Effect of stocking sub-yearling Atlantic salmon on the habitat use of sub-yearling rainbow trout

Effect of stocking sub-yearling Atlantic salmon on the habitat use of sub-yearling rainbow trout

Journal of Great Lakes Research 42 (2016) 116–126 Contents lists available at ScienceDirect Journal of Great Lakes Research journal homepage: www.el...

2MB Sizes 0 Downloads 97 Views

Journal of Great Lakes Research 42 (2016) 116–126

Contents lists available at ScienceDirect

Journal of Great Lakes Research journal homepage: www.elsevier.com/locate/jglr

Effect of stocking sub-yearling Atlantic salmon on the habitat use of sub-yearling rainbow trout☆ James H. Johnson 1 Tunison Laboratory of Aquatic Science, USGS—Great Lakes Science Center, 3075 Gracie Road, Cortland, NY 13045 USA

a r t i c l e

i n f o

Article history: Received 15 December 2014 Accepted 27 October 2015 Available online 24 November 2015 Index words: Atlantic salmon Rainbow trout Habitat

a b s t r a c t Atlantic salmon (Salmo salar) restoration in the Lake Ontario watershed may depend on the species' ability to compete with naturalized non-native salmonids, including rainbow trout (Oncorhynchus mykiss) in Lake Ontario tributaries. This study examined interspecific habitat associations between sub-yearling Atlantic salmon and rainbow trout as well as the effect of salmon stocking on trout habitat in two streams in the Lake Ontario watershed. In sympatry, Atlantic salmon occupied significantly faster velocities and deeper areas than rainbow trout. However, when examining the habitat use of rainbow trout at all allopatric and sympatric sites in both streams, trout habitat use was more diverse at the sympatric sites with an orientation for increased cover and larger substrate. In Grout Brook, where available habitat remained constant, there was evidence suggesting that trout may have shifted to slower and shallower water in the presence of salmon. The ability of sub-yearling Atlantic salmon to affect a habitat shift in rainbow trout may be due to their larger body size and/or larger pectoral fin size. Future studies examining competitive interactions between these species during their first year of stream residence should consider the size advantage that earlier emerging Atlantic salmon will have over rainbow trout. Published by Elsevier B.V. on behalf of International Association for Great Lakes Research.

Introduction Prior to 1900 Atlantic salmon (Salmo salar) was the dominant migratory salmonid species in Lake Ontario (Webster, 1982). The decline of Atlantic salmon in the Lake Ontario basin coincided with increased European settlement (Huntsman, 1944). Several causal factors have been implicated in the decline, and eventual extirpation of salmon by the turn of the century. Those factors that have been cited most often include severing access to natal streams (mill dams), overfishing, and deforestation (Parsons, 1973). Even before Atlantic salmon became extirpated, management agencies began experimenting with introduction of Pacific salmonids into Lake Ontario with the first release of rainbow trout (Oncorhynchus mykiss) occurring about 1878 (MacCrimmon and Gots, 1972). Continued release of rainbow trout (steelhead) established naturalized populations of the species in several tributaries of Lake Ontario (MacCrimmon and Gots, 1972). By the 1970s juvenile rainbow trout (steelhead) were the numerically dominant salmonid in most of New York's higher quality tributaries that entered Lake Ontario (Johnson and Ringler, 1981). Expanding populations of sea lamprey (Petromyzon marinus) in the Great Lakes beginning in the 1930s led to the decimation of large fish species including lake trout (Salvelinus namaycush) (Christie and Goddard, 2003). It was not until sea lamprey control was achieved in ☆ Communicated by E. Rutherford E-mail address: [email protected]. 1 Tel.: + 607 753 9391x7530.

Lake Ontario in the 1960s that management agencies began restoration efforts for native species such as lake trout as well as a new round of Pacific salmonid introductions (Parsons, 1973). However, it was not until 1987 that efforts were initiated to reestablish Atlantic salmon in Lake Ontario (Stanfield and Jones, 2003). The delay in attempting to restore Atlantic salmon in Lake Ontario was likely due to the overwhelming success of the Pacific salmon introduction program and recognition that wild salmon populations were declining globally in spite of ongoing restoration efforts (Parrish et al., 1998). Renewed binational interest in restoring Atlantic salmon in Lake Ontario led to several studies that addressed anticipated biological impediments that could impact successful reintroduction. One of the major biological impediments identified was the high densities of naturalized Pacific salmonid juveniles that were present in historic Atlantic salmon nursery streams (Jones and Stanfield, 1993). Because of similar juvenile life history, habitat requirements, and current levels of juvenile densities in Lake Ontario tributaries, the non-native species of most concern was rainbow trout (Johnson and Wedge, 1999). Stanfield and Jones (2003) found that habitat conditions greatly influenced competitive interactions between juvenile Atlantic salmon and rainbow trout in tributaries along the northern shoreline of Lake Ontario. Dietrich et al. (2008) examined the effects of stocking high and low densities of Atlantic salmon on abundance and growth of sub-yearling rainbow trout in two Lake Ontario tributaries and concluded that salmon stocking may impact trout. In work done in a Finger Lake tributary in New York that is within the Lake Ontario watershed, Coghlan and Ringler (2005) found that, as stream temperatures increase, Atlantic salmon became increasingly

http://dx.doi.org/10.1016/j.jglr.2015.11.002 0380-1330/Published by Elsevier B.V. on behalf of International Association for Great Lakes Research.

