Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 545 2018 NORTHEASTERN NATURALIST 25(4):545–570 Finfish Diversity and Distribution in a Bor eal, Macrotidal Bay Jeffrey D. Vieser1, Gayle Barbin Zydlewski1,*, and James D. McCleave1 Abstract - Cobscook Bay is an 11- km2 geographically complex, boreal, and macrotidal bay in eastern Maine. The physical environment, primary producers, and invertebrate assemblage of the bay are well-characterized, but no contemporary data exist on its finfish assemblage. We sampled the finfish assemblage of Cobscook Bay from 2011 to 2013 in May, June, August, and September to create a baseline dataset suitable for future comparisons. We also examined the composition, diversity, and annual changes in the assemblage. We sampled in the subtidal and intertidal zones using seines (n = 390), fyke nets (n = 72), and benthic (n = 112) and pelagic (n = 111) trawls; sampling was divided among the bay’s 3 different sub-bays. We collected more than 60,000 individuals from 46 species. We employed species richness, Simpson’s index of diversity, and non-metric multidimensional scaling (with the Bray–Curtis and Horn–Morisita indices) to examine spatial and temporal variation of finfish assemblages throughout the bay. Our analysis suggested that data collected in the subtidal pelagic were not a representative sample of that assemblage. Therefore, we considered 2 assemblages: the intertidal and subtidal benthic. Assemblage composition and species’ relative abundances were different at diel, monthly, and annual timescales and were associated with changes in the catch rate of ubiquitous species. In the intertidal, these species included Gasterosteus aculeatus (Threespine Stickleback), Clupea harengus (Atlantic Herring), Alosa pseudoharengus (Alewife), and Menidia menidia (Atlantic Silverside). In the subtidal, the common species were Atlantic Herring and Pseudopleuronectes americanus (Winter Flounder). Statistical analyses indicated that both spatial and temporal factors were significant predictors of assemblage evenness. The sampling design, albeit complex, was sufficient to capture these differences and characterize these assemblages. Implications for future studies are that the study design must be sufficiently complex to capture the anticipated spatial and temporal variability inherent in such dynamic environments. Furthermore, given recent warming trends in the Gulf of Maine, this study’s results suggest the importance of thoroughly understanding local temporal and ecosystem variability. Introduction Of all types of estuaries and embayments of the North Atlantic Ocean, macrotidal systems are the least understood. Their high energy creates numerous challenges for scientists (Shields et al. 2011) and complicates standard sampling practices such as finfish netting, though studies do exist (MacDonald et al. 1984, Tyler 1971). Lower-energy bays are regularly assessed for seasonal patterns and changes in finfish-assemblage diversity, composition, and density over time by documenting species presence and abundance and then applying diversity indices and ordination techniques (Alemany et al. 2009, Greenstreet and Hall 1996, 1School of Marine Sciences, University of Maine, Orono, ME 04469. *Corresponding author - gayle.zydlewski@maine.edu. Manuscript Editor: Joseph Rachlin Northeastern Naturalist 546 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 Jouffre and Inejih 2005, Jung and Houde 2003, Pyle 2012, Sherman et al. 2012). These data can subsequently serve as baselines for future comparisons. The applicability of such assessment techniques to macrotidal environments has not been thoroughly explored but would be timely in light of changing anthropogenic influences in such regions (Melvin and Cochrane 2012, Shields et al. 2011, Viehman et al. 2015). Diversity metrics, such as Simpson’s index of diversity and non-metric multidimensional scaling (NMDS), generally do not provide inherent measures of confidence. This shortcoming has been rectified using multiple techniques and internal comparisons of results. An examples of this approach is the use of species richness, the Shannon–Weiner diversity index, abundance–biomass comparison curves, and correspondence analysis to describe the Chesapeake Bay finfish assemblage (Jung and Houde 2003). From 2011 to 2013, we conducted surveys of Cobscook Bay, a boreal, macrotidal bay in the northeastern US to (1) characterize Cobscook Bay’s finfish assemblage, including both the subtidal and intertidal components; (2) analyze spatial and temporal variation of the assemblages within the complex bay; and (3) evaluate the utility of standard diversity measures in a high-energy environment. Successful application of this approach in Cobscook Bay would enable a baseline understanding of that system and would demonstrate the utility of standard-sampling practices in a macrotidal system. Subsequent examinations of results could then be used to make meaningful observations about assemblage diversity in macrotidal environments, e.g., when evaluating potential environmental effects of tidal-energy extraction (Shields et al. 2009, Viehman et al. 2015). Important in this approach are the appropriate scope and scale at which meaningful observations are necessary for suitable understanding of the system in question. Field-site Description The surface area of Cobscook Bay (Fig. 1) at high water is about 111 km2 (Larsen et al. 2004) with an average tidal volume of 0.54 km3 (Brooks 2004) which, when flowing through the bay’s intricate 3-dimensional structure, results in tidal currents that reach ~2 m s-1 in some locations (Brooks 1992, 2004). Cobscook Bay was the focus of a series of physical, chemical, and biological studies in the late 1990s and early 2000s (Larsen 2004), but the bay’s finfish assemblage was not studied. Materials and Methods We distributed our sampling effort among Cobscook Bay’s 3 sub-bays (Inner, Central, and Outer; Fig. 1). For our analysis, we subdivided Inner Bay into Dennys Bay to the north and Whiting Bay to the south, and Central Bay into Pennamaquan River (bay) and East Bay to the north and South Bay to the south. We did not subdivide Outer Bay. Sampling occurred from 2011 through 2013 in May, June, August, and September because scientists and local fishermen hypothesized that finfish would be most abundant then (Collette and Klein-MacPhee 2002, MacDonald et al. Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 547 1984). Samples were centered on neap tides to maximize available sampling time and standardize the areas sampled. Intertidal sampling Intertidal-site selection was not random due to a scarcity of suitable sampling locations. No site was accessible in the southern region of Outer Bay. At the request of local fishers, we sampled 1 additional site at the mouth of the Pennamaquan River (the northwestern-most site in Central Bay, Fig. 1). We used beach seines and fyke nets in the intertidal zone. The seine was 30.5 m long by 1.8 m tall with a bag at its center. The mesh was a diamond pattern with 3.2-mm mesh. Fyke nets were 1.2 m in height and had one wing 9.1-m in length and one 18.3-m in length. Mesh size was 2.5 cm throughout the net, wings, and cod end. The cod end consisted of 3 sections, each separated by a tapered throat with an opening of 30.5-cm in diameter. Figure. 1. Map of finfish sampling locations around Cobscook Bay, ME, 2011–2013 (NOAA National Geophysical Data Center 2011). The sub-bays from left to right are Inner, Central, and Outer Bays. Northeastern Naturalist 548 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 We conducted all seine sampling on the ebbing tide to keep the areas sampled consistent. Seining was done from shore by wading. When possible, we made multiple tows at each site based upon the amount of suitable habitat, the slope of the site, and the amount of time spent handling fish. There were 3 occasions in which weather conditions or large catches limited sampling to 1 tow at a site. We sampled a diversity of habitats, including mudflat, cobble, Ascophyllum nodosum (L.) Le- Jolis, (Rockweed), or Spartina alterniflora Loisel (Saltmarsh Cordgrass, hereafter termed Cordgrass). We calculated catch per unit effort (CPUE) as individuals per 2 min of sampling to standardize the data with that reported by other seine tows and subtidal samples (trawls). In 2012 and 2013, we used fyke nets to target larger individuals in the northern site in each sub-bay, excluding the Pennamaquan River site (Fig. 1). We manually deployed 2 nets during low tide to sample from mid-flood to mid-ebb tide. The cod ends were set parallel to the shore and the wings of each net were set angled away from the cod end in a ‘Y’ shape. We included data collected with fyke nets only in the analysis of overall species composition due to low catch and variable effort. Subtidal sampling We fished benthic and pelagic trawls from the F/V Pandalus in the subtidal zone of each of the 3 sub-bays (Fig. 1). The benthic-net dimensions were a 13.7-m headrope, 10.7-m footrope, and no breastlines. The main body of the net had square mesh with openings of 5.1 cm, which decreased to 2.5 cm in the cod end. The pelagic-net had a 12.2-m footrope, 12.2-m headrope, and a 12.2-m breastline; mesh that began as 10.2-cm squares, with side panels that tapered to 5.1-cm squares; and a cod end extension that had a mesh consisting of 2.5-cm squares. Most trawls were ~20 min in duration. Trawls in Inner Bay were geographically limited and were occasionally reduced to as little as 13 min. We calculated CPUE as number of individuals caught per 10 min of sampling. The trawls were centered on high tide in 2011 and 2013, and on low tide in 2012. We employed the GPS on board the F/V Pandalus to record the start and end positions of benthic and pelagic trawls. To maximize bottom area sampled, we performed benthic trawls on either side (usually ±30 min) of slack tide (estimated using the closest NOAA-predicted site). The captain followed similar transects when benthic trawling to keep geographic distance traveled similar among trawls (Fig. 1). Pelagic trawls were performed immediately preceding or following benthic trawls based upon the tidal stage. Tidal stage and variability in water velocity resulted in changes in geographic distance covered during pelagic trawls. The captain adjusted the boat’s engine speed to maintain similar sampling volumes regardless of the distance traveled while fishing. Nighttime sampling We conducted nighttime sampling in the intertidal and subtidal to investigate diel differences in the finfish assemblages throughout the bay (Table 1). We combined these nighttime data with daytime data in analyses of the assemblage and its spatial and temporal variability. We conducted separate analyses to examine diel Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 549 differences in species richness and species’ relative abundance for Cobscook Bay for the replicate samples. Fish handling We immediately placed captured fishes into a holding container. We identified individuals to the species level and used MS222 to anesthetize up to 30 individuals from each species to be weighed (to 0.01 g) and measured (to nearest 1 mm) in a laboratory setting. Additional individuals were counted and released. We weighed (to nearest 10 g), measured (mm), and released larger individuals in the field. We anesthetized and later identified, using keys from Collette and Klein-MacPhee (2002) and descriptions in Scott and Scott (1988), any fishes that we could not identify in the field. Diversity indices We documented finfish diversity in Cobscook Bay using species richness and Simpson’s index of diversity, as defined by the equation: Simpsons index of diversity = 1 - Σ(n[n - 1] / N[N - 1]), where n = total number of individuals of a species, and N = total number of individuals of all species. Values can vary between 1 (highly diverse) and 0 (no diversity) (Berger and Parker 1970, Simpson 1949). Species richness is independent of abundance (Colwell 2009). Simpson’s index of diversity is a measure of the evenness of species’ relative abundances within an assemblage (Simspson 1949); thus, effort must be incorporated. The samples collected by each gear (benthic trawl, pelagic trawl, and seine) were analyzed independently for Simpson’s index due to differences in catchability. We made all calculations and statistical comparisons in R, version 3.0.2 (R Core Team 2015). Spatial scales used to partition these data for analyses included sub-bay (Inner, Central, and Outer), tidal zone (intertidal and subtidal), habitat (e.g., Cordgrass and mudflat), and individual sampling locations. We further partitioned subtidal data by benthic and pelagic catch. Temporal scales considered were years (2011, 2012, and 2013), months (May, June, August, and September), and diel cycle (day and night). Table 1. Night finfish-samples taken in Cobscook Bay, ME, 2011–2013. T = trawl (benthic and pelagic), F = fyke net, and S = seine. We conducted 2 sets of trawls in Outer Bay. Subtidal night trawls were not done in Inner Bay for safety reasons. No night sampling was done in Whiting Bay or the Pennamaquan River. Sampling was consistent throughout all months in 2012 and 201 3. Inner Bay Central Bay Year Month (Denny’s) East South Outer Bay 2011 May T, S, F Jun T, S, F T Aug F T, F T S