shelf groupings consisting of <200 m and >200 m sea floor depth, (3) month (June, July, August, September, October, and November; May was excluded due to a small [n = 4] sample size), and (4) year (2004, 2005, and 2006). 2.6. Indicator species analysis (ISA) ISA was used to contrast the dominant taxa (same data set used for the NMS) for all of the MRPP factors with 5000 randomizations used in the Monte Carlo test (Dufrêne and Legendre, 1997). The ISA

80

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

Table 4 Pearson’s R2 and Kendall’s tau NMS axes correlation values for significant environmental variables and the dominant taxa for each axis. Significant correlations (p-value <0.01) are in bold type and the highest correlation for each axis is underlined. Variable

Distance offshore Sea floor depth krill 50 m temperature 30 m temperature 5 m temperature 5–30 m temperature 50 m salinity 30 m salinity 5 m salinity 30-5 m salinity E-W Ekman transport N-S Ekman transport A. felis A. elongatus C. calyx C. sordidus C. stigmaeus Crangon spp. D. theta E. mordax G. zachirus G. borealis G. onyx L. ringens Liparidae L. ochotensis L. opalescens L. exilis M. productus M. pacificus O. borealijaponicus P. jordani R. nicholsii Sebastes spp. S. similis S. leucopsarus S. californiensis T. macropus T. crenularis T. pacificus

Axis 1

Axis 2

Axis 3

R2

s

R2

s

R2

s

0.52 0.34 <0.01 0.16 0.15 0.12 <0.01 0.30 0.16 0.05 0.14 <0.01

0.53 0.53 0.13 0.28 0.25 0.21 0.02 0.34 0.23 0.01 0.19 0.03

<0.01 <0.01 0.05 <0.01 0.02 0.02 0.06 <0.01 <0.01 <0.01 <0.01 0.11

0.044 0.07 0.23 0.01 0.07 0.11 0.18 0.07 0.03 0.10 0.06 0.19

<0.01 <0.01 0.02 0.04 0.10 0.05 <0.01 0.03 0.11 0.05 0.01 0.07

0.08 0.07 0.18 0.15 0.27 0.19 0.01 0.19 0.23 0.17 0.08 0.17

0.02

0.05

0.03

0.12

0.06

0.19

0.55 0.04 0.15 0.14 0.08 0.07 0.52 0.06 <0.01 0.28 0.22 0.13 0.08 0.13 0.03 0.11 0.04 0.05 0.15 0.11 0.02 0.27 0.28 0.55 0.25 0.05 0.69 0.06

0.58 0.18 0.32 0.24 0.17 0.21 0.64 0.14 0.02 0.46 0.38 0.33 0.21 0.36 0.15 0.26 0.29 0.17 0.32 0.27 0.05 0.38 0.47 0.65 0.47 0.21 0.67 0.19

<0.01 0.03 0.00 0.14 0.17 <0.01 <0.01 0.01 0.08 <0.01 0.02 <0.01 0.04 0.01 <0.01 0.01 0.33 <0.01 <0.01 0.01 0.20 0.02 0.01 <0.01 <0.01 <0.01 0.01 0.03

0.03 0.12 0.01 0.32 0.34 0.10 0.07 0.10 0.24 0.03 0.16 0.05 0.17 0.09 0.02 0.03 0.50 0.01 0.05 0.05 0.40 0.11 0.11 0.07 0.05 0.04 0.12 0.11

<0.01 0.19 <0.01 <0.01 0.01 0.04 <0.01 0.01 <0.01 <0.01 0.01 0.01 0.08 0.01 <0.01 0.04 <0.01 <0.01 <0.01 0.02 0.02 0.01 0.16 0.02 <0.01 <0.01 <0.01 0.08

0.08 0.27 0.04 0.06 0.07 0.15 0.05 0.17 0.05 0.07 0.05 0.16 0.21 0.15 0.04 0.15 0.06 0.04 0.02 0.09 0.16 0.04 0.45 0.11 0.08 0.04 0.06 0.22

for analysis. A single trawl was removed that contained less than 10% of the taxa and two fish taxa with one occurrence; thus the resulting matrix was comprised of 46 trawls and 44 taxa. Cluster analysis, MRPP, and ISA were also conducted on this data set. Factors for MRPP were survey (SAIP, OSU), year (1976, 1981, 2003, 2004, and 2005), latitude (44.00°N, 44.65°N), and month (July, August, and September). 3. Results 3.1. Taxonomic composition of micronekton An estimated total catch of over 17,000,000 (350,000 excluding euphausiids) individuals from 76 micronekton taxa and 43 families was collected during 2004–2006 (Table 2). Teleost fishes were the most diverse micronekton group, with 28 families and over 50 species collected. At the same time, more than half of all teleost taxa were represented by late-larval and juvenile stages, primarily of bottom species. These included various rockfish, flatfish, poachers, eelpouts, gobies, and others with pelagic development strategies (Table 2). Pleuronectidae generally showed the highest diversity (10 species), followed by Myctophidae (7), Osmeridae (3), Paralichthyidae (2), Stomiidae (2), and Scorpaenidae (2). Remaining families were represented by just one species. However, late larval and juvenile rockfishes, not identified to species level in this study, probably comprised the highest unrecorded diversity considering the numerous rockfish species known from our area. There are between 36 and 40 species of rockfish estimated to be in this area (Love et al., 2002). Based on concentration estimates, Myctophidae were the dominant family within the teleost group, with three dominant species, Diaphus theta (comprising 35.6% of all fish), Tarletonbeania crenularis (20.6%) and Stenobrachius leucopsarus (7.7%), comprising well over half of the fishes collected. Less abundant taxa at the family level were the Scorpeanidae, represented by late larval/juvenile rockfishes (17.4%), and Merluccidae represented by age-0 Pacific hake (14.3%). Remaining taxa totaled less than 3% of all fishes collected (Fig. 3). Overall, our sampling yielded representatives of only 6 true open-ocean families (Myctophidae, Bathylagidae, Paralepidae, Stomiidae, Trachipteridae, and Centrolophidae) with most other taxa being clearly neritic in their adult distribution. In 20% or more of all trawls, only 11 fish taxa were collected, while 30 fish taxa were found in less than 5% of the trawls. Late larval/juvenile rockfishes were most frequently found, followed by the three dom-

Fish

26.41 17% 21.74 14%

1.48 1%

Lanternfishes (a,j)

0.88 1%

Rockfishes (l) Pacific hake (j) Northern anchovy (a)

99.08 66%

Sanddabs (l) Other

2.19 1%

Squids

Crustaceans

1.36 10%

11.44 86%

0.55 4%

Abraliopsis felis Gonatidae Other

15701.91 98.94%

162.60 1.03% 4.92 0.03%

Euphausiidae

Sergestes similis Other

Fig. 3. Dominant taxonomic groups averaged by number of individuals per haul and standardized to individuals per 106 m3. Percent of composition within each group is also presented; (j) indicates late larvae and young-of-the-year juveniles; (a) indicates adult stage.