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

117

Table 3 Mean depth, velocity, substrate size, and percent total cover (± standard error) of subyearling rainbow trout (RBT), sub-yearling Atlantic salmon (ATS), and available habitat (AH) during pre-stocking, 2-week post stocking, and 9-week post stocking periods in Orwell Brook and Grout Brook. Orwell Brook-Tubbs, Grout Brook-Lower, and Grout BrookMill were sympatric sites and Orwell Brook-Orwell and Grout Brook-Scott were allopatric sites. Depth (cm)

Velocity (cm/s)

Substrate

Total Cover (%)

15.7 ± 0.5 17.1 ± 0.5 25 ± 1.4 – 16 ± 0.7 22.4 ± 1.2 15.6 ± 0.9 11.5 ± 0.7 19.2 ± 0.8

29.6 ± 1.1 13.2 ± 0.5 41.4 ± 2.7 – 18.6 ± 1.1 68.9 ± 3.7 16.8 ± 1.1 12.8 ± 0.9 44.5 ± 2.2

6.1 ± 0.01 6.1 ± 0.02 6.1 ± 0.03 – 6.2 ± 0.02 6.2 ± 0.02 6 ± 0.03 6.2 ± 0.02 6.1 ± 0.02

9.3 ± 0.5 11.1 ± 0.4 11.1 ± 0.9 – 11.9 ± 0.6 10.9 ± 0.8 5.4 ± 0.6 4.3 ± 0.5 4.9 ± 0.5

Orwell Brook-Orwell RBT pre-stocking 14 ± 0.4 RBT week 2 post 10.5 ± 0.4 RBT week 9 post 20.7 ± 0.6 AH pre-stocking 10.9 ± 0.5 AH week 2 post 9.1 ± 0.8 AH week 9 post 21.7 ± 0.6

25 ± 1.2 11.5 ± 0.7 40.2 ± 2.7 21.9 ± 1.4 7.5 ± 0.7 55.9 ± 2.9

6.1 ± 0.02 6.1 ± 0.02 6.1 ± 0.02 6.1 ± 0.03 6.2 ± 0.03 6.1 ± 0.02

11 ± 0.6 8.8 ± 0.4 12.9 ± 0.7 6.7 ± 0.7 4.3 ± 0.5 8.1 ± 0.5

Grout Brook-Lower RBT pre-stocking RBT week 2 post RBT week 9 post ATS pre-stocking ATS week 2 post ATS week 9 post AH pre-stocking AH week 2 post AH week 9 post

18.8 ± 0.5 19 ± 0.6 21.4 ± 0.7 – 18.9 ± 0.9 19.7 ± 1.5 20.4 ± 1.3 17.9 ± 1 17.9 ± 0.9

34.9 ± 1.2 36.5 ± 1.2 31.0 ± 1.3 – 45.8 ± 2.5 52.9 ± 3.1 24.4 ± 1.7 26.4 ± 1.9 26.9 ± 2

5.7 ± 0.03 5.7 ± 0.03 5.9 ± 0.03 – 5.9 ± 0.03 6 ± 0.03 5.7 ± 0.05 5.6 ± 0.05 5.7 ± 0.05

7.4 ± 0.5 8.7 ± 0.4 8.5 ± 0.4 – 8.8 ± 0.7 7.3 ± 0.8 4.4 ± 0.6 4.1 ± 0.5 3.5 ± 0.4

Grout Brook-Mill RBT pre-stocking RBT week 2 post RBT week 9 post ATS pre-stocking ATS week 2 post ATS week 9 post AH pre-stocking AH week 2 post AH week 9 post

15.2 ± 0.5 17.1 ± 0.5 19.6 ± 0.7 – 16.9 ± 0.4 16.8 ± 0.6 14.4 ± 0.6 13.3 ± 0.6 13.4 ± 0.6

35.1 ± 1.9 36.6 ± 1.3 30.3 ± 1.8 – 42.5 ± 1.2 47.9 ± 3.7 28 ± 1.7 26.1 ± 1.8 30.2 ± 2.2

6.2 ± 0.02 6.1 ± 0.02 6.2 ± 0.03 – 6.2 ± 0.01 6.2 ± 0.02 6.2 ± 0.02 6.1 ± 0.02 6.1 ± 0.03

10.2 ± 0.8 11.9 ± 0.6 13 ± 0.8 – 14.2 ± 0.6 13.9 ± 1.3 7.4 ± 0.8 7.3 ± 0.9 7.3 ± 0.8

Grout Brook-Scott RBT pre-stocking RBT week 2 post RBT week 9 post AH pre-stocking AH week 2 post AH week 9 post

12.9 ± 0.3 12.6 ± 0.6 15.3 ± 0.7 11.3 ± 0.7 10.5 ± 0.6 13.9 ± 0.8

21.6 ± 1 24.7 ± 1.3 36.3 ± 2 16.1 ± 1.1 12.1 ± 0.9 19.9 ± 1.9

6.1 ± 0.02 6.2 ± 0.04 5.9 ± 0.03 5.9 ± 0.04 5.9 ± 0.04 5.9 ± 0.04

9.1 ± 0.6 10.1 ± 0.5 15.9 ± 1.3 6.5 ± 0.6 6.5 ± 0.7 5.5 ± 0.7

Orwell Brook-Tubbs RBT pre-stocking RBT week 2 post RBT week 9 post ATS pre-stocking ATS week 2 post ATS week 9 post AH pre-stocking AH week 2 post AH week 9 post

Fig. 1. Location of allopatric and sympatric sites in Grout Brook and Orwell Brook in Central New York.

favored over rainbow trout in terms of competitive interactions. Trophic interactions between the two species in Lake Ontario tributaries are less well known. However, Johnson and Waldt (2014) provided evidence that juvenile rainbow trout may subtly shift to a more drift feeding strategy in sympatry with juvenile Atlantic salmon. Habitat studies on juvenile Atlantic salmon and rainbow trout in sympatry are rare. Hearn and Kynard (1986) observed that although yearling Atlantic salmon and rainbow trout used similar habitat in tributaries of the White River, Vermont, sub-yearling salmon occupied deeper and swifter water than sub-yearling trout. Stanfield and Jones (2003) found that densities of sub-yearling Atlantic salmon were generally higher at sites with greater abundance of rock cover and lower

Table 1 Number of habitat observations on sub-yearling Atlantic salmon and sub-yearling rainbow trout in allopatric and sympatric sites in Orwell Brook, New York. Orwell Brook Sympatric sites 1 week pre-salmon stocking Atlantic salmon 0+ Rainbow trout 0+

– 327

2 weeks post salmon stocking 209 313

9 weeks post salmon stocking 150 145

Allopatric sites 1 week pre-salmon stocking – 243

2 weeks post salmon stocking – 310

9 weeks post salmon stocking – 97

Table 2 Number of habitat observations on sub-yearling Atlantic salmon and sub-yearling rainbow trout in allopatric and sympatric sites in Grout Brook, NY. Grout Brook