Sep F T, S, F T F 2012 May, Jun, Aug, Sep S, F T, S, F T T, S, F 2013 May, Jun, Aug, Sep S, F T, S, F T T, S, F Northeastern Naturalist 550 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 There is not an inherent way to estimate confidence for either species richness or Simpson’s indices, making it difficult to draw useful conclusions from them. To estimate confidence for observations of species richness, we employed the abundance-based coverage estimator (ACE). ACE estimates the number of species in an environment based upon the number of times they were observed in a data set, generates confidence intervals around estimates, and requires few samples (n ≥ 5) (Chao 1987, Chao and Lee 1992). We also employed rarefaction curves, generated by repeatedly re-sampling the data (n = 100) without replacement, and plotting the number of species represented per number of individuals sampled, to evaluate species richness values. These curves have 2 distinct regions: 1 of rapid increase, and 1 approaching an asymptote. We visually inspected curves to determine if an asymptote was approached, which would indicate that a sampling area was exhaustively sampled (Gotelli and Colwell 2001). For Simpson’s index, we followed a multi-step process to analyze these data and compare species composition over spatial and temporal gradients. The first step was an evaluation of normality conducted using the Kolmogorov–Smirnov (KS) and Shapiro–Wilk (SW) tests in the R package “stats” under sample size constraints of 0, 1, 5, 10, 20, and 30 individuals (R Core Team 2015). Both tests indicated that our data were non-normal. For each gear, we used the dataset identified by the KS and SW tests as closest to normal (minimum sample size of 10 individuals) for all density-dependent analyses. We employed permutation ANOVA to test the predictive power of the different factors being considered and for the presence of significant interactions. This approach performs well in situations where sample sizes are small (Anderson 2001). Three sets of factors were considered for the intertidal and subtidal data. The first was year (only 2012 and 2013 because so few night samples were taken in 2011), month, and diel; the second was year, month, and sub-bay; and the third was year and month. For each set of factors, we ran permutation ANOVA that allowed for interactions and immediately eliminated factors without predictive power (P ≥ 0.1). We then reran the model and eliminated all remaining non-significant (P > 0.05) factors until only factors with significant predictive power (P ≤ 0.05) remained. We also employed the Akaike information criterion (AIC) to compare models and ensure that we selected the optimal model. We conducted post-hoc analysis using Tukey’s “honest significant difference” method to investigate the within-group effects of all significant group-level factors. Post-hoc analysis of significant interaction terms was conducted with linear regression (R Core Team 2015). We graphically compared assemblage composition with NMDS (McCune et al. 2002), with respect to space (sub-bay, tidal zone, sampling site, and habitat), sampling gear, and time (year, month, season, and diel cycle). We used 2 separate indices to optimize the ordinations: the Bray–Curtis dissimilarity index, which measures difference in species composition between 2 sites, and the Horn–Morisita overlap index, which measures proportional species overlap (Horn 1966, Morisita 1959). We ran ordinations using the metaMDS function (100 random starts) in the statistical package “vegan” in R (Oksanen et al . 2015). Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 551 We ran ordinations for both indices in 2 and 3 dimensions, and considered as optimal a stress value of 0.1, based on McCune et al. (2002). Stress refers to the difference in the distance between points in multidimensional space compared to the difference in 2- or 3-dimensional ordination space. The test preferred 2-dimensional iterations when stress was less than 0.1; otherwise we used the ordination that produced the lowest stress (McCune et al. 2002). We tested ordinations for significant differences among grouping parameters (e.g., month or sub-bay) by examining the standard deviation and center of within-group points and plotting them along best-fit x- and y-axes. We plotted the resulting ellipses, and if no overlap occurred, considered assemblages to be significantly different (Oksanen et al. 2015). Results During this study, we captured 60,476 individual fish from 46 species. Scientific names and taxonomic authors of all fishes caught in this study are given in Table 2. Threespine Stickleback (n = 19,126), Atlantic Herring (n = 18,086), and Winter Flounder (n = 8302) were the 3 most abundant species, comprising 31.7%, 30.0%, and 13.8% of the total catch, respectively. Overall, we conducted 390 capture events with seines, 112 with benthic trawls, 111 with pelagic trawls, and 72 with fyke nets, respectively. All gear types captured fish, and the most abundant species differed for each tidal zone sampled by the 4 gear types. In the intertidal zone, the most abundant species were Threespine Stickleback (n = 18,879) for seine sampling and Microgadus tomcod (Atlantic Tomcod; n = 121) for fyke sampling. Winter Flounder (n = 8291) dominated the subtidal benthic sampling , and Atlantic Herring (n = 10,907) was the most frequently caught species in subtidal pelagic sampling. We captured 32 species in 2011, 36 in 2012, and 40 in 2013. Half of the species were rare, i.e., fewer than 40 individuals captured in all 3 y combined (Table 2). Calculated rarefaction curves approached asymptotes for seine (Fig. 2) and subtidal benthic sampling (Fig. 3). Rarefaction curves calculated from subtidal pelagic samples did not approach clear asymptotes (Fig. 3). Observed species richness fell within the 95% confidence intervals of the ACE-1 estimates for species richness, including those in the subtidal pelagic zone (Table 3). In the intertidal zone, an average seine tow (2 min) captured 64.9 (min–max = 0–3822) individuals. Threespine Stickleback accounted for 60.9% of the catch for all years. In the subtidal zone, 10 min of benthic trawling captured between 0 and 2614 individuals, averaging 77.7 individuals, and 10 min of pelagic trawling captured between 0 and 1252 individuals, averaging 50.9 individuals. The total subtidal catch was primarily split between 2 species: Atlantic Herring (59.6%) and Winter Flounder (27.1%). In the benthic samples, Winter Flounder (36.8%) and Atlantic Herring (34.6%) were the most frequently caught species. The pelagic samples were dominated by Atlantic Herring, which comprised 97.2% of