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

6

No. of Euphausiidae and Sergestes similis per 10 m

3

400 10000 300 6

No. of taxa per 10 m

3

1000

100

200

10 100 1

Ma y0 6n =4 Jun 06 n= Au 7 g0 6n =1 Se 8 p0 6n =2 0

Jun 05 n= 15 Jul 05 n= Au 11 g0 5n =2 Se 0 p0 5n =1 Oc t 05 4 n= 21

0.1 Jun 04 n= Au 5 g0 4n =1 Se 3 p0 4n =1 No 5 v0 4n =2 0

0

81

Crangonidae, and Caridea comprising <2% based on total concentration estimates (Fig. 3). Euphausiids were present in most trawls (94%), and oceanic Sergestes similis occurred in 50% of samples (Table 2). The 10 dominant taxa tended to vary highly between years (Fig. 4). Diaphus theta appeared to have favorable recruitment in 2004, as many of the individuals collected were juveniles (Fig. 4 and Table 5). In 2006, age-0 Pacific hake and late-larval/juvenile rockfishes were widespread during the year and abundant throughout the summer, suggesting strong production. A large number of age-0 Pacific hake were captured in three tows in June 2004 and five tows in August 2004, but were not encountered in the later cruises (Fig. 4). Species diversity and evenness indices were generally higher after the fall transition in each year (Fig. 5). 3.2. Nonmetric multidimensional scaling (NMS) analysis

Month, year, and no. of trawls by cruise Diaphus theta Merluccius productus Tarletonbeania crenularis Sebastes spp. Abraliopsis felis

Stenobrachius leucopsarus Pandalus jordani Engraulis mordax Other Euphausiidae spp. Sergestes similis

Fig. 4. Average monthly abundance (number/106 m3) summaries of the 10 most dominant micronekton taxa captured in SAIP surveys from 2004 to 2006. Euphausiid and sergestid species average abundance were log transformed and plotted as lines to prevent truncation of the other taxa.

inant lanternfish species, a number of larval flatfishes, and larval gobies (Table 2). Six families and eight species of cephalopods were identified in trawls, with gonatids represented by three species, and other families by single species. Cephalopod micronekton was exceedingly dominated by one enoplotheuthid squid, Abraliopsis felis (85.7% of all squids), and two gonatid squids (10.2% of all squids) represented by Gonatus onyx and Gonatopsis borealis. Remaining taxa totaled 4.1% of the total cephalopod catch (Fig. 3). A similar pattern was evident in terms of frequency of occurrence in the samples, with A. felis being the most frequently collected squid in all years, followed by G. onyx and G. borealis. Less frequently occurring species were Chiroteuthis calyx, larval octopods, and Onychoteuthis borealijaponicus, with other species occurring in <10% of all trawls (Table 2). Crustaceans were the least diverse micronekton component, with only 5 families and 6-7 species collected during all sampled years. At the same time, crustaceans dominated the micronekton collections numerically, comprising 98.9% of all taxa. Euphausiids, mostly represented by two common Oregon species, Euphausia pacifica and Thysanoessa spinifera, were the dominant crustacean taxa (98.9%), with other groups such as Sergestidae, Pandalidae,

The three-dimension NMS ordination solution (Figs. 6–8) of species grouped by haul showed a distinct offshore–onshore gradient. Variance explained by each axis was 1 = 46.4%, 2 = 21.6%, and 3 = 18.2%, for a total of 86.2%. In general, distance from shore, sea floor depth, and the cross-shelf gradient, which are three highly inter-correlated variables, essentially described the same ecological gradient and were the most significant indicators of preferred habitat. The NMS results showed sea floor depth and distance from shore to be the most significant variables, with each explaining approximately 50% of the variance along axis 1 (Fig. 7 and Table 4). Fig. 8 clearly indicates that the dominant taxa, with the exception of age-0 Pacific hake, were highly separated along axis 1, which was strongly associated with depth. In fact, nearly all (n = 25) of the taxa were strongly associated with axis 1 (p < 0.01), and 20 of the taxa were correlated (s > 20%) with this axis (Table 4). Significant (p < 0.01) parametric Pearson’s r and non-parametric Kendall tau correlations agreed among variables in general, but non-parametric correlations were typically higher (Table 4). Axis 1 was most strongly associated with distance offshore (s = 0.53) and station depth (s = 0.53). Axis 1 was also moderately associated with temperature at all depths (s = 0.19–0.34), salinity at 50 m and 30 m, and the gradient strength of the salinity halocline (Table 4). Axis 2 was moderately associated with euphausiid concentration (s = 0.23), gradient strength of the thermocline (s = 0.18), and east-west Ekman transport (s = 0.19). Axis 3 was moderately associated with temperature (s = 0.15–0.23) and salinity at 50 and 30 m, and with both east-west and north-south Ekman transports (Fig. 7 and Table 4). Nearly all of the taxa showed statistically significant correlations to various degrees with axis 1 (Table 4). Lanternfishes and the squid (A. felis) had the strongest relationship with axis 1. Axis

Table 5 Results of the multi-response permutation procedure (MRPP) and indicator species analysis (ISA) for latitudinal (transect), cross-shelf (coastal vs. offshore), seasonal (month), and annual (2004, 2005, and 2006) differences in composition of the dominant (top 28) micronekton taxa. The MRPP A-statistic is chance-corrected within-group agreement and describes within-group homogeneity, compared to random expectation. The significant indicator taxa are listed with the category with which each taxon is associated in parentheses. Factor

MRPP A-statistic

p-value

Significant indicator taxa (p-value < 0.05)

Latitude

0.017

<0.001

S. leucopsarus, S. californiensis, S. similis, T. crenularis, D. theta (Newport, NH); M. pacificus, E. mordax (Columbia, CR)

Cross-shelf

0.128

<0.001

All but G. zachirus, R. nicholsii, and M. productus were significant (p-value <0.02) (see Table 3 for groupings)

Montha

0.04

<0.001

G. zachirus, M. productus (June); L. rigens (July); E. mordax, Liparidae (August); O. borealijaponicus, Sebastes spp. (September); A. elongatus, Crangon spp., R. nicholsii (October); C. sordidus, C. stigmaeus, D. theta, and T. macropus (November)

Year

0.061

<0.001

C. sordidus, D. theta, S. californiensis, T. crenularis (2004); R. nicholsii (2005); G. zachirus, L. exilis, M. productus, P. jordani, Sebastes spp. (2006)

a

May was exclude from ISA due to a small sample size (n = 4).

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

2 was strongly associated with age-0 Pacific hake (s = 0.50) and late-larval blackeye goby (R. nicholsii). Axis 2 was also moderately associated with both larval sanddab species, larval rex sole, and late-larval snailfish. Axis 3 was strongly associated with S. similis (s = 0.45), and moderately associated with whitebait smelt, bay shrimp, and late larval snailfish. NMS bubble plots for the nine dominant micronekton taxa are presented in the axes 1–2 plane (Fig. 8), which explained the most variance (68%). Among these species, T. crenularis, A. felis, S. leucopsarus, D. theta, and S. similis were strongly associated with axis 1 in the negative (offshore) direction, while P. jordani was positively (nearshore) associated with axis 1. In general, E. mordax and Sebastes spp. were negatively (offshore) associated with axis 1, but had some concentrations in the positive (nearshore) direction. M. productus appeared to have no relationship with axis 1 but was negatively associated with axis 2.

1.4 Evenness Diversity

1.2 1 0.8 0.6 0.4 0.2

Sep 06 n=20

Jun 06 n=7

Aug 06 n=18

May 06 n=4

Oct 05 n=21

Sep 05 n=14

Jul 05 n=11

Aug 05 n=20

Jun 05 n=15

Sep 04 n=15

Nov 04 n=20

Jun 04 n=5

0 Aug 04 n=13

Diversity and evenness index

82

Cruise

Fig. 5. Diversity and evenness of micronekton taxon (n = 64; taxa excluded because they were most likely unidentified individuals of lower taxonomic categories already included or unquantifiable) grouped by cruise. Diversity measured by the Shannon–Wiener diversity index (H’) and evenness measured by the Shannon Index of Evenness (J0 ). n = number of trawls per cruise.

Two distinct groups, broadly defined here as offshore and inshore assemblages, were outlined in cluster analysis based on concentrations for 28 taxa, including 19 fish, 6 cephalopod, and 3 crustacean taxa (Fig. 9). The largest group consisted of 19 micronekton taxa found primarily offshore. This diverse assemblage also included the most abundant and frequently collected species. Several sub-groups were recognizable within this cluster. The adults of true oceanic fish taxa (i.e., lanternfishes, deep-sea smelts, and barracudinas), cephalopods, and crustaceans (Sergestes similis) were combined here with late-larval/juvenile stages of Sebastes spp., larval sanddabs, and R. nicholsii: all taxa with high dispersal capabilities and either high concentrations (i.e., Sebastes spp.) or frequency of occurrence values (F.O.). A second smaller group of nine micronekton taxa was composed primarily of nearshore species with low concentrations and frequency of occurrence in trawls ranging from 4 to 11%. Two exceptions were late larval rex sole (moderate F.O.) and age0 Pacific hake (high concentration and moderate F.O.).