Atlantic salmon 0+ Rainbow trout 0+

Sympatric sites Lower Mill – – 166 348

Lower 202 220

Mill 128 363

Lower 143 152

Mill 68 254

Allopatric sites 1 week pre-salmon stocking – 225

2 weeks post salmon stocking – 215

9 weeks post salmon stocking – 133

118

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

densities were associated with wood cover and fines. Johnson and Chalupnicki (2014) examined the habitat use of yearling Atlantic salmon along with that of three species of Pacific salmonids in two Lake Ontario tributaries and concluded that the habitat used by Atlantic salmon was most similar to that used by sub-yearling Chinook salmon (Oncorhynchus tshawytscha). Although Hearn and Kynard (1986) examined the habitat use of sub-yearling Atlantic salmon and sub-yearling rainbow trout in

sympatry in Vermont, their studies were done in artificial stream channels that were stocked with salmon (hatchery) and trout (wild). Consequently, there have been no studies undertaken that have examined how the stocking of sub-yearling hatchery Atlantic salmon may influence the habitat use of sub-yearling rainbow trout in high quality natural stream environments. The objective of this study was to assess the effects of stocking Atlantic salmon on the habitat use of sub-yearling rainbow trout.

Fig. 2. Distribution of habitat variables (depth, velocity, substrate size, cover) used by sub-yearling rainbow trout (RBT = black bars) and sub-yearling Atlantic salmon (ATS = grey bars) shown in the bars at the Tubbs site on Orwell Brook. Distribution of available habitat shown in the connecting line. Left vertical panels = pre-stocking, middle vertical panels = 2 weeks post-stocking, right vertical panels = 9 weeks post-stocking.

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

119

Fig. 3. Distribution of habitat variables (depth, velocity, substrate size, cover) used by sub-yearling rainbow trout (RBT = black bars) at the Orwell site on Orwell Brook. Distribution of available habitat shown in the connecting lines. Left vertical panels = pre-stocking, middle vertical panels = 2 weeks post-stocking, right vertical panels = 9 weeks post-stocking.

Methods Two high quality streams were chosen to evaluate the effects of stocking Atlantic salmon on the habitat of sub-yearling rainbow trout. Orwell Brook is a tributary of the Salmon River which discharges into the southern shore of Lake Ontario in Oswego County, N.Y. The second Stream, Grout Brook, is a tributary of Skaneateles Lake (located in the Lake Ontario watershed), in Cortland County, N.Y. (Fig. 1). In Orwell Brook two representative 0.8 km sections (one allopatric—rainbow trout only, and one sympatric) were sampled. Three 0.8 km study sections were established in Grout Brook, two sympatric and one allopatric. The study sections were selected after examining about 10 km of each

stream. The study sections in each stream were approximately 3 km apart and in each stream the allopatric section was located upstream of the sympatric section(s). All five study sections had similar habitat characteristics including gradient, substrate composition, mean width, and depth, water temperature, and riparian vegetation. Generally, for each study reach stream gradients were 2.5–2.8%, stream width was 4.5 m–5.2 m, the substrate consisted of a mixture of gravel, cobble, and rubble with very little fine sediment; maximum summer water temperatures were 18–20 °C. Both Orwell Brook and Grout Brook support high densities of naturalized rainbow trout (Johnson and Douglass, 2009; Johnson et al., 2013). Although the rainbow trout in both streams are migratory,

120

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

Fig. 4. Distribution of habitat variables (depth, velocity, substrate size, cover) used by sub-yearling rainbow trout (RBT = black bars) and sub-yearling Atlantic salmon (ATS = grey bars) shown in the bars at the Lower Grout Brook site. Distribution of available habitat shown in the connecting line. Left vertical panels = pre-stocking, middle vertical panels = 2 weeks poststocking, right vertical panels = 9 weeks post-stocking.

there is no stream resident population; the fish in Orwell Brook are considered to be steelhead whereas those present in Grout Brook are considered rainbow trout. In both streams lake-run adults spawn in the spring with emergence of fry beginning in early June and peaking in mid-June. Because the size of the pectoral fin of juvenile Atlantic salmon (Arnold et al., 1991) and juvenile rainbow trout (Bisson et al., 1988) may play a role in competitive interactions between the species, the area of the pectoral fin was determined for 25 individuals of each species using a digital optical system. Juvenile salmonid habitat assessments were made on three occasions at each of the five sample sites. The first observation was made in mid-July, one week prior to stocking Atlantic salmon, and subsequently, sub-yearling rainbow trout habitat was examined at all five sites. Sub-yearling Atlantic salmon (West Grand Lake strain) were stocked at a density of 3 fish/m2 at the sympatric sites in each stream and averaged 66 mm, total length (TL). At this time rainbow trout averaged approximately 53 mm, TL. Additional habitat observations were made in early August (two weeks post Atlantic salmon stocking) and early October (9 weeks post salmon stocking).

Juvenile salmonid habitat use was examined using the spotelectrofishing technique (Bovee, 1986). Working stealthily upstream, a numbered, weighted buoy was placed at each site a juvenile salmonid was collected. The method is effective in waters that are too shallow (b12 cm mean depth) to snorkel when the distance between sample sites is sufficiently spaced (i.e. ≥ 3 m) to minimize fish disturbance (Heggenes et al., 1990). At each site a buoy was placed the number, species, and age group (rainbow trout) were recorded. Water depth, water velocity (measured at a depth of 60% from the water's surface), percent cover, and substrate size were recorded. Water depth was measured with a calibrated wading rod and water velocity with a Marsh–McBirney model 201d digital flow meter. The amount of cover and substrate size was estimated visually. Cover was recorded in 5% increments as total available cover within four fish lengths of the location of the buoy. Substrate size was estimated using a modified Wentworth particle size scale ranging from 1 (detritus) to 8 (bedrock) (Orth et al., 1981). Available habitat was determined at all study sites during each of the three