sampled individuals. Assemblages captured in the intertidal (seines) were significantly different from those captured in the subtidal zone (benthic and pelagic trawls) (Fig. 4a). Only 1 species, Atlantic Herring, was found to be abundant in both intertidal and subtidal habitats. Subtidal benthic and pelagic assemblages also differed significantly (Fig. 4b). Northeastern Naturalist 552 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 Table 2. Catch of all fish species captured in Cobscook Bay, ME, 2011–2013, by month, year, and total. The gear column shows all gears capturing each species: S = seine, B = benthic trawl, P = pelagic trawl, and F = fyke net. For each species, gear type marked with an asterisk (*) indicates that it captured 33% or more of the total catch. [Table continued on following page.] 2011 2012 2013 Species May Jun Aug Sep May Jun Aug Sep May Jun Aug Sep Total Gear Gasterosteus aculeatus L. (Threespine Stickleback) 233 429 614 308 895 904 8469 4803 109 1848 157 357 19,126 B,P,S* Clupea harengus L. (Atlantic Herring) 225 820 545 24 2561 1230 0 3 52 6153 1056 5417 18,086 B*,P*,S,F Pseudopleuronectes americanus Walbaum (Winter 156 251 461 271 772 892 130 163 2443 1600 798 365 8302 B*,P,S,F Flounder) Alosa pseudoharengus (Wilson) (Alewife) 0 1 1 13 0 735 289 88 5 0 1642 573 3347 B,P,S*,F Menidia menidia (L.) (Atlantic Silverside) 3 10 18 25 70 66 37 1814 79 26 17 536 2701 S* Fundulus heteroclitus (L.) (Mummichog) 29 25 148 69 197 196 133 301 133 394 371 471 2467 S*,F Gasterosteus wheatlandi Putnam (Blackspotted 78 109 69 11 222 237 716 562 62 143 153 74 2436 P,S* Stickleback) Merluccius bilinearis (Mitchill) (Silver Hake) 1 7 8 0 32 216 8 2 5 59 207 47 592 B*,P Myoxocephalus octodecemspinosus (Mitchill) 27 13 21 1 87 86 4 6 140 123 61 9 578 B* (Longhorn Sculpin) Osmerus mordax (Mitchill) (Rainbow Smelt) 238 13 12 25 31 118 9 18 9 55 17 22 567 B*,P,S*,F Myoxocephalus aenaeus (Mitchill) (Grubby) 46 10 3 3 46 54 6 8 138 66 36 29 445 B*,P Urophycis tenuis (Mitchill) (White Hake) 0 2 20 28 0 5 21 28 0 0 121 203 428 B*,P Melanogrammus aeglefinus (L.) (Haddock) 0 0 40 6 0 0 0 0 0 0 208 135 389 B* Microgadus tomcod (Walbaum) (Atlantic Tomcod) 0 17 24 28 2 16 25 22 3 26 65 46 274 B,S*,F* Urophycis chuss (Walbaum) (Red Hake) 0 2 2 4 11 7 2 11 23 52 84 15 213 B*,S Apeltes quadracus (Mitchill) (Fourspine Stickleback) 0 1 21 33 0 0 32 10 2 0 0 0 99 S* Peprilus triacanthus (Peck) (Butterfish) 0 0 7 5 0 1 50 11 0 0 1 8 83 B*,P Hemitripterus americanus (Gmelin) (Sea Raven) 12 1 6 3 8 6 1 0 0 3 1 2 43 B* Pollachius virens (L.) (Pollock) 0 10 12 3 0 0 0 5 0 2 5 6 43 B,S*,F* Pungitius pungitius (L.) (Ninespine Stickleback) 0 1 1 0 0 0 12 3 0 0 5 20 42 S* Lumpenus lampretaeformis (Walbaum) (Snakeblenny) 0 4 3 0 15 6 0 0 5 0 2 0 35 B* Gadus morhua L. (Atlantic Cod) 4 9 2 3 6 4 0 0 2 0 0 2 32 B* Scophthalmus aquosus (Mitchill) (Windowpane) 0 0 0 0 0 1 0 0 13 9 3 0 26 B*,P Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 553 Table 2. , continued. 2011 2012 2013 Species May Jun Aug Sep May Jun Aug Sep May Jun Aug Sep Total Gear Myoxocephalus scorpius (L.) (Shorthorn Sculpin) 3 4 1 1 0 3 0 0 0 4 2 0 18 B*,P Hippoglossus hippoglossus (L.) (Atlantic Halibut) 0 0 0 1 2 6 1 1 0 0 3 2 16 B* Scomber scombrus L. (Atlantic Mackerel) 0 0 10 0 0 0 3 2 0 0 1 0 16 B*,P* Alosa aestivalis (Mitchill) (Blueback Herring) 0 0 0 1 0 2 2 2 2 0 0 1 10 B,S*,F Leucoraja ocellate (Mitchill) (Winter Skate) 2 0 2 0 0 2 0 0 1 1 0 1 9 B* Ulvaria subbifurcata (Storer) (Radiated Shanny) 0 1 1 0 0 1 1 1 2 1 0 0 8 B* Cyclopterus lumpus L. (Lumpfish) 0 2 1 0 0 1 0 0 0 1 1 2 8 B*,P* Malacoraja senta (Garman) (Smooth Skate) 0 0 0 0 2 0 0 0 0 0 3 0 5 B* Liopsetta putnami (Gill) (Smooth Flounder) 0 0 0 0 0 0 0 0 0 0 4 1 5 S* Zoarces americanus (Bloch and Schneider) (Ocean 0 0 0 0 0 0 0 0 0 2 2 0 4 B* Pout) Pholis gunnellus (L.) (Rock Gunnel) 0 1 1 0 0 1 0 0 0 0 0 0 3 B* Leucoraja erinacea (Mitchill) (Little Skate) 0 0 0 1 0 1 0 0 0 0 0 1 3 B* Squalus acanthias L. (Spiny Dogfish) 0 0 0 0 0 0 3 0 0 0 0 0 3 B* Liparis atlanticus (Jordan and Evermann) (Atlantic 0 0 0 0 0 0 0 0 2 1 0 0 3 B*,P* Seasnail) Hippoglossoides platessoides (Fabricius) (American 0 0 0 0 0 0 0 0 1 1 0 0 2 B* Plaice) Triglops murrayi Günther (Moustache Sculpin) 0 0 0 0 0 0 0 0 0 2 0 0 2 B* Enchelyopus cimbrius (L) (Fourbeard Rockling) 0 0 0 0 1 0 0 0 0 0 0 0 1 B* Raja eglanteria Bosc in Lacepède (Clearnose Skate) 0 0 0 0 0 0 0 1 0 0 0 0 1 B* Brosme brosme (Ascanius) (Cusk) 0 0 0 0 0 0 0 0 1 0 0 0 1 B* Ammodytes americanus DeKay (American Sand Lance) 0 1 0 0 0 0 0 0 0 0 0 0 1 P* Lophius americanus Valenciennes in Cuvier and 0 0 0 0 0 1 0 0 0 0 0 0 1 P* Valenciennes (Goosefish) Anguilla rostrata (Lesueur) (American Eel) 0 0 0 0 0 0 0 0 0 0 0 1 1 S* Morone americana (Gmelin) (White Perch) 0 0 0 0 0 0 0 0 0 0 0 1 1 F* Total 1057 1744 2054 867 4960 4798 9954 7865 3232 10,572 5026 8347 60,476 Northeastern Naturalist 554 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 Figure 2. Rarefaction curves of species richness for intertidal seine samples taken in Cobscook Bay, ME, 2011–2013. All seine curves reached a visual asymptote. Intervals of 1 standard deviation are represented by the gray areas. Richness estimates were generated by n = 100 random re-samples of the data. Table 3. Species-richness estimates and the calculated ACE-1 (abundance-based coverage estimates) and 95% confidence intervals for each year of sampling in the intertidal and subtidal areas of Cobscook Bay, ME. Area Year Species Richness ACE-1 Intertidal 2011 11 11 ± 0.0 2012 12 14 ± 4.7 2013 14 18 ± 6.7 Benthic 2011 24 26 ± 3.3 2012 26 33 ± 7.2 2013 31 35 ± 4.2 Pelagic 2011 12 16 ± 5.3 2012 7 11 ± 6.5 2013 11 23 ± 15.4 Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 555 Temporal patterns in diversity No clear pattern of species richness was evident among years and months (Fig. 5). Eighteen species of the total 46 for the study were absent from May samples, 10 were absent from June, 11 were absent from August, and 13 were absent from September. In contrast, 2 species were unique to May samples (Enchelyopus cimbrius [Fourbeard Rockling] and Brosme brosme [Cusk]); 3 to June samples (Lophius americanus [Goosefish], Triglops murrayi [Moustache Sculpin], and Ammodytes americanus [American Sandlance]); 1 to August samples (Squalus acanthias [Spiny Dogfish]); and 3 to September samples (Morone americana [White Perch], Anguilla rostrata [American Eel], and Raja eglanteria [Clearnose Figure 3. Gear-specific rarefaction curves of species richness for samples taken in Cobscook Bay, ME, 2011–2013. All subtidal benthic curves reached a visual asymptote. Subtidal pelagic curves did not reach clear asymptotes. Intervals