Cross-shelf < 200 m > 200 m

C. stigmaeus C. sordidus

A. elongatus

R. nicholsii E. mordax Sebastes spp.

Axis 2

A. felis C. calyx D. theta G. borealis L. rigens M. pacificus O. borealijaponicus S. californiensis S. leucopsarus S. similis T. crenularis T. macropus

3.3. One-way cluster analysis

G. onyx L. ochotensis

G. zachirus

Liparidae Crangon spp.

L. opalescens L. exilis P. jordani T. pacificus

M. productus

3.4. Two-way cluster analysis Based on results of MRPP (see below), two-way clusters were grouped by year to investigate interannual variability within the community (Fig. 10). For all years, the trawls strongly clustered into two distinct assemblages (<200 m and >200 m), and in general, the assemblage subgroups appeared to be best defined by month (Fig. 10).

Axis 1

Fig. 6. Nonmetric multidimensional scaling overlay of species weighted averages along axes 1–2. (+) indicates location of weighted species average. Each dot represents the assemblage structure at each station.

30 m salinity 50 m salinity

Axis 3

50 m salinity

Axis 3

Axis 2

Cross-shelf < 200 m > 200 m

30 m salinity

30 m temp. 30 m temp. 50 m temp. Distance offshore & Sea floor depth Axis 1

Distance offshore & Sea floor depth Axis 1

Axis 2

Fig. 7. Nonmetric multidimensional scaling ordination with overlay of collected environmental variables (cut-off r2 > 0.15) for axes 1–2, 1–3, and 2–3. The angle and lengths of lines indicate the direction and strength of the environmental variables in relation to the ordination scores. Variance explained by each axis is 1 = 46.4%, 2 = 21.6%, and 3 = 18.2%, for a total of 86.2%.

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

83

Cross-shelf < 200 m > 200 m

T. crenularis

S. leucopsarus

A.felis

S. similis

Sebastes spp.

E. mordax

M. productus

Axis 2

D. theta

P. jordani

Axis 1 Fig. 8. Nonmetric multidimensional scaling bubble plots for the nine dominant micronekton taxa along the most significant plane (Axes 1–2 explaining 68% of the variance). Size of the points is proportional to the concentration per trawl for each species.

Taxonomic communities showed some variability between years at the assemblage level, but in general each year showed results similar to those of the one-way cluster analyses for all years at the highest group level. The offshore assemblage subgroup consisting of one squid (A. felis), three lanternfishes (D. theta, S. leucopsarus, and T. crenularis), a pelagic shrimp (S. similis), and larval rockfishes (Sebastes spp.) grouped closely (>75% information remaining) for each year. The late-larval sanddabbs (C. sordidus and C. stigmaeus), and late larval blackeye goby (R. nicholsii) grouped closely (>70% information remaining) for all years, but in 2004 these taxa were associated with a nearshore assemblage subgroup, while in 2005–2006 they were associated with the offshore assemblage. A major offshore assemblage from all years combined was composed of four cephalopods (C. calyx, O. borealijaponicus, G. borealis, and G. onyx), one lanternfish (S. californiensis), slender barracudina (L. ringens), eared blacksmelt (L. ochotensis), and longfin dragonfish (T. macropus). This assemblage had high subgroup variability between years. A smaller cluster, mostly nearshore, was composed of two crustacean taxa (Crangon spp. and P. jordani), one cephalopod (L. opalescens), and several fishes (G. zachirus, M. productus, Liparidae spp., L. exilis, and T. pacificus). This assemblage is described in Fig. 9, and it also showed variability in subgroups between years (Fig. 10). 3.5. Multi-response permutation procedures (MRPP) and indicator species analysis (ISA) The MRPP analysis revealed that the cross-shelf gradient was highly significant (p < 0.001) and yielded the strongest taxonomic

association (A-statistic = 0.128) of any of the factors tested (Table 5). Significant A-statistic values are typically <0.1 in studies of community ecology (McCune and Grace, 2002). Given the particularly dynamic nature of the taxa, collection times, and locations in this study, we conclude that this A-statistic indicated a moderate to moderately strong relationship. The ISA results fully supported the MRPP results, revealing 25 taxa that were highly associated (p < 0.02) with the cross-shelf gradient (Tables 3 and 5). Latitude, month, and year all yielded statistically significant (p < 0.001) values, but with weaker associations (A-statistics < 0.1) and fewer significant indicator species. 3.6. Comparison between SAIP and OSU experimental midwater trawls Catch data from the OSU trawls are summarized in Table 6. A total of 31 taxa and 14 families are represented in this data set, with eleven taxa and 5 families encountered in the OSU survey that were not present in any of the SAIP survey samples. NMS ordination and cluster results from the two surveys show distinct groupings between the SAIP and OSU surveys (Fig. 11). Data from both surveys split into two groups along axis 1 of the NMS, which explained 47% of the variance, and the OSU survey split into two distinct groups by year along axis 2. The SAIP data showed no clear secondary split in the NMS. Two large groups (SAIP and OSU surveys combined) separated early in the cluster analysis, supporting the NMS results. The OSU survey data also split into annual groups within the cluster dendrogram. However, in contrast to the NMS results, samples from the SAIP survey split approximately into annual groups within the dendrogram.

84

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

Information remaining (%) 0

25

50

75

100

Abraliopsis felis Tarletonbeania crenularis Stenobrachius leucopsarus Diaphus theta Sebastes spp. Sergestes similis Chiroteuthis calyx Onychoteuthis borealijaponicus Gonatopsis borealis Symbolophorus californiensis Lestidiops rigens Gonatus onyx Lipolagus ochotensis Tactostoma macropus Citharichthys sordidus Citharichthys stigmaeus Rhinogobiops nicholsii Engraulis mordax Microstomus pacificus Allosmerus elongatus Crangon spp. Loligo opalescens Glyptocephalus zachirus Merluccius productus Pandalus jordani Liparidae Lyopsetta exilis Thaleichthys pacificus

Offshore

Coastal

Fig. 9. Cluster analysis showing relationships among the dominant taxa.

The MRPP analysis revealed that survey (OSU, SAIP), year (1976, 1981, 2004, 2005, and 2006), and month were highly significant (p < 0.001) and that latitude (NH and HH transects) was not significant (p < 0.398). Year yielded the strongest taxonomic association (A-statistic = 0.19) of any of the factors tested followed by survey (A-statistic = 0.087; Table 7). The ISA results using year as a factor supported the MRPP results, revealing 20 taxa that were highly associated (p < 0.05) with specific years (Tables 7). Month yielded a weaker association (A-statistics < 0.050) and fewer significant indicator species. 4. Discussion 4.1. Taxonomic composition and assemblages The taxa area curve for the 2004–2006 data set indicated that our sample size was nearing the asymptote in expected number of taxa; thus more sampling during this period would probably not have yielded large numbers of additional taxa. Numbers of taxa estimated were lower than the total number of species, since not all taxa were identified to the species level. It is likely that many taxa collected in the OSU survey and absent from the SAIP survey were not likely to have been encountered with future sampling unless environmental conditions such as SST reverted to their previous conditions. However, the addition of a single species in one year can be significant and should not be ignored. For example, larval arrowtooth flounder (Reinhardtius stomias) were collected for the first time during the May 2006 cruise, and preliminary results from the 2007 and 2008 SAIP survey data suggest that numbers of this important piscivore are increasing along the west coast. In general it appears that diversity and evenness are lower in June, peak by August, and then slightly decline through fall, with a few exceptions (Fig. 5). The May 2006 cruise suggested that diversity and evenness might be highest in spring, but this is speculative, since only a few samples were collected in one year. However, a peak in diversity and evenness during August is supported