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

121

Fig. 5. Distribution of habitat variables (depth, velocity, substrate size, cover) used by sub-yearling rainbow trout (RBT = black bars) and sub-yearling Atlantic salmon (ATS = grey bars) shown in the bars at the Mill site on Grout Brook. Distribution of available habitat shown in the connecting line. Left vertical panels = pre-stocking, middle vertical panels = 2 weeks poststocking, right vertical panels = 9 weeks post-stocking.

periods when juvenile salmonid habitat was quantified. Water depth, water velocity, percent cover, and substrate size were recorded along 30 transects, spaced 25 m apart to establish available habitat within each stream reach during each sampling period. The distributions of salmonid habitat and available habitat variables were compared using a non-parametric Kruskal–Wallis one-way analysis of variance (Statistix 8.0, Tidepool Scientific, Tallahassee, FL). When differences were detected, Dunn's multiple comparison test was used to differentiate significant groups. Differences in the area of the pectoral fin between species were assessed using the analysis of covariance (Zar, 2010). Principal component analysis (PCA) was used to examine the ordination of salmonid habitat and available habitat variables (Canoco for

Windows 4.5, Wageningen, the Netherlands). An alpha level of P b 0.05 was used to detect significance. Results A total of 3831 observations were made on salmonid habitat including 2911 on sub-yearling rainbow trout and 920 on sub-yearling Atlantic salmon (Tables 1 and 2). Most of the observations (70%) were at the three sympatric sites with 30% occurring at the two allopatric sites. Sixty percent of the salmonid habitat observations were made in the three study sections in Grout Brook whereas 40% of the observations were made at the two sites in Orwell Brook.

122

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

Fig. 6. Distribution of habitat variables (depth, velocity, substrate size, cover) used by sub-yearling rainbow trout (RBT = black bars) shown in the bar at the Scott site on Grout Brook. Distribution of available habitat shown in the connecting lines. Left vertical panels = pre-stocking, middle vertical panel = 2 weeks post-stocking, right vertical panels = 9 weeks post-stocking.

Available habitat varied more over the three sample dates in Orwell Brook than in Grout Brook (Table 3, Figs. 2–6). Available substrate size was the habitat variable that varied the least in Orwell Brook. Consequently, because habitat conditions were more stable in Grout Brook, observed changes in the habitat use of sub-yearling rainbow trout at the sympatric sites may more likely be attributed to the presence of sub-yearling Atlantic salmon. At allopatric and sympatric sites in both streams rainbow trout almost always were associated with more cover than was generally available within the stream. In both allopatric and sympatric situations, rainbow trout generally occupied areas that were deeper and with faster flows than available habitat in both streams (Tables 3, 4). Sub-yearling Atlantic salmon were always associated with higher velocities (p ≤ 0.01) and more cover (p ≤ 0.01) than was generally available within the stream reach (Tables 3, 4, 5; Figs. 2, 4, 5). Salmon also occupied mean depths that were greater than available at the Orwell Brook Tubbs site (9 weeks post stocking) and the Grout Brook Mill site (p ≤ 0.01). There was no difference in mean depth occupied by rainbow trout and Atlantic salmon at 2 weeks post-stocking at any

sympatric site. At the lower Grout Brook site, which had the smallest substrate size of all five of the stream reaches sampled, sub-yearling Atlantic salmon were associated with larger substrate size than was available (all p b 0.01) (Tables 3, 4; Fig. 4) within the stream section. In sympatry, sub-yearling Atlantic salmon used faster velocity waters than sub-yearling rainbow trout (p ≤ 0.01) during each time period at all sites (Tables 3, 4, 5; Figs. 2, 4, 5). The mean depth occupied by rainbow trout was greater than used by Atlantic salmon in sympatry at the Orwell Brook Tubbs site and the Grout Brook Mill site (9 weeks post stocking) (p = 0.01) but not at the lower Grout Brook site. At two of the sympatric sites (i.e. Orwell Brook Tubbs, lower Grout Brook), Atlantic salmon were associated with larger substrate than rainbow trout (both p ≤ 0.01) (Tables 3, 4, 5; Figs. 2, 4). There was little difference in the amount of cover used by either species in sympatry. Whereas differences in available habitat among sampling periods in Orwell Brook may have masked changes in the habitat use of subyearling rainbow trout due to the presence of sub-yearling Atlantic salmon, more uniform habitat conditions in Grout Brook allowed for a more critical examination of fish habitat use among sampling periods.

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

123

Table 4 Dunn's T-test statistic for seasonal habitat use (depth, velocity, substrate index, percent cover) for sub-yearling rainbow trout (RBT) and sub-yearling Atlantic salmon (ATS) and available habitat (AH) for Grout Brook.

Grout Brook—Mill Site RBT Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) ATS Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) AH Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) RBT vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) ATS vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) RBT v ATS Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) Grout Brook—Lower Site RBT Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) ATS Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) AH Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) RBT vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) ATS vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) RBT v ATS Prestocking 1 week post stocking 9 weeks post stocking p

Depth (cm)

Velocity (cm/s)

Substrate index

Percent cover

2.07 2.47⁎ 4.53⁎ b0.01 10.7 (2)

2.31⁎ 3.32⁎ 1.01 b0.01 7.06 (2)

0.38⁎ 2.44⁎ 2.06 0.0266 3.81 (2)

8.51⁎ 0.56 7.95⁎ b0.01 61.9 (2)

– 0.24 – 0.8655 0.03 (1)

– 0.36 – 0.719 0.13 (1)

– 1.54 – 0.1159 2.53 (1)

– 0.77 – 0.43 0.62 (1)

0.81 0.6 1.41 0.3568 1.03 (2)

1.33 1.28 0.05 0.305 1.19 (2)

1.03 0.12 0.91 0.5023 0.69 (2)

0.07 0.35 0.28 0.9286 0.07 (2)

3.09 8.34⁎ 5.98⁎

0.88 0.06 0.15 0.7273 0.56 (5)

20.36 5.99⁎ 4.9⁎

b0.01 16.1 (5)

3.77 4.58⁎ 3.41 b0.01 4.99 (5)