of 1 standard deviation are represented by the gray areas. Richness estimates were generated by n = 100 random re-samples of the data. Northeastern Naturalist 556 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 Skate]). All species unique to a month were represented by a single individual, except for Moustache Sculpin (n = 2) and Spiny Dogfish (n = 3) (Table 2). Diel differences in intertidal species richness were small (≤ 2 species) and inconsistent. Subtidal species richness was always higher at night (Fig. 5), driven by the capture of rare species (less than 40 individuals) in the night benthic trawls. The difference ranged from 1 to 11 species and averaged more than 6 species. We detected no significant differences when examining diel differences with NMDS (Bray-Curtis: 3 dimensions, stress = 0.134). Observed assemblage evenness (Simpson’s index) in the intertidal was variable among months and years. Species evenness was highest in a different month each year and varied from 0.230 to 0.768 (Fig. 6, top left). ANOVA did not identify any group-level factors with predictive power (Table 4). In the subtidal benthic assemblage, peak evenness occurred in June in 2011 and August in 2012 and 2013, and varied from 0.226 to 0.765 (Fig. 6, top right). ANOVA indicated that month was a significant group-level factor (Table 4). Tukey’s comparisons subsequently indicated that evenness in both August and September was significantly higher than evenness in May (Table 5). In the subtidal pelagic assemblage, calculations of Simpson’s index were inconsistent due to small sample sizes and precluded meaningful comparisons among months and years. Diel differences of species evenness in the intertidal did not present discernable patterns across months or between 2012 and 2013, when adequate night samples were taken. ANOVA identified time of day as a significant group-level factor in Figure 4. Finfish assemblage similarity by zone for B = benthic subtidal samples, P = pelagic subtidal samples, and S = intertidal samples taken in Cobscook Bay, ME, 2011–2013. Points represent a year and month combination. (Left) NMDS ordination with samples partitioned by gear. The Bray–Curtis dissimilarity index run in 3 dimensions produced a stress of 0.101. (Right) NMDS ordination with subtidal samples partitioned by gear. The Bray–Curtis dissimilarity index run in 3 dimensions produced a stress of 0.096. The ellipses show 95% confidence intervals. Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 557 predicting assemblage diversity, also indicating the presence of significant interactions between time of day and both month and year (Table 4). Tukey’s comparisons indicated that average evenness was greater at night (Table 5). Further statistical analysis through linear regression suggested that the influence of night sampling on observed evenness was stronger in 2013 (Table 5). In the subtidal benthic zone, diversity was either similar between days and nights, or was comparably higher at night (Fig. 6, bottom right). ANOVA identified month and diel as significant group-level factors for predicting subtidal species evenness (Table 4). Tukey’s comparisons indicated that assemblage evenness was generally higher at night. Also, average evenness was significantly higher in August and September as compared to May samples (Table 5). In the subtidal pelagic zone, data inconsistencies precluded statistical analyses. Figure 5. Species richness by year and month (top) and by day and night (bottom) for intertidal (seine and fyke; left) and subtidal (benthic and pelagic; right) zones in Cobscook Bay, ME, 2011–2013. Northeastern Naturalist 558 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 Spatial patterns in diversity In the intertidal zone, species richness was similar among the 3 sub-bays (Fig. 7). The average species richness was 6.42 in Inner Bay (120 samples), 7.17 in Central Bay (196 samples), and 5.08 in Outer Bay (74 samples). In contrast, subtidal species richness varied among sub-bays. In Inner Bay, richness averaged 5.09 (48 samples), in Central bay 13.42 (93 samples), and in Outer Bay 11.00 (82 samples) (Fig. 7). Species evenness varied over space in the intertidal and subtidal finfish assemblages (Fig. 8). We did not discern any general pattern in these differences. ANOVA identified sub-bay as a significant group-level factor and indicated the presence of significant interactions between sub-bay and year (Table 4). Tukey’s comparison indicated that average evenness in Outer Bay was higher than the average evenness Figure 6. Simpson’s index by month and year from 2011–2013 (top) and by day and night from 2012–2013 (bottom) for intertidal (left) and benthic subtidal (right) zones in Cobscook Bay, ME. Bootstrapped (n = 1000) error bars represent ± 2 standard deviations. Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 559 in Central Bay. Subsequent regression analysis indicated that average evenness in Central Bay was lower in 2013 (Table 5). ANOVA did not identify spatial factors as having significant predictive power with respect to assemblage evenness in the subtidal (Table 4). The inconsistency of data in the subtidal pelagic zone precluded those data from meaningful statistical analysis. The CPUEs in subtidal samples differed widely among sub-bays across months and years, but pelagic zone CPUEs were lowest in Outer Bay, except in June 2012, and benthic zone CPUEs were lowest in Inner Bay, except in September 2011 Table 4. ANOVA table showing the group-level F-statistics and their associated P-values. Analysis results for all group-level factors are shown. Interaction terms identified as significant are also shown. Significant factors (P ≤ 0.05) are marked with an asterisk (*). The + symbol identifies 2 significant F-statistics produced by identical models. Assemblage Scale Factor F-value P-value Intertidal Temporal Year 1.286 0.279 Month 1.102 0.349 Spatial Year 1.411 0.246 Month 1.210 0.308 Subbay* 3.622* 0.029* Year: Subbay* 3.589* 0.002* Diel Year 1.343 0.248 Month 1.794 0.151 Diel* 21.625* >0.001* Diel: Month* 2.333* 0.035* Diel: Year* 4.366* 0.014* Subtidal Benthic Temporal Year 0.345 0.709 Month* 4.285*+ 0.007* Spatial Year 0.432 0.651 Month* 4.285*+ 0.007* Subbay 0.218 0.804 Diel Year 0.032 0.860 Month* 4.412* 0.008* Diel* 12.365* >0.001* Table 5. Results of post-hoc analysis of average within-group effects. We employed Tukey’s test to examine differences between the mean effects of within-group factors. Linear regression was used to examine difference between the mean effects of in-group interaction terms (indicated with an asterisk [*]). Only significant (P ≤ 0.05) differences between paired