by Auth and Brodeur (2006), who found ichthyoplankton diversity to peak in April. A lag would be expected between ichthyoplankton and pelagic late larval/juvenile fish that eventually settle to demersal habitats. The low diversity and evenness observed in June 2004 may have resulted from low sampling effort, with two trawls during the day and three at night far offshore (Table 1). During a major upwelling event in July 2005, both species diversity and evenness decreased to their lowest recorded levels. Two caveats in comparing diversity and evenness between years are the variability in sampling effort between years and the underestimation of diversity if some taxa are not identified to the species level. In almost all cases species that grouped into an offshore, nearshore, or mixed assemblage agreed between the different multivariate tests. However, northern anchovy, four species of late larval flatfishes (Pacific sanddab, speckled sanddab, Dover sole, and rex sole), blackeye goby, and age-0 Pacific hake disagreed to varying levels between tests. Northern anchovy, late larval sanddabs, and Dover sole grouped into offshore assemblages in all analyses except the 2004 two-way cluster, in which case they were associated with the nearshore group. Mixed assemblage species (Pacific hake, blackeye goby, and rex sole) were forced into cluster groups, but it was clear from NMS and MRPP results that these species were widely distributed across the sampling area. 4.2. Temporal, spatial, and environmental variability Analysis of spatio-temporal variations suggests that micronekton assemblages had variable concentrations that were relatively stable as onshore/offshore assemblages during 2004-2006 (Table 5). Cross-shelf differences were greatest, followed by interannual, monthly, and latitudinal differences. In all 3 years, we observed persistent offshore and inshore assemblages, with a limited number of key species driving interannual changes in community structure. This result was expected, considering that drastic changes in the NCC were evident in these years (Brodeur et al., 2006; Barth et al., 2007).

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

75

50

25

0

A

85

Cross shelf

100

> 200 m

Matrix coding Min

Max

Information remaining (%) 0

25

50

75

100

A. felis S. leucopsarus T. crenularis S. similis D. theta Sebastes spp. G. borealis S. californiensis G. onyx L. rigens O. borealijaponicus C. calyx L. ochotensis T. macropus A. elongatus G. zachirus M. pacificus M. productus C. sordidus C. stigmaeus R. nicholsii E. mordax Liparidae L. opalescens

< 200 m

HH05 AUG HH35 AUG HH05 SEP HH15 AUG CR20 NOV CR10 NOV NH15 NOV NH15 AUG HH15 SEP HH25 SEP HH20 NOV NH15 SEP CR30 NOV HH11 NOV WB19 NOV WB09 NOV NH85 JUN NH65 JUN NH55 JUN NH35 AUG NH25 AUG HH45 AUG NH75 AUG NH65 AUG NH55 AUG NH45 AUG NH45 SEP HH35 SEP NH55 SEP NH35 SEP HH45 SEP NH75 SEP NH65 SEP HH55 SEP HH65 SEP NH35 NOV NH45 NOV NH65 NOV HH46 NOV HH37 NOV CR40 NOV CR50 NOV WB40 NOV WB30 NOV WB50 NOV HH75 SEP NH25 SEP NH25 NOV HH29 NOV

Fig. 10. Two-way cluster analysis of taxon by station for samples collected in 2004 (A), 2005 (B), and 2006 (C). Trawls are designated by transect, distance from shore (nautical miles), and month abbreviations.

Many environmental variables were found to relate to assemblages best in non-linear relationships, and some of these variables may simply have been correlated with distance from shore, and had no relationship with the assemblages. Ekman transport may have also played a role in the location of some species, driving them offshore during periods of strong upwelling. The highly statistically significant p-values (<0.001) seen for latitude and month were probably not biologically meaningful, since chance-corrected within-group values were low (A-statistics < 0.05). By latitude, lanternfishes were found to be the most

significant family occurring along the NH transect, though this was probably due to the larger number of samples collected from offshore stations along this transect. Dover sole and northern anchovy appeared to be important along the CR transect and may be related to the Columbia River plume. For the month factor at the taxa level, several late larval/juvenile fishes appeared to be important throughout the season (Table 5). The biological significance of the year effect was less clear at the community level, and late larval/juvenile fishes tended to be the significant indicators as taxa. High levels of late larval/juvenile fishes in some years

86

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

50

25

0

B

75

Cross shelf

100

> 200 m

Matrix coding Min

Max

Information remaining (%) 0

25

50

75

100 WB10 JUL CR10 JUL WB09 OCT WB19 OCT CR10 OCT WB20 JUL HH20 JUL WB20 SEP HH11 JUL WB10 SEP HH29 JUN NH15 JUN HH11 AUG HH20 AUG NH15 OCT CR10 SEP HH20 OCT NH15 AUG CR10 AUG HH29 AUG CR20 AUG WB20 AUG WB20 JUN WB10 JUN HH20 JUN NH25 AUG CR20 SEP WB10 AUG WB30 OCT HH29 OCT CR10 JUN CR20 JUN CR20 OCT HH11 OCT NH55 SEP CR50 OCT CR40 OCT WB30 JUN HH37 OCT CR40 JUN NH35 JUN HH37 JUL WB40 JUN CR30 JUL CR40 JUL HH37 JUN NH45 JUN CR30 JUN WB30 JUL WB40 JUL WB40 AUG NH55 AUG NH45 AUG NH45 SEP WB50 OCT WB50 AUG CR40 AUG CR50 AUG CR30 AUG CR30 SEP WB30 SEP CR50 SEP WB40 OCT CR40 SEP WB40 SEP WB50 SEP CR30 OCT NH35 OCT NH45 OCT NH55 OCT NH65 OCT HH46 OCT NH25 JUN WB30 AUG NH35 SEP NH25 OCT NH35 AUG HH46 AUG HH37 AUG NH25 SEP

A. felis D. theta T. crenularis S. leucopsarus Sebastes spp. S. similis C. sordidus R. nicholsii C. stigmaeus E. mordax C. calyx G. borealis S. californiensis O. borealijaponicus L. rigens M. pacificus G. zachirus G. onyx L. ochotensis A. elongatus Cragon spp. Liparidae L. exilis T. pacificus M. productus P. jordani

< 200 m

Fig. 10 (continued)

suggested strong year classes and may be used to index year class strength (e.g. Phillips et al., 2007). Temperature was positively correlated with the offshore assemblage, while salinity was negatively correlated with it at various depths. The reverse was true for the nearshore assemblage in a non-linear relationship (Table 4). This likely occurred because cold,

high-salinity water was upwelled nearshore, resulting in higher temperatures and lower salinity levels offshore relative to coastal waters. Thus, this apparent relationship to onshore and offshore assemblages was probably related to transport and geography. Some late larval/juvenile fishes (sanddabs, rex sole, snailfishes, age-0 Pacific hake, and blackeye goby) and krill concentrations

87

0

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

75

50

25

C

Cross shelf

100

> 200 m

Matrix coding Min

Max

Information remaining (%) 0

25

50

75

100

A. felis T. crenularis S. leucopsarus Sebastes spp. D. theta S. similis C. calyx G. borealis O. borealijaponicus G. onyx L. ochotensis E. mordax S. californiensis T. macropus C. sordidus C. stigmaeus R. nicholsii L. rigens M. pacificus Cragon spp. L. opalescens G. zachirus L. exilis Liparidae M. productus P. jordani T. pacificus