– 5.05⁎ 4.78⁎

– 8.07⁎ 6.55⁎

– 6.34⁎ 0.03⁎

b0.01 13.6 (3)

0.047 20.1 (3)

– 0.89 0.52 0.126 1.91 (3)

– 0.18 3.26⁎ b0.01 4.89 (3)

– 0.054⁎ 0.053⁎ b0.01 11.4 (3)

– 1.06 1.17 0.06 2.51 (3)

– 0.27 1.13 0.08 2.18 (3)

b0.01 5.58 (2)

1.43 1.77 0.34 0.1401 2.0

0.98 4.6⁎ 5.58⁎ b0.01 18.3 (2)

3.70⁎ 1.45 5.15⁎ b0.01 17.0 (2)

– 1.50 – 0.8575 0.03 (1)

– 20.40⁎ – 0.0161 5.9 (1)

– 1.47 – 0.1223 2.41 (1)

– 6.73 – 0.3806 0.77 (1)

1.60 0.71 0.89 0.2678 1.03 (2)

4.43 5.56 9.98 0.8325 0.18 (2)

1.58 1.44 0.15 0.1894 1.67 (2)

1.40 12.1 1.96 0.5574 0.59 (2)

2.67 2.78 3.5⁎

4.10⁎ 4.90⁎ 5.53⁎

5.62⁎ 7.05⁎ 8.91⁎

0.0139 2.87 (5)

b0.01 11.0 (5)

3.53 4.46 0.19⁎ b0.01 5.98 (5)

– 0.53 0.03 0.6682 0.52 (3)

– 0.27⁎ 0.26⁎

– 7.09⁎ 5.17⁎

– 6.71⁎ 5.44⁎

b0.01 29.3 (3)

b0.01 15.3 (3)

b0.01 20.1 (3)

– 0.59 2.01 0.05

– 0.16⁎ 0.2⁎ b0.01

– 0.07⁎ 0.19⁎ b0.01

– 2.02 0.22⁎ b0.05

0.37 3.1⁎ 2.7⁎

b0.01 10.8 (5)

b0.01 21.8 (3)

b0.01 21.7 (5)

(continued on next page)

124

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

Table 4 (continued)

Chi-square (df) Grout Brook—Scott Site RBT Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) AH Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) RBT vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df)

Depth (cm)

Velocity (cm/s)

Substrate index

Percent cover

2.56 (3)

19.2 (3)

17.4 (3)

0.81 (3)

2.56⁎ 5.16⁎ 2.59⁎ b0.01 14.3 (2)

0.96 4.85⁎ 5.81⁎ b0.01 19.5 (2)

2.46⁎ 5.37⁎ 2.91⁎ b0.01 15.7 (2)

1.97 2.48⁎ 4.44⁎ b0.01 11.0 (2)

1.22 4.81 3.59 b0.01 12.9 (2)

2.5⁎ 2.26⁎ 0.24 0.0155 4.2 (2)

1.47 0.58 0.89 0.2796 1.28 (2)

0.002 1.47 1.48 0.1903 1.66 (2)

6.24 3.43⁎ 4.57⁎ b0.01 7.18 (5)

2.37⁎ 4.43⁎ 3.72⁎ b0.01 31.0 (5)

1.63⁎ 2.48⁎ 3.31 b0.01 6.62 (5)

3.34 4.17⁎ 6.65⁎ b0.01 23.5 (5)

⁎ Values significantly differ down a column (p b 0.05).

At the allopatric site (Scott) in Grout Brook, the water velocities used by sub-yearling rainbow trout increased over the sampling periods (p ≤ 0.01) (Tables 3, 4; Figs. 2, 6). Conversely, although there was no difference in water velocities used by sub-yearling rainbow trout between pre-stocking and two week post stocking periods, by week 9 water velocities used by rainbow trout were lower (p b 0.01) (Tables 3, 4; Figs. 4, 5) than during the pre-stocking period at the lower sympatric site on Grout Brook. During this same period water velocities occupied by sub-yearling Atlantic salmon increased (p = 0.002, t = 3.19, df = 131) (Table 3, Figs. 4, 5) between week 2 and week 9 at this sympatric site. Using principal component analysis (PCA), axis 1 explained 89.5% of the variation in habitat variables and axis 2 explained 7.7% (Fig. 7). Available habitat polygons between allopatric and sympatric sites overlapped substantially, suggesting that habitat was similar between allopatric and sympatric sites. Consequently, observed differences in the habitat use of sub-yearling rainbow trout between allopatric and sympatric sites are most likely due to the presence of Atlantic salmon and not differences in available habitat. Rainbow trout habitat and available habitat at the allopatric sites were similar, suggesting that little habitat selection was occurring. However, rainbow trout habitat at the sympatric sites was much more diverse, and based on the PCA vectors rainbow trout used far more cover and larger size substrate (compared to rainbow trout at the allopatric sites). Atlantic salmon habitat at the sympatric sites was most highly associated with high water velocities and deeper areas (Fig. 7). A linear regression model for pectoral fin area and fish length explained 59% of data variation for sub-yearling Atlantic salmon and 83% for sub-yearling rainbow trout (Fig. 8). Comparison of the elevation of the regression lines revealed that the area of the pectoral fin of Atlantic salmon was significantly larger (n = 60, p b 0.001, t = 2.01) than that of rainbow trout at all sizes. Furthermore, significant differences in the slope of the lines indicated that the size of the pectoral fin of subyearling Atlantic salmon became proportionally larger (n = 60, p b 0.001, t = 6.97) as fish grew. Discussion The ability of juvenile Atlantic salmon to compete with other naturalized non-native juvenile salmonid species in tributaries will play a large role in determining the success of restoration efforts in the Lake Ontario drainage. If Atlantic salmon are found to be inferior competitors, restoration efforts may be limited to areas that other salmonid species cannot access. This may require selective passage of adult Atlantic salmon, or stocking of juveniles into these limited habitats. As previously