in-group means are shown. Assemblage Model Term Pair Estimate P–value Intertidal Subbay Outer:Central 0.088 0.041 Subbay : Year *Central:2013 -0.168 0.004 Diel N:D 0.143 >0.001 Diel : Year *N:2013 0.133 0.008 Subtidal Benthic Month August:May 0.177 0.024 September:May 0.184 0.019 Diel N:D 0.162 0.002 Month (w/ Diel) August:May 0.231 0.009 September:May 0.220 0.014 Northeastern Naturalist 560 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 and 2013 (Table 6). NMDS did not identify any significant trends despite these differences (Bray–Curtis: 3 dimensions, stress = 0.147). We explored additional subdivisions of these data using NMDS, e.g., intertidal habitat types (Horn–Morisita: 3 dimensions, stress = 0.15), the subtidal pelagic with respect to month (Bray-Curtis: 3 dimensions, stress = 0.047), the subtidal benthic with respect to sub-bay (Bray-Curtis: 3 dimensions, stress = 0.133), and diel differences (Bray-Curtis: 3 dimensions, stress = 0.134). We observed no significant differences or discernible patterns. Discussion Our objectives for this study were to characterize Cobscook Bay’s finfish assemblage in the absence of existing data and test the utility of various diversity Figure 7. Species richness for intertidal (left) and benthic and pelagic subtidal (right) zones in Cobscook Bay, ME, 2011–2013, by sub-bay, month, and year. In the subtidal, hatching represents daytime species richness overlaid onto the total for sub-bays and months in which night sampling occurred. Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 561 Figure 8. Simpson’s index for sub-bays of Cobscook Bay, ME, 2011–2013 by intertidal (left) and benthic and pelagic subtidal (right) zone, month, and year. Bootstrapped (n = 1000) error bars represent ± 2 standard deviations. Table 6. Catch per unit effort of benthic and pelagic tows in each sub-bay for all sampling events in Cobscook Bay, ME, 2011–2013. 2011 2012 2013 Gear Sub-bay May Jun Aug Sep May Jun Aug Sep May Jun Aug Sep Benthic Inner 0.44 1.06 0.51 1.08 0.18 0.63 0.64 0.03 0.05 0.1 4.11 6.58 Central 0.88 5.14 5.94 3.73 6.44 6.39 1.21 1.58 4.33 4.1 11.55 73.53 Outer 3.23 1.23 2.76 0.73 7.28 9.14 1.46 1.29 24.58 19.93 13.33 3.99 Pelagic Inner 3.70 2.55 0.00 0.00 30.22 10.03 0.00 0.00 0.12 58.71 10.75 0.00 Central 5.51 6.26 0.71 0.3 15.41 2.37 0.00 0.01 0.67 38.06 0.60 0.60 Outer 0.56 0.00 0.00 0.00 1.28 2.85 0.00 0.01 0.01 13.79 0.16 0.01 Northeastern Naturalist 562 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 measures in tidally energetic waters. Calculated rarefaction curves and the abundance- based coverage estimators (ACE) suggest our study was effective in this macrotidal system for sampling in the intertidal and subtidal benthic areas. This study sampled and documented a variety of species in the subtidal pelagic areas of Cobscook Bay. However, non-asymptotic rarefaction curves and a lack of catch in several months suggest that data collected there do not provide a representative sample of the subtidal pelagic assemblage. Species richness, Simpson’s index, the Bray–Curtis index of dissimilarity, and the Horn–Morisita index provided insight into the diversity of Cobscook Bay’s finfish assemblage. These indices identified 3 distinct assemblages: the intertidal, subtidal benthic, and subtidal pelagic. They also provided qualitative measures of assemblage diversity in each and quantitative measures of assemblage diversity in the intertidal and subtidal benthic zones. In the intertidal zone, species richness was generally correlated with sampling effort. This finding was expected because the catch of rare species correlates with sampling effort, and species richness is sensitive to rare species (Chao et al. 2005, Ebner et al. 2008, Gotelli and Colwell 2001). Intertidal sampling effort was most intense in Central Bay, and richness was greatest there in 201 1 and 2013. In the subtidal, most captured species were rare (n ≤ 40 individuals total), and only 21 of 39 species were captured in all 3 years of the study. Species richness (across all gears) always reached its annual peak in either June or August, an observation consistent with previous studies performed in coastal Maine waters (Ojeda and Dearborn 1990). In Cobscook Bay, Melanogrammus aeglefinus (Haddock), Urophycis tenuis (White Hake), and Peprilus triacanthus (Butterfish) were almost exclusively captured in August and September. Inshore movement in response to seasonally warming water (NERACOOS 2014) is consistent with the literature on these species (Collette and Klein-MacPhee 2002, Tyler 1971). Subtidal benthic species richness was consistently higher in nighttime samples. This result can be explained by increased nighttime catchability resulting from a combination of fishes’ reduced ability to avoid nets at night (Glass and Wardle 1989) and diel changes in their distribution throughout the water column. Viehman et al. (2015) observed significant diel differences in the distribution of fish in Cobscook Bay’s subtidal water column using hydroacoustics, whereby fish made fuller use of the water column at night, becoming most homogeneous across depths during the nighttime slack tide. Analysis of assemblage diversity with Simpson’s index did not reveal predictable trends in Cobscook Bay with respect to space or time. In the intertidal, estimates of assemblage diversity were generally driven by the most abundant intertidal species. Larger differences in evenness could generally be attributed to large catches (n ≥ 600 individuals) of Threespine Stickleback, Atlantic Silverside, and Alewife. When an aggregation of 1 species was sampled, it reduced diversity. When aggregations of 2 or more species were sampled, diversity was higher. Overall, we sampled 11 schools or aggregations—6 in Central Bay and 5 in Inner Bay. These catches may be attributable to spawning aggregations of Threespine Stickleback and schooling Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 563 behavior of Atlantic Silverside and Alewife (Ardekani et al. 2013, Collette and Klein-MacPhee 2002). The species that drove diversity in the subtidal benthic assemblage were variable. Winter Flounder were generally ubiquitous and typically dominated the subtidal benthic catch. Assemblage evenness thus reflected the cumulative abundance of other species relative to Winter Flounder. The exception to this trend was September 2013, when we sampled a large number (n = 5200) of Atlantic Herring in 1 benthic trawl. Several limitations may have affected this study’s results. In the intertidal, we chose sampling locations based on the presence of seineable slope and substrate, and accessibility by vehicle. We identified and sampled only 1 