< 200 m

WB20 MAY CR20 JUL CR10 SEP CR20 SEP WB20 SEP HH29 JUL NH15 SEP WB10 SEP WB30 MAY NH25 SEP WB20 JUN WB30 JUN CR20 JUN HH11 JUL HH20 JUL CR10 JUN WB20 JUL CR10 JUL HH20 SEP HH11 SEP HH29 SEP WB40 MAY CR50 JUL NH55 JUL CR40 JUL CR30 JUL WB40 JUL WB50 JUL WB50 SEP WB30 SEP CR40 JUN WB30 JUL CR30 JUN HH37 JUL NH45 JUL CR30 SEP WB40 SEP CR40 SEP CR50 SEP NH35 JUL HH37 SEP HH46 SEP NH55 SEP NH45 SEP NH35 SEP NH25 JUL NH15 JUL

Fig. 10 (continued)

were related to east-west Ekman transport in a non-linear relationship (Table 4). Sanddabs and blackeye goby appeared to be passively transported offshore, while rex sole, krill, and especially age-0 Pacific hake moved in a shoreward direction during periods of relaxed upwelling. 4.3. Comparison with previous micronekton work in the NCC The summer ocean off Oregon and southern Washington is characterized by intermittent strong coastal upwelling, and usually coincides with the highest catches of mesopelagic fishes (Pearcy, 1964). Early micronekton work at similar latitudes in the NCC covered more seasons and had a broader coverage of the water col-

umn, but sampling was performed with much smaller pelagic nets, such as a 1.8-m Isaacs–Kidd midwater trawl (Pearcy, 1964; Pearcy and Laurs, 1966; Pearcy et al., 1977). Ours was the first broad-scale study of NCC micronekton since the late 1970s (Pearcy et al., 1977; Pearcy, 1976). In addition, our systematic micronekton sampling with a large pelagic trawl covered the spectrum of micronekton diversity in the near-surface assemblage. However, we apparently targeted only the most vigorous vertical migrants that enter the upper 50-m layer at night. The early 1960s studies yielded 40 species of mesopelagic fishes (Pearcy, 1964), while our collections include only 11–12 midwater species. Apparently, our trawls, covering only the epipelagic layer at night, excluded some common midwater representatives, such

88 Table 6 Trawl summaries (phylogenentic order) of Oregon State University open/closing midwater night trawls off Oregon in July 1976 and September 1981. NH65  44.65°N, 125.6°W; HH51 (formerly Siuslaw 51)  44.00°N, 125.32°W; HH41 (formerly Siuslaw 41)  44.00°N, 125.12°W; and HH31 (formerly Siuslaw 31)  44.00°N, 124.95°W. (j) indicates late larvae and young-of-the-year juveniles; (a) indicates adult. Class/family

Common name

Scientific name

Date

192020141616161616181818July-76 July-76 July-76 September- September- September- September- September- September- September- September- September81 81 81 81 81 81 81 81 81

Depth

0– 55 m

Location NH65 Nemichthyidae

Slender snipe eel

48–50 m

30 m

30–50 m

45–50 m

50 m

0–50 m

37 m

27 m

37 m HH31

NH65

NH65

NH65

NH65

NH65

NH65

NH65

NH65

HH51

HH41

0

3

0

0

0

0

0

0

0

0

0

0

Total 3

0 0 0 0 14 0 1

0 0 0 1 83 0 0

0 0 0 0 36 0 0

0 0 0 0 0 0 0

0 1 0 0 0 0 0

0 0 0 0 0 0 0

0 0 0 0 0 0 0

0 0 0 0 0 1 0

0 131 0 0 28 1 1

3 0 0 0 0 0 0

0 0 21 0 0 0 0

0 0 2 0 0 0 0

3 132 23 1 161 2 2

1

0

0

0

0

0

0

0

0

0

0

0

1

4

56

10

0

0

0

0

0

0

49

0

0

119

0

6

4

0

0

0

0

0

0

10

0

0

20

16 0

0 3

1 1

0 0

0 0

0 0

0 0

0 0

0 0

27 0

0 0

0 0

44 4

1 47 0

16 137 3

33 108 0

26 95 0

0 124 0

6 825 0

5 780 0

0 343 0

3 530 0

160 323 0

138 118 0

13 6 0

401 3436 3

2

1

0

0

0

0

0

0

4

0

0

0

7

131

155

73

235

81

374

537

90

416

221

68

0

2381

0

0

0

1

9

57

35

29

167

234

91

0

623

7

69

79

108

15

79

91

129

54

193

61

8

893

0

0

0

0

0

0

0

0

3

0

0

0

3

3

2

0

1

0

0

0

0

0

0

0

0

6

5 5 1 0 0 0

111 7 2 0 0 4

7 2 0 0 0 0

0 0 0 0 0 0

0 0 0 0 1 0

0 0 0 0 0 0

0 0 0 0 0 0

0 0 0 0 0 0

1 0 0 0 0 0

0 0 0 0 0 0

0 0 0 0 0 0

0 0 0 1 0 0

124 14 3 1 1 4

6

32

13

0

0

0

0

0

0

0

0

0

51

0 6

1 85

1 39

0 0

0 0

0 0

0 0

0 0

2 0

0 0

0 0

0 2

4 132

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

Nemichthys scolopaceus Pale snipe eel Nemichthys larseni Engraulididae Northern anchovy (j,a) Engraulis mordax Argentinoidei Bathylagus/Nansenia Argentinoidei Microstomatidae White pencilsmelt Nansenia candida Popeye blacksmelt (j,a) Lipolagus ochotensis Phosichthyidae Panama lightfish Vinciguerria lucetia Sternoptychidae Short silver hatchetfish Argyropelecus hemigymnus Silvery hatchetfish Argyropelecus sladeni Stomiidae Longfin drangonfish (a) Tactostoma macropus Shiny loosejaw Aristostomias scintillans Pacific viperfish Chauliodus macouni Pacific blackdragon (a) Idiacanthus antrostomus Slender barracudina (j,a) Lestidiops ringens Myctophidae California headlightfish (j,a) Diaphus theta Pinpoint lampfish (j,a) Nannobrachium regale Broadfin lampfish (j,a) Nannobrachium ritteri Northern lampfish (j,a) Stenobrachius leucopsarus California lanternfish (a) Symbolophorus californiensis Blue lanternfish (j,a) Tarletonbeania crenularis California flashlightfish (a) Protomyctophum crockeri Bigeye lanternfish (j,a) Protomyctophum thompsoni Scorpaenidae Rockfish (j) Sebastes spp. Liparididae Snailfish (j) Liparidae Northern ronquil (j) Ronquilus jordani Centrolophidae Medusafish (a) Icichthys lockingtoni Bothidae Left-eyed flounders Bothidae Flathead sole (j) Hippoglossoides elassodon Dover sole (j) Microstomus pacificus C-O or Curlfin sole (j) Pleuronichthys sp. Rex sole (j) Glyptocephalus zachirus

30 m 10– 15 m

89

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

100

75

Information remaining (%) 50

25

0

1981

Axis 2

2004-06

1976 Axis 1

SAIP

Axis 3

Year 1976 1981 2004 2005 2006

OSU Stress = 14.6 Axis 1 Chaining = 3.5% Fig. 11. Nonmetric multidimensional scaling ordination and cluster analysis results for 46 comparable trawls and 46 fish taxa between the OSU (Oregon State University 1976 and 1981) and SAIP (Stock Assessment Improvement Program 2004–2006) surveys off central Oregon (44.00–44.65°N and 57–139 km offshore). Variance explained by each axis is 1 = 46.9%, 2 = 15.1%, and 3 = 21.4%, for a total of 83.4%.