mentioned, the non-native species that may offer the most competition for Atlantic salmon is rainbow trout, and several studies have been conducted on these species in sympatry in the Lake Ontario drainage examining survival and growth (Jones and Stanfield, 1993; Coghlan and Ringler, 2005), densities (Stanfield and Jones, 2003; Dietrich et al., 2008), and diet (Coghlan et al., 2007; Johnson and Waldt, 2014). This study is the first that examined habitat differences between the two species that was carried out in the Lake Ontario watershed. Van Zwol et al. (2012) reported a hierarchy of dominance among juvenile brown trout (Salmo trutta), Atlantic salmon, and rainbow trout with brown trout being the most dominant and Atlantic salmon the least dominant in artificially created stream channels. Conversely, Blanchet et al. (2006) found that juvenile rainbow trout significantly affected juvenile brown trout habitat selection and survival in a combined field and laboratory experiment in France. In a study done in experimental stream channels, Blanchet et al. (2008) found that interspecific competition between sub-yearling Atlantic salmon and rainbow trout had no effect on salmon feeding or growth. However, they also found that interspecific competition was greater during diurnal periods and postulated that this could expose Atlantic salmon to increased predation risk during the day. In sympatry, sub-yearling Atlantic salmon have been reported to use higher water velocities than rainbow trout (Hearn and Kynard, 1986) and brown trout (Heggenes et al., 2002). In Orwell Brook and Grout Brook sub-yearling Atlantic salmon occupied significantly faster water velocities than sub-yearling rainbow trout at the sympatric sites during both the two week and nine week post stocking assessment periods. In Grout Brook, an increase in the water velocities occupied by sub-yearling Atlantic salmon between week 2 and week 9 suggests that salmon were still acclimating to the stream environment and by week 9 they were more effectively using faster water velocities. At the allopatric site in Grout Brook, the water velocities occupied by sub-yearling rainbow trout were highest at the nine week assessment period whereas the velocities used by trout fry at the two sympatric sites declined. Diminished use of high velocity areas by subyearling rainbow trout in sympatry with Atlantic salmon is likely due to morphological adaptations of salmon to reside in fast water. The body shape and pectoral fin size of both juvenile Atlantic salmon (Arnold et al., 1991) and rainbow trout (Bisson et al., 1988) is thought to enhance each species' ability to maintain station in faster water. The area of the pectoral fin of sub-yearling Atlantic salmon in Grout Brook and Orwell Brook was approximately 78% larger than similar size sub-yearling rainbow trout and this likely contributed to salmon occupying faster velocities in sympatry with rainbow trout. Although water velocity was the main habitat variable that differed between sub-yearling Atlantic salmon and sub-yearling rainbow trout

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

125

Table 5 Dunn's T-test statistic for seasonal habitat use (depth, velocity, substrate index, percent cover) for subyearling Rainbow trout (RBT) and subyearling Atlantic salmon (ATS) and available habitat (AH) for Orwell Brook.

Orwell Brook-Tubbs RBT Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) ATS Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) AH Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) RBT vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) ATS vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) RBT v ATS Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df)

Orwell Brook-Orwell RBT Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) AH Prestocking vs. 1 week post stocking 1 week post stocking vs. 9 weeks post stocking Prestocking vs. 9 weeks post stocking p Chi-square (df) RBT vs. AH Prestocking 1 week post stocking 9 weeks post stocking p Chi-square (df) ⁎ Values significantly differ (p b 0.05).

Depth (cm)

Velocity Substrate Percent (cm/s) index cover

1.3 6.22⁎

7.39⁎ 10.6⁎

4.0 1.2

1.97⁎ 0.3

7.4⁎ b0.01 29.3 (2)

3.25⁎ b0.01 158 (2)

2.8 0.06 39.9

34.9⁎ b0.01 7.08 (2)

– 4.98⁎

– 9.94⁎

– 1.5

– 0.5

– b0.01 27.4 (2)

– b0.01 160 (1)

– – 0.2375 0.643 1.234 (1) 0.22 (1)

3.91⁎ 8.67⁎

2.11 11.2⁎

2.6⁎ 0.42

4.76⁎ b0.01 39.1 (2)

9.11⁎ 2.19 b0.01 0.0167 92.1 (2) 4.12 (2)

0.69 0.3744 0.98 (2)

4.12 9.25⁎ 2.11⁎ b0.01 21.4 (5)

9.26⁎ 15.54 17.28 b0.01 110 (5)

1.54⁎ 1.79⁎ 1.85 b0.01 5.99 (5)

6.43⁎ 7.43⁎ 7.99⁎ b0.01 34.3 (5)

– 5.65 6.89⁎ b0.01 25.0 (3)

– 20.88⁎ 10.1⁎ b0.01 140 (3)

– 0.48 0.15⁎ 0.03 2.99 (3)

– 9.77⁎ 0.82⁎ b0.01 53.4 (3)

– 4.96⁎ 6.21⁎ b0.01 19.8 (3)

– 24.53⁎ 13.74⁎ b0.01 229 (3)

– 0.49⁎ 0.2⁎ 0.01 3.53 (3)

– 0.95 0.01 0.67 0.51 (3)

7.33⁎ 16.9⁎

6.84⁎ 10.1⁎

2.69 0.68

2.04⁎ 4.75⁎

9.62⁎ b0.01 167 (2)

3.22⁎ 2.02 b0.01 0.07 98.3 (2) 3.83 (2)

2.71⁎ b0.01 14.1 (2)

4.95⁎ 15.3⁎

7.48⁎ 15.4⁎

3.11 0.12

3.69⁎ 5.97⁎

10.4⁎ b0.01 236 (2)

7.89⁎ b0.01 216 (2)

3.23 b0.01 7.24 (2)

2.28 b0.01 21.4 (2)

9.64⁎ 2.69 16.32 b0.01 85.5 (5)

7.69 7.31 22.6⁎ b0.01 131 (5)

3.24 1.77⁎ 0.05 b0.01 3.32 (5)

9.72⁎ 8.87⁎ 2.02⁎ b0.01 23.8 (5)