viable location in Outer Bay, 2 in Inner Bay, and 3 in Central Bay. We do not know that species composition and relative abundance in these locations were representative of those areas. Rockweed and Cordgrass habitat frequently tangled the seine or lifted it off the bottom, which created gaps through which we observed fish were escaping. Additionally, species such as Cryptacanthodes maculatus Storer (Wrymouth) likely were able to avoid capture by hiding in the available habitat (C. Bartlett, Maine Sea Grant, Eastport, ME, pers. comm.; Collette and Klein-MacPhee 2002). Wrymouth are undoubtedly present within Cobscook Bay, having been observed by local fishermen and scientists (C. Bartlett, pers. comm.), but they were rare in nearby Passamaquoddy Bay (Cooper and Blanchard 2016, Macdonald et al. 1984, Tyler 1971), perhaps for the same reason. Subtidal benthic trawling locations were restricted to smooth substrates and may not have properly represented species inhabiting rocky substrates. Dangerous conditions also precluded night sampling in Inner Bay, which may have reduced the effectiveness of sampling performed there; CPUE was always highest in either Central or Outer bay and was generally more than double Inner Bay’s. The subtidal gear also failed to ubiquitously sample fish of all sizes and species effectively. We captured only 75 individuals ≥30 cm. In addition, we rarely captured species (n = 16) known to be abundant in the area, such as Scomber scombrus (Atlantic Mackerel) (C. Bartlett, pers. comm.; Viehman et al. 2015). It is likely that Atlantic Mackerel were able to avoid the gear (Glass and Wardle 1989) because their swimming speeds (3.5–3.8 body lengths s-1; Misund 1993) greatly exceed the speed of the F/V Pandalus while trawling. Despite these potential biases, our results can be considered representative of Cobscook Bay’s intertidal and subtidal benthic finfish assemblages. Species richness, when combined with the species lists, was sensitive to rare species and captured the seasonal movements of several northwest Atlantic species. Simpson’s index was useful for detecting changes in the relative abundance of species and suggested that intertidal diversity was heavily influenced by aggregations, while subtidal benthic diversity was influenced by the relative abundance of Winter Flounder. These observations relied on the accompanying species lists. This combination of metrics has been used in other broad-scale descriptive studies of embayments to observe similar changes in finfish assemblages (Jung and Houde 20 03, Peet 1974). Northeastern Naturalist 564 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 Our results demonstrate the utility of these descriptive tools in macrotidal systems and support their continued use. An external point of comparison for the finfish composition of Cobscook Bay is Passamaqoddy Bay, an embayment 15 km north of Cobscook Bay. Passamaquoddy is also a macrotidal system with strong (~2 m s-1) tidal currents in channels and passages (Brooks 1992, 2004). Given its proximity to Cobscook Bay, it is plausible that their finfish assemblages are similar. This comparison is speculative, however, because extensive differences exist between the physical environments of both bays. In Passamaquoddy Bay 3% of the tidal volume is freshwater (Brooks 1992), and the benthic habitat is 70% mud and sand (Lotze and Milewski 2004, MacDonald et al. 1984). In contrast, less than 1% of Cobscook Bay’s tidal volume is freshwater, and the benthic habitat is composed of 70% rock and gravel (Kelley and Kelley 2004). Cooper and Blanchard (2016) recently extensively sampled Passamaquoddy Bay’s subtidal benthic finfish assemblage using a benthic trawl. Their work was largely presented in qualitative fashion, which precludes in-depth comparisons with Cobscook Bay. However, some general similarities are apparent. For example, Winter Flounder was ubiquitous in both bays; however, they were never the most abundant species in Passamaquoddy Bay. Few Merluccius bilinearis (Silver Hake) were caught in either bay in 2011. They were the dominant species in Passamaquoddy Bay in 2012 and 2013 and were abundant in Cobscook Bay over the same period. Atlantic Herring were observed sporadically in the subtidal benthic zones of both bays. Alewife, however, were infrequently observed in Cobscook Bay’s subtidal benthic zone, but were abundant in Passamaquoddy Bay in 2011, 2013, and 2014. Some species present in catches in Cobscook Bay were absent in Passamaquoddy Bay, e.g., Lumpenus lampretaeformis (Snakeblenny) and Pholis gunnellus (Rock Gunnel). Similarly, some species present in catches in Passamaquoddy Bay were absent in Cobscook Bay, e.g., Sebastes fasciatus Storer (Acadian Redfish), Glyptocephalus cynoglossus (L.) (Witch Flounder), and Limanda ferruginea (Storer) (Yellowtail Flounder). Passamaquoddy Bay was the focus of a more comprehensive study of assemblage dynamics in the late 1970s and early 1980s conducted by MacDonald et al. (1984). Their study included intertidal sampling with seines and subtidal benthic sampling with trawls. Threespine Stickleback, Atlantic Silverside, and Fundulus heteroclitus (Mummichog) were among the most abundant species in intertidal areas for both their study and ours, while Winter Flounder, Atlantic Herring, and Silver Hake were abundant in both subtidal benthic zones. The authors described seasonal patterns in Passamaquoddy Bay’s intertidal and subtidal benthic assemblage compositions. Intertidal and subtidal species richness were consistently highest in the summer months. This finding was consistent with observations made in Cobscook Bay and was generally attributed to the offshore movement of most species during the winter months. Noteworthy differences are also apparent between both studies’ results, which is not surprising because the work of MacDonald et al. (1984) is now dated. They observed 24 species that were not observed in Cobscook Bay. They also labeled several species as common or abundant that were rarely or never observed in Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 565 Cobscook Bay such as Gadus morhua (Atlantic Cod), Zoarces americanus (Ocean Pout), Hippoglossoides platessoides (American Plaice), and Limanda ferruginea (Storer) (Yellowtail Flounder). Those species, and several others, experienced substantial population declines in the years following the work of MacDonald et al. (1984) (Lotze and Milewski 2004; NEFSC 2012, 2014; Pershing et al. 2015). Today, Atlantic Cod, Ocean Pout, and Yellowtail Flounder are all considered overfished (NOAA Fisheries 2018). American Plaice was declared overfished in 2004, and though not currently