as several species of Cyclothone, Protomyctophum thompsoni, Chaulioudus macouni, many stomiids, melamphaeids, and some other deep-dwelling, non-migrant species or species with limited migratory capabilities. In addition, our work covered a relatively narrow region of the continental shelf and slope compared to studies that extended further offshore (Pearcy, 1964; Pearcy et al., 1977). Previous reports off Oregon list three species of lanternfishes accounting for 76% of the total catch, with S. leucopsarus ranking first, followed by D. theta, and T. crenularis (Pearcy, 1964; Pearcy et al., 1977; Brodeur et al., 2003). In the SAIP survey, D. theta, T. crenularis, late-larval Sebastes spp., and age-0 M. productus made

up the bulk of the catch. Stenobrachius leucopsarus was found to be the most abundant fish species in previous NCC studies but the fifth most abundant fish species in our study. Pearcy et al. (1977) found peak abundance of S. leucopsarus (a semi-migrant species) in near-surface waters (0–50 m) at night, a finding in contrast with our results. This may suggest that S. leucopsarus now has a reduced biomass locally, or that conditions in recent years have become favorable at greater depths, such that the ratio of migratory individuals has decreased. Yet another very abundant midwater species, Tactostoma macropus, usually ranking fourth after myctophids (Pearcy, 1964), was rarely collected in our

Table 7 Results of the multi-response permutation procedure (MRPP) and indicator species analysis for survey (OSU, SAIP), annual (1976, 1981, 2004, 2005, and 2006), latitudinal (NH, HH transects), and seasonal (month) differences in composition of the micronekton fish taxa between comparable SAIP and OSU trawls. The MRPP A-statistic is chance-corrected within-group agreement and describes within-group homogeneity, compared to random expectation. The significant indicator species are listed with the category with which each species is associated in parentheses. Factor

p-Value

Significant indicator species (p < 0.05)

Survey

0.087

<0.001

T. macropus, A. scintillans, C. macouni, L. ringens, S. californiensis, P. thompsoni, M. pacificus, Liparis spp., Pleuronichthys spp. (OSU Survey); C. sordidus, C. stigmaeus, Sebastes spp. (SAIP survey)

Year

0.193

<0.001

L. ochotensis, T. macropus, A. scintillans, C. macouni, I. antrostomus, L. ringens, N. ritteri, P. thompsoni, Microstomus pacificus, Pleuronichthys spp., Glyptocephalus zachirus, Liparis spp., Ronquilus jordani (1976); S. californiensis (1981); Pleuronectiform spp. (2004); R. nicholsii, C. stigmaeus (2005); Nannobrachium spp., Sebastes spp. (2006)

Latitude Month

MRPP A-statistic

<0.001

0.398

0.050

<0.001

No significant species L. ochotensis, T. macropus, A. scintillans, C. macouni, I. antrostomus, Liparis spp., R. jordani, M. pacificus, Pleuronichthys spp., G. zachirus (July); C. stigmaeus, E. mordax (August)

90

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

study. This species, which has a similar migration pattern to S. leucopsaurus, apparently now has reduced abundance in our sampling area or has a somewhat deeper center of distribution at night, and was less effectively sampled in our epipelagic trawls. Gonatus squid species made up 60% of the squid catch, and Abraliopsis sp. comprised 20% of the squid micronekton collected in the OSU survey from 1961 to 1963 (Pearcy, 1965). In our survey, A. felis dominated the squid catch, comprising 85% of the total, followed by Gonatidae at 10%. The pelagic shrimp composition was comparable between OSU and SAIP surveys in that S. similis was the dominant species, followed by P. jordani, which tended to dominate nearshore tows. Thus, with the exception of pelagic shrimps, species composition appeared somewhat different in the recent surveys. This could be due to differences in sampling methodology; however, we believe it is more likely due to environmental regime shifts that occurred in the northern California Current during the decades between the two survey periods (Brodeur et al., 2003; Peterson and Schwing, 2003). The comparison of a subsample of the SAIP data with the limited data from OSU opening/closing midwater trawls provided by Dr. Pearcy suggests a that change in the offshore micronekton assemblage has occurred. These assemblages were quite different between the two surveys, as was apparent from results of the NMS, MRPP, and cluster analyses (Fig. 11 and Table 7). An example of this was that late larval sanddabs (Citharicthys spp.) were frequently captured in the SAIP survey and completely absent from the OSU survey. Another difference was that late larval rockfish were captured in greater numbers and smaller sizes in the SAIP survey than in the OSU survey. Also, mesopelagic taxa not captured in any of the SAIP trawls (n = 183) were captured in the OSU experimental trawls at similar depths. Two possible explanations for the differences in mesopelagic taxa between surveys are (1) contamination of OSU midwater trawls, and (2) a change in oceanic conditions such that midwater taxa were no longer present in the epipelagic zone in sufficient numbers to be captured in the SAIP survey. Biological and physical changes resulting from the regime shifts of 1976-1977 and 1989 are the most likely environmental factor resulting in differences in species composition (Brodeur et al., 2003). However, contamination of the OSU trawls is a possibility. The opening and closing of midwater trawls, which were sampled to deeper depths in the OSU survery, may have trapped some deeper taxa in the net mesh. These taxa may then have ended up in the cod end at depths other than where they were captured (Pearcy, 1983). Seven of the mesopelagic taxa captured only in the OSU survey have documented depth distributions to shallower than 30 m, and the remaining three, which have been rarely captured, have documented occurrences at depths above 200 m (Love et al., 2005). Contamination in the shallower tows from deep water was less likely in 1981 because the nets were closed as they increased in depth, but more likely in 1976 because the net was fished from depth to the surface (William Pearcy, Oceanic/Atmospheric Sciences, Oregon State University, 104 COAS Administration Bldg, Corvallis, Oregon 97331 pers. comm., 2008). It is also possible that many semi-migrant mesopelagics did not move up as high in the water column, or had reduced biomass, during the recent surveys. The two species that best support the idea that semi-migrant mesopelagic fish are now in reduced abundances in the upper 50 m of the water column are S. leucopsarus and T. macropus. 4.4. Potential biases It should be noted that several potential biases were introduced into the data due to sampling, but these were accounted for when possible. Due to several uncontrollable factors (e.g., poor weather

and vessel unavailability), consistent sampling of stations between cruises was not possible (Table 1). In addition, the net we used had different catchability coefficients for different taxa of micronekton based on size or escape ability. For example, mesh selected for the smaller micronekton may have extruded through the larger meshes near the mouth opening, while larger nekton were retained in the codend if they were not able to evade the mouth opening. We adjusted for this bias by using a fourth-root transformation of the data, which down-weighted the dominant taxa, but still allowed density to influence the results. Finally, most Sebastes spp. were unidentifiable to species based on morphology or meristics, and subsequently were placed into a single group. Future genetic analysis will assist in determining key species within this group, but will not likely change the ultimate outcome of the results since rockfishes were so closely associated with the offshore assemblage. Mean length of Sebastes spp. was 17.2 mm, and although most were at the low end of the typical range of micronekton (2–10 cm), we felt they were an important enough taxon to be included in the analysis. 4.5. Implications The NCC is an extremely productive and dynamic ecosystem that has undergone extensive changes on several scales in the past decade (Brodeur et al., 2005a). Examples are the recent invasion of Humboldt squid (Dosidicus gigas), dramatically increased occurrences of age-0 Pacific hake, dramatic changes in forage fishes, major seabird mortality events, frequent hypoxic conditions, and several environmental anomalies (Pearcy, 2002; Brodeur et al., 2003, 2005a, 2006; Peterson and Schwing, 2003; Sydeman et al., 2006; Barth et al., 2007; Field et al., 2007; Phillips et al., 2007; Chan et al., 2008). Our study spanned the highly variable period from summer to fall 2004–2006, and we attempted to define the dominant pelagic micronekton assemblages during this period to understand which physical and environmental conditions affected these assemblages. As management of species occupying the California Current (Field and Francis, 2006) and those of other coastal ecosystems (Pikitch et al., 2004; Rosenberg and McLeod, 2005; Francis et al., 2007) moves toward a more holistic, ecosystem-based approach, knowledge of these micronekton assemblages and their interactions with commercially exploited stocks will become indispensable. For example, our findings indicate that juvenile rockfishes show a high degree of spatial overlap with several dominant myctophid species which are not commercially utilized. Thus an estimation of any potential trophic interactions (i.e., competition, predation) occurring between these taxa is clearly warranted.