0.61 1.3

Fig. 7. Ordination of sub-yearling salmonid habitat and available habitat at allopatric and sympatric sites in Grout Brook and Orwell Brook, NY. OO is Orwell Brook, Orwell site, OT is Orwell Brook, Tubbs site, GL is Grout Brook, lower site, GM is Grout Brook, mill site, GS is Grout Brook, Scott site, A is Atlantic salmon, R is rainbow trout, AH is available habitat, Pre—is one week pre salmon stocking, PO2 is two weeks post salmon stocking and PO9 is nine weeks post salmon stocking. The colored polygon envelopes represent the outline of available habitat centroids at allopatric (blue) and sympatric sites (green), rainbow trout habitat at allopatric (red) and sympatric sites (black), and Atlantic salmon habitat at sympatric sites (orange).

in sympatry, there was also variation in depth and substrate size. Rainbow trout generally used deeper areas whereas Atlantic salmon often were associated with larger size substrate. The habitat variable that differed least between the two species was the amount of cover with which they were associated. The larger size of substrate used by Atlantic salmon compared to rainbow trout may suggest the use of substrate as cover (Rimmer et al., 1984) because salmon were larger than trout. The observations specific to water depth in this study contrast with those of Hearn and Kynard (1986) who reported that, in sympatry, sub-yearling Atlantic salmon occupied deeper areas than sub-yearling rainbow trout. Volpe et al. (2001) suggested that, because adult Atlantic salmon spawn in the fall and rainbow trout in the spring, that earlier emerging salmon fry should have a size advantage over trout fry during the first summer of stream residence. Because larger size often confers a competitive advantage (Young, 2004), Atlantic salmon may have a competitive advantage over rainbow trout as sub-yearlings in Lake Ontario tributaries. However, this possibility must be balanced with the findings of Van Zwol et al. (2012) who found that juvenile rainbow trout were more dominant than juvenile Atlantic salmon of similar size. Johnson et al. (2010) reported the first evidence of natural reproduction of

Fig. 8. Relationship between pectoral fin area and fish length for sub-yearling Atlantic salmon (ATS) and sub-yearling rainbow trout (RBT).

126

J.H. Johnson / Journal of Great Lakes Research 42 (2016) 116–126

Atlantic salmon in the Salmon River, NY in over a century. Based on the size of the wild sub-yearling Atlantic salmon that were collected in the Salmon River (personal observation, author) salmon may be about 10–15 mm larger than sub-yearling rainbow trout during their first year of residence in Lake Ontario tributaries. This is approximately the same size differential between the sub-yearling Atlantic salmon and rainbow trout in this study. Based on other studies (Noakes, 1980; Young, 2004), this size differential should offer a competitive advantage to sub-yearling Atlantic salmon over sub-yearling rainbow trout in Lake Ontario tributaries. When comparing the habitat use of sub-yearling rainbow trout at allopatric and sympatric sites, the data from Grout Brook suggest that the presence of Atlantic salmon influenced the habitat use of trout. Specifically, when salmon were present, rainbow trout shifted to slower and shallower areas. However, this apparent habitat shift did not occur in Orwell Brook. Moreover, when comparing rainbow trout habitat at all allopatric and sympatric sites PCA suggested that cover and substrate size were the habitat variables most affected by the presence of Atlantic salmon. Although these findings are consistent with previous studies (Hearn and Kynard, 1986) on the habitat use of these species in sympatry, the shift in the habitat use of rainbow trout due to the addition of Atlantic salmon had not been reported previously. As mentioned previously, the body morphology of Atlantic salmon may also confer an additional competitive advantage over trout in fast water. This study demonstrated that, based on the likely size differential between the two species during their first six months of stream residence, that Atlantic salmon will influence the habitat use of rainbow trout. Collectively, this study along with the previous studies on growth, survival, and diet of these species in sympatry in the Lake Ontario drainage is an essential first step towards restoring Atlantic salmon. Habitat based studies such as this one are important because of the number of non-native salmonid species that are now naturalized in Lake Ontario tributaries and speculation that the habitat in many of these streams may be more suitable for the non-natives than for Atlantic salmon (Stoneman and Jones, 2000; Dietrich et al., 2008). Until Atlantic salmon establish successful spawning populations in Lake Ontario tributaries, studies examining interspecific interactions in streams will require the use of hatchery reared salmon. When examining interspecific interactions between Atlantic salmon and rainbow trout during their first year of stream residence, special attention should be given to the size of fish used in order to reflect the most likely biological interactions. Acknowledgments I thank Jean Adams and James McKenna for their helpful comments and Marc Chalupnicki for assistance with data analysis. This article is contribution 1958 of the U. S. Geological Survey Great Lakes Science Center. Any use of trade, product, or firm names is for descriptive purposes only and does not imply endorsement by the U. S. Government. References Arnold, G.P., Webb, P.W., Holford, B.H., 1991. The role of the pectoral fins in station holding of Atlantic salmon parr (Salmon salar). J. Exp. Biol. 156, 625–629. Bisson, P.A., Sullivan, K., Nielsen, J.L., 1988. Channel hydraulics, habitat use, and body form of juvenile coho salmon, steelhead, and cutthroat trout in streams. Trans. Am. Fish. Soc. 117, 262–273. Blanchet, S., Loot, G., Grenouillet, G., Brosse, S., 2006. Competitive interactions between native and exotic salmonids: a combined field and laboratory demonstration. Ecol. Freshw. Fish 115, 1–11. Blanchet, S., Páez, D.J., Bernatchez, L., Dodson, J.J., 2008. An integrated comparison of captive-bred and wild Atlantic salmon (Salmo salar): implications for supportive breeding programs. Biol. Conserv. 141 (8), 1989–1999.