overfished, it remains in a rebuilding plan (Mayo and Terceiro 2005, NOAA Fisheries 2018). Environmental variables may be contributing to the observed differences between Passamaquoddy Bay’s finfish assemblage in early 1980s (MacDonald et al. 1984) and Cobscook Bay’s finfish assemblage in the early 2010s. Our study overlapped with an anomalously warm year in 2012 (Chen et al. 2014, 2015; Mills et al 2013). Conditions in that year may have altered the behavior of some species. Butterfish were common in both Cobscook Bay and Passamaquoddy Bay in 2012. They were rare in both bays in 2011 and 2013 (Cooper and Blanchard 2016). Haddock were common in Cobscook Bay in 2011 and 2013. None were captured throughout all of 2012. At the broader scale, the Gulf of Maine is experiencing warming rates that are greater than 99% of the world’s oceans (e.g., Mills et al. 2013, Pershing et al 2015). Other species may be affected by the rapid warming in the Gulf of Maine and its influence on other environmental factors, e.g., changes in currents, salinity, and pH. Klein and colleagues’ (2017) analysis of effects of climate change on 4 commercially important species suggested that Winter Flounder (the most dominant benthic species in Cobscook Bay) is most vulnerable to future climate change, followed by Atlantic Cod (rare in Cobscook Bay); Haddock and Yellowtail Flounder were considered least vulnerable. Hare et al. (2016) projected future climate influences on >62 marine and diadromous fishes on the northeastern US continental shelf. Of those species captured in the Cobscook Bay study, 6 were projected to have high to very high climate exposure and biological sensitivity to climate-related factors, at least 15 had high to very high projected change in distribution, and 18 were projected to experience negative overall effects due to climate change. Winter Flounder and Hippoglossus hippoglossus (Atlantic Halibut) occurred in all 3 categories. Butterfish were projected to have very high potential for distributional change with climate-change direction being positive. We close with the speculation that several species abundant in Cobscook Bay, especially those with small body sizes there (e.g., Threespine and Gasterosetus whatlandi [Blackspotted Stickleback], Mummuichog, Atlantic Silverside, juvenile Alewife, and juvenile Osmerus mordax [Rainbow Smelt]) may become even smaller in body size because of 2 temperature-related constraints on scope for growth. First, warmer ocean temperature tends to favor the abundance of small-bodied, less nutritious zooplankton prey (Debertin et al. 2018, Moore and Folt 1993) leading to slower growth and perhaps increased mortality of finfish. Second, the 2-dimensional structure of fish gills constrains how much oxygen can be supplied to a growing Northeastern Naturalist 566 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 3-dimensional body, more of which must be used for metabolism rather than growth with increased temperature (Pauly and Cheung 2017). Conclusions Cobscook Bay is a dynamic physical environment with a diverse finfish assemblage. We observed large natural variation among years, months, and even individual days. We detected patterns and trends over space and time such as annual increases in species richness during the summer months. The baseline dataset provided by this study demonstrates an effective approach for sampling and describing a macrotidal finfish assemblage, extending the work presented in Special Issue 2 of the Northeast Naturalist (Larsen 2004) into the vertebrate realm. It can be used as a reference frame for any future investigations of Cobscook Bay’s intertidal and subtidal benthic finfish assemblages. Others have emphasized the importance of such baseline studies (Bull et al. 2014, Edgar et al. 2004). This research suggests that there are 3 distinct finfish assemblages in Cobscook Bay: the benthic and pelagic subtidal and the intertidal. We observed significant spatial and temporal differences in the intertidal, but no trends or patterns were identifiable aside from patterns in the relative abundance of individual species. As such, thorough documentation of the bay’s intertidal finfish assemblage can only be reliably achieved by widely sampling in space and time. Implications for any future studies conducted in this or similar settings is that a complex physical environment requires a study design sufficiently complex to capture the anticipated spatial and temporal variability finfish assemblage. The composition and relative abundance of species in the subtidal benthic assemblage also varied significantly over both spatial and temporal scales. It is important that any attempt to monitor the finfish assemblage accounts for and continues to investigate this variation as the bay (1) continues to be fished for Placopecten magellanicus (Gmelin) (Sea Scallop), Homarus americanus H. Milne Edwards (American Lobster), Strongylocentrotus droebachiensis (O.F. Müller) (Green Sea Urchin), and other invertebrates; (2) experiences residential and commercial development (e.g., tidal power); and (3) is af fected by global climate change. Acknowledgments We thank the many students, technicians, and others, especially Megan Altenritter, Garrett Staines, and Haley Viehman, who were willing to get their hands dirty sampling at all hours of the day and night and processing captured fishes. The expertise of Steve Brown and his crew on the F/V Pandalus allowed successful sampling in the subtidal zone. Chris Bartlett, of Maine Sea Grant and Maine Cooperative Extension, provided advice on the local environment and facilitated communication with the fishing community. This work was supported in part by an award from the Department of Energy under award number DE-EE0003647. The views expressed herein are those of the authors and do not necessarily reflect the views of Ocean Renewable Power Company or any of its sub-agencies. This project was supported by the USDA National Institute of Food and Agriculture, Hatch (or McIntire Stennis, Animal Health, etc.) project number #ME0-031716 through the Maine Agriculture and Forest Experiment Station, Publication # 3623. Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 567 Literature Cited Alemany, D., E.M. Acha, and O. Iribarne. 2009. The relationship between marine fronts and fish diversity in the Patagonian Shelf Large Marine Ecosystem. Journal of Biogeography 36:2111–2124. Anderson, M.J. 2001. Permutation tests for univariate or multivariate analysis of variance and regression. Canadian Journal of Fisheries and Aquatic Sciences 58:626–639. Ardekani, R., A.K. Greenwood, C.L. Peichel, and S. Tavaré. 2013. Automated quantification of the