5. Conclusions This study represents the first time vertebrate and invertebrate micronekton taxa have been studied together in terms of their community structure in the NCC region. The micronekton assemblages in the epipelagic zone of this region were dynamic and highly variable. Distance from shore, sea floor depth, and the cross-shelf boundary were the variables that best characterized the habitats used by different micronekton assemblages. Three distinct assemblages appeared to be present based on their crossshelf distributions: (1) a complex, tightly associated offshore assemblage composed of species-rich micronekton in high densities dominated by lanternfishes, A. felis squid, S. similis shrimp, and larval rockfishes, (2) a simple, low-diversity, species-poor assemblage occupying the coastal environment, and (3) an assemblage composed of larval and juvenile species that shifts offshore to nearshore as the season progresses. This third assemblage is

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

dominated by age-0 Pacific hake, which appears to be a new feature of the NCC (Phillips et al., 2007). Acknowledgments We wish to thank Toby Auth, Paul Peterson, Jesse Lamb, Paul Bentley, Bob Emmett, and the crew of the research and fishing vessels that participated in the cruises and in processing of samples in the laboratory. We would like to thank Bill Pearcy for providing unpublished data. Also, we would like to thank Bill Pearcy, Toby Auth, Ed Casillas and Doug Reese for comments on earlier versions of this manuscript. Funding for the collection and analysis of the data comes from the NOAA Northwest Fisheries Science Center and NOAA’s Stock Assessment Improvement Program. References Auth, T.D., Brodeur, R.D., 2006. Distribution and community structure of ichthyoplankton off the coast of Oregon, USA, in 2000 and 2002. Marine Ecology Progress Series 319, 199–213. Aron, W., 1962. The distribution of animals in the Eastern North Pacific and its relationship to physical and chemical conditions. Journal of the Fisheries Research Board of Canada 19, 271–314. Bakun, A., 1993. The California Current, Benguela Current, and Southwestern Atlantic Shelf Ecosystems: A Comparative Approach to Identifying Factors Regulating Biomass Yields. pp. 199–221. In: Sherman, et al. (Ed.), Large Marine Ecosystems, Stress, Mitigation and Sustainability, Washington, DC, p. 376. Barber, R.T., Smith, R.L., 1981. Coastal upwelling ecosystems. pp. 31–68. In: Longhurst, A.L., (Ed.), Analysis of Marine Ecosystems, Academic press, p. 741. Barth, J.A., Menge, B.A., Lubchenco, J., Chan, F., Bane, J.M., Kirincich, A.R., McManus, M.A., Nielsen, K.J., Pierce, S.D., Washburn, L., 2007. Delayed upwelling alters nearshore coastal ocean ecosystems in the northern California Current. Proceedings of the National Academy of Sciences 104, 3719–3724. Blackburn, M., 1968. Micronekton of the eastern tropical Pacific ocean: faunal composition, distribution and relation to tuna. Fishery Bulletin 67, 71–115. Bosley, K.L., Lavelle, J.W., Brodeur, R.D., Wakefield, W.W., Emmett, R.L., Baker, E.T., Rehmke, K.M., 2004. Biological and physical processes in and around Astoria Submarine Canyon, Oregon, USA. Journal of Marine Systems 50, 21–37. Brodeur, R.D., Pearcy, W.G., Ralston, S., 2003. Abundance and distribution patterns of nekton and micronekton in the Northern California Current Transition Zone. Journal of Oceanography 59, 515–534. Brodeur, R.D., Fisher, J.P., Emmett, R.L., Morgan, C.A., Casillas, E., 2005a. Species composition and community structure of pelagic nekton off Oregon and Washington under variable oceanographic conditions. Marine Ecology Progress Series 298, 41–57. Brodeur, R.D., Seki, M.P., Pakhomov, E.A., Suntsov, A.V., 2005b. Micronekton – What are they and why are they important? PICES Press 13, 7–11. Brodeur, R.D., Yamamura, O., (Eds.), 2005. Micronekton of the North Pacific. PICES Scientific Report 30. p. 115. Brodeur, R.D., Ralston, S., Emmett, R.L., Trudel, M., Auth, T.D., Phillips, A.J., 2006. Recent trends and anomalies in pelagic nekton abundance, distribution, and apparent recruitment in the northeast Pacific Ocean. Geophysical Research Letters 33, L22S08. doi:10.1029/2006GL026614. Cartes, J.A., Hidalgo, M., Papiol, V., Massuti, E., Moranta, J., in press. Changes in the diet and feeding of the hake Merluccius merluccius at the shelf-break of the Balearic Islands: Influence of the mesopelagic-boundary community. Deep-Sea Research I. doi:10.1016/j.dsr.2008.09.009. Chan, F., Barth, J.A., Lubchenco, J., Kirincich, A., Weeks, H.A., Peterson, W.H., Menge, B.A., 2008. Novel emergence of anoxia in the California Current large marine ecosystem. Science 319, 920. Clarke, K.R., Ainsworth, M., 1993. A method of linking multivariate community structure to environmental variables. Marine Ecology Progress Series 92, 205– 219. Clarke, K.R., Warwick, R.M., 2001. Change in marine communities: an approach to statistical analysis and interpretation. Second ed. Plymouth, UK, PRIMER-E Ltd. p. 172. Cury, P., Bakun, A., Crawford, R.J.M., Jarre, A., Quinones, R.A., Shannon, L.J., Verheye, H.M., 2000. Small pelagics in upwelling systems: patterns of interaction and structural changes in ‘‘wasp-waist” ecosystems. ICES Journal of Marine Science 57, 603–618. Dufrêne, M., Legendre, P., 1997. Species assemblages and indicator species: the need for a flexible asymmetrical approach. Ecological Monographs 67, 345–366. Emmett, R.L., Brodeur, R.D., Orton, P.M., 2004. The vertical distribution of juvenile salmon and associated fishes in the Columbia River plume. Fisheries Oceanography 13, 392–402. Field, J.C., Francis, R.C., 2006. Considering ecosystem-based fisheries management in the California Current. Marine Policy 30, 552–569. Field, J.C., Baltz, K., Phillips, A.J., Walker, W.A., 2007. Range expansion and trophic interactions of the jumbo squid, Dosidicus gigas, in the California Current. California Cooperative Fisheries Investigation Reports 48, 131–146.