Bovee, K.D., 1986. Development and evaluation of habitat suitability criteria for use in the instream flow incremental methodology. Washington, DC: U.S. Fish and Wildlife Service. Instream Flow Information Paper No. 21. Biol. Rep. 86 (7), 235. Christie, G.C., Goddard, C.I., 2003. Sea Lamprey International Symposium (SLIS II): Advances in the Integrated Management of Sea Lamprey in the Great Lakes. J. Great Lakes Res. 29 (Supplement 1), 1–14. Coghlan Jr., S.M., Ringler, N.H., 2005. Temperature-dependent effects of rainbow trout on growth of Atlantic salmon parr. J. Great Lakes Res. 31, 386–396. Coghlan Jr., S.M., Connerton, M.J., Ringler, N.H., Stewart, D.J., Mead, J.V., 2007. Survival and growth responses of juvenile salmonines stocked in Eastern Lake Ontario tributaries. Trans. Am. Fish. Soc. 136, 56–71. Dietrich, J.P., Bowlby, J.N., Morrison, B.J., Jones, N.E., 2008. The impacts of Atlantic salmon on stocking rainbow trout in Barnum House Creek, Lake Ontario. J. Great Lakes Res. 34, 495–505. Hearn, W.E., Kynard, B.K., 1986. Habitat utilization and behavioral interaction of juvenile Atlantic salmon (Salmo salar) and rainbow trout (S. gairdneri) in tributaries of the White River of Vermont. Can. J. Fish. Aquat. Sci. 43, 1988–1998. Heggenes, J., Brabrand, A., Saltveit, S.J., 1990. Comparison of three methods for studies of stream habitat use by young brown trout and Atlantic salmon. Trans. Am. Fish. Soc. 119, 101–111. Heggenes, J., Saltveit, S.J., Bird, D., Grew, R., 2002. Static habitat partitioning and dynamic selection by sympatric young Atlantic Salmon and Brown Trout in south-west England streams. J. Fish Biol. 60, 72–86. Huntsman, A.G., 1944. Why did Ontario salmon disappear? Trans. R. Soc. Can. Third Ser. 5, 83–102. Johnson, J.H., Chalupnicki, M.A., 2014. Interspecific habitat associations of juvenile salmonids in lake Ontario tributaries: implications for Atlantic salmon restoration. J. Appl. Ichthyol. 30, 853–861. Johnson, J.H., Douglass, K.A., 2009. Diurnal stream habitat use of juvenile Atlantic salmon, brown trout and rainbow trout in winter. Fish. Manag. Ecol. 16, 352–359. Johnson, J.H., Ringler, N.H., 1981. Natural reproduction and juvenile ecology of Pacific salmon and rainbow trout in tributaries of the Salmon River, New York. N.Y. Fish Game J. 28, 49–60. Johnson, J.H., Waldt, E.M., 2014. Examination of the influence of juvenile Atlantic salmon on the feeding mode of juvenile steelhead in lake Ontario tributaries. J. Great Lakes Res. 40, 370–376. Johnson, J.H., Wedge, L.R., 1999. Intraspecific competition in tributaries: prospectus for restoring Atlantic salmon in Lake Ontario. Great Lakes Res Rev. 4 (2), 11–17. Johnson, J.H., Nack, C.C., McKenna Jr., J.E., 2010. Migratory salmonid redd habitat characteristics in the Salmon River, New York. J. Great Lakes Res. 36, 387–392. Johnson, J.H., McKenna Jr., J.E., Douglass, K.A., 2013. Movement and feeding ecology of recently emerged steelhead in Lake Ontario tributaries. J. Appl. Ichthyol. 29, 222–225. Jones, M.L., Stanfield, L.W., 1993. Effects of exotic juvenile salmonines on growth and survival of juvenile Atlantic salmon (Salmo salar) in a Lake Ontario tributary. In: Gibson, R.J., Cutting, R.E. (Eds.), Production of Juvenile Atlantic Salmon, Salmo salar, in Natural Waters. Canadian Special Publication of Fisheries and Aquatic Sciences 118. Ontario, Canada. MacCrimmon, H.R., Gots, B.L., 1972. Rainbow Trout in the Great Lakes. Ontario Ministry of Natural Resources, Toronto, Ontario, p. 66. Noakes, D.L.G., 1980. Social behavior in young charrs. In: Balon, E.K. (Ed.), Charrs: Salmonid Fishes of the Genus Salvelinus. Dr. W. Junk, The Hague, pp. 683–701. Orth, D.L., Jones, R.N., Maughan, O.E., 1981. Consideration in the development of habitat suitability criteria. Pp. 124–133. In: Armantrout, N.B. (Ed.), Acquisition and Utilization of Aquatic Habitat Inventory Information. Western Division. American Fisheries Society, Portland, OR (273 pp.). Parrish, D.L., Behnke, R.J., Gephard, S.R., McCormick, S.D., Reeves, G.H., 1998. Why aren't there more Atlantic salmon (Salmo salar)? Can. J. Fish. Aquat. Sci. 55 (S1), 281–287. Parsons, J.W., 1973. History of salmon in the Great Lakes, 1850–1970. U. S. Bur. Sport Fish. Wildlife, Tech. Paper, No. 68. Rimmer, D.M., Paim, U., Saunders, R.L., 1984. Changes in the selection of microhabitat by juvenile Atlantic salmon (Salmo salar) at the summer–autumn transition in a small river. Can. J. Fish. Aquat. Sci. 41, 469–475. Stanfield, L., Jones, M.L., 2003. Factors influencing rearing success of Atlantic salmon stocked as fry and parr in a Lake Ontario tributary. N. A. J. Fish. Manag. 23, 1175–1183. Stoneman, C.L., Jones, M.L., 2000. The influence of habitat features on the biomass and distribution of three species of southern Ontario stream salmonines. Trans. Am. Fish. Soc. 129 (3), 639–657. Van Zwol, J.A., Neff, B.D., Wilson, C.C., 2012. The effect of competition among three salmonids on dominance and growth during the juvenile stage. Ecol. Freshw. Fish 21, 533–540. Volpe, J.P., Anholt, B.R., Glickman, B.W., 2001. Competition among juvenile Atlantic salmon (Salmo salar) and steelhead (Oncorhynchus mykiss): relevance to invasion potential in British Columbia. Can. J. Fish. Aquat. Sci. 58, 197–207. Webster, D.A., 1982. Early history of the Atlantic salmon in New York, N. Y. Fish Game J. 29, 26–44. Young, K.A., 2004. Asymmetric competition, habitat selection, and niche overlap in juvenile salmonids. Ecology 85, 137–149. Zar, J.H., 2010. Biostatistical Analysis. 5th edition. Pearson, Prentice Hall, Upper Saddle River, New Jersey.