91

Fisher, J.P., Pearcy, W.G., 1983. Reproduction, growth, and feeding of the mesopelagic fish Tactostoma macropus (Melanostomiatidae). Marine Biology 74, 257–267. Francis, R.C., Hixon, M., Clarke, E., Murawski, S., Ralston, S., 2007. Ten commandments for ecosystem-based fishery scientists. Fisheries 32, 217–233. Greenlaw, C.F., Pearcy, W.G., 1985. Acoustical patchiness of mesopelagic micronekton. Journal of Marine Research 43, 163–178. Hidaka, K., Kawaguchi, K., Murakami, M., Takahashi, M., 2001. Downward transport of organic carbon by diel migratory micronekton in the western equatorial Pacific: its quantitative and qualitative importance. Deep-Sea Research I 48, 1923–1939. Hickey, B.M., Banas, N.S., 2003. Oceanography of the US Pacific Northwest coastal ocean and estuaries with application to coastal ecology. Estuaries 26, 1010– 1031. Jefferts, K., 1982. Zoogeography and systematics of cephalopods of the Northeastern Pacific Ocean. Ph.D. Dissertation, Oregon State University, Corvallis. p. 291. Jefferts, K., Burczynski, J., Pearcy, W.G., 1987. Acoustical assessment of squid (Loligo opalescens) off the central Oregon coast. Canadian Journal of Fisheries and Aquatic Sciences 44, 1261–1267. Kalish, J.M., Greenlaw, C.F., Pearcy, W.G., Holliday, D.V., 1986. The biological and acoustical structure of sound scattering layers off Oregon. Deep-Sea Research I 33, 631–653. Krygier, E.E., Pearcy, W.G., 1981. Vertical distribution and biology of pelagic decapod crustaceans off Oregon. Journal of Crustacean Biology 1, 70–95. Legaard, K., Thomas, A.C., 2006. Spatial patterns of seasonal and interannual variability in chlorophyll and surface temperature in the California Current. Journal of Geophysical Research 111, C06032. doi:10.1029/2005JC003282. Love, M.S., Yoklavich, M., Thorsteinson, L., 2002. The Rockfishes of the Northeast Pacific. University of California Press, Berkeley, California. pp. 405. Love, M.S., Mecklenburg, C.W., Mecklenburg, T.A., Thorsteinson, L.K., 2005. Resource inventory of marine and estuarine fishes of the West Coast and Alaska: a checklist of North Pacific and Arctic Ocean species from Baja California to the Alaska–Yukon border. United States Department of the Interior, United States Geological Survey, Seattle, Washington. p. 276. MacArthur, R.H., MacArthur, J.W., 1961. On bird species diversity. Ecology 42, 594– 598. McCune, B., Grace, J.B., 2002. Analysis of ecological communities. MjM Software Design, Gleneden Beach, Oregon, 300. McCune, B., Mefford, M.J., 1999. PC-ORD, Multivariate Analysis of Ecological Data, Users Guide. MjM Software Design, Gleneden Beach, Oregon, 237. Nishida, S., Pearcy, W.G., Nemoto, T., 1988. Feeding habits of mesopelagic shrimps collected off Oregon. Bulletin of the Ocean Research Institute, University of Tokyo 26, 99–108. Pearcy, W.G., 1964. Some distributional features of mesopelagic fishes off Oregon. Journal of Marine Research 22, 83–102. Pearcy, W.G., 1965. Species composition and distribution of pelagic cephalopods from the Pacific Ocean off Oregon. Pacific Science XIX 2, 261–266. Pearcy, W.G., 1970. Vertical migration of the ocean shrimp, Pandalus jordani: A feeding and dispersal mechanism. California Fish and Game 56, 125–129. Pearcy, W.G., 1972a. Distribution and diel changes in the behavior of pink shrimp, Pandalus jordani, off Oregon. Proceedings of the National Shellfisheries Association 62, 15–20. Pearcy, W.G., 1972b. Distribution and Ecology of Oceanic Animals off Oregon. pp. 351–376. In: Pruter, A.T., Alverson, D.L., (Eds.), The Columbia River Estuary and Adjacent Ocean Waters: Bioenvironmental Studies, University of Washington Press, Seattle, Washington. p. 896. Pearcy, W.G., 1976. Seasonal and inshore-offshore variations in the standing stock of micronekton and macrozooplankton off Oregon. Fishery Bulletin 74, 70–79. Pearcy, W.G., 1983. Quantitative assessment of the vertical distributions of micronektonic fishes with opening/closing midwater trawls. Biological Oceanography 2, 289–310. Pearcy, W.G., 2002. Marine nekton off Oregon and the 1997-98 El Niño. Progress in Oceanography 54, 399–403. Pearcy, W.G., Laurs, R.M., 1966. Vertical migration and distribution of mesopelagic fishes off Oregon. Deep-Sea Research I 13, 153–165. Pearcy, W.G., Forss, C.A., 1966. Depth distribution of oceanic shrimps (Decapoda; Natantia) off Oregon. Journal of the Fisheries Research Board of Canada 23, 1135–1143. Pearcy, W.G., Forss, C.A., 1969. The oceanic shrimp Sergestes similis off the Oregon coast. Limnology and Oceanography 14, 755–765. Pearcy, W.G., Lorz, H.V., Peterson, W., 1979. Comparison of the feeding habits of migratory and non-migratory Stenobrachius leucopsarus (Myctophidae). Marine Biology 51, 1–8. Pearcy, W.G., Krygier, E.E., Mesecar, R., Ramsey, F., 1977. Vertical distribution and migration of oceanic micronekton off Oregon. Deep-Sea Research I 24, 223–245. Peterson, W.T., Schwing, F.B., 2003. A new climate regime in northeast Pacific ecosystems. Geophysical Research Letters 30 (17), 1896. doi:10.1029/ 2003GL017528. Pereyra, W.T., Pearcy, W.G., Carvey, F.E., 1969. Sebastodes flavidus, a shelf rockfish feeding on mesopelagic fauna, with consideration of the ecological implications. Journal of the Fisheries Research Board of Canada 26, 2211–2215. Phillips, A.J., Ralston, S., Brodeur, R.D., Auth, T.D., Emmett, R.L., Johnson, C., Wespestad, V.G., 2007. Recent pre-recruit Pacific hake (Merluccius productus) occurrences in the northern California Current suggest a northward expansion of their spawning area. California Cooperative Fisheries Investigation Reports 48, 215–229.

92

A. Jason Phillips et al. / Progress in Oceanography 80 (2009) 74–92

Pikitch, E.K., Santora, C., Babcock, E.A., Bakun, A., Bonfil, R., Conover, D.O., Dayton, P., Doukakis, P., Fluharty, D., Heneman, B., Houde, E.D., Link, J., Livingston, P.A., Mangel, M., McAllister, M.K., Pope, J., Sainsbury, K.J., 2004. Ecosystem-based fishery management. Science 305, 346–347. Pielou, E.C., 1969. An Introduction to Mathematical Ecology. John Wiley and Sons, New York. 286 pp. Renfro, W.C., Pearcy, W.G., 1966. Food and feeding apparatus of two pelagic shrimps. Journal of the Fisheries Research Board of Canada 23, 1971–1975. Rosenberg, A.A., McLeod, K.L., 2005. Implementing ecosystem-based approaches to management for the conservation of ecosystem services. Marine Ecology Progress Series 300, 270–274. Smoker, W., Pearcy, W.G., 1970. Growth and reproduction of the lanternfish Stenobrachius leucopsarus. Journal of the Fisheries Research Board of Canada 27, 1265–1275. Sydeman, J.W., Bradley, R.W., Warzybok, P., Abraham, C.L., Jahncke, J., Hyrenbach, K.D., Kousky, V., Hipfner, J.M., Ohman, M.D., 2006. Planktivorous auklet Ptychoramphus aleuticus responses to ocean climate, 2005: unusual atmospheric blocking? Geophysical Research Letters 33, L22S09, 10.1029/2006GL026736.

Tyler Jr., H.R., Pearcy, W.G., 1975. The feeding habits of three species of lanternfishes (family Myctophidae) off Oregon, USA. Marine Biology 32, 7–11. Watanabe, H., Kawaguchi, K., 2003. Decadal change in abundance of surface migratory myctophid fishes in the Kuroshio region from 1957 to 1994. Fisheries Oceanography 12, 100–111. Willis, J.M., 1984. Mesopelagic fish faunal regions of the northeast Pacific. Biological Oceanography 3, 167–185. Willis, J.M., Pearcy, W.G., 1980. Spatial and temporal variations in the population size structure of three lanternfishes (Myctophidae) off Oregon, USA. Marine Biology 57, 181–191. Willis, J.M., Pearcy, W.G., 1982. Vertical distribution and migration of fishes of the lower mesopelagic zone off Oregon. Marine Biology 70, 87–98. Wooster, W.S., Reid, J.L., 1963. Eastern boundary currents. In: Hill, M.N. (Ed.), The Sea, Vol. 2. Wiley, pp. 253–280. Yamamura, O., Inada, T., 2001. Importance of micronekton as food of demersal fish assemblages. Bulletin of Marine Science 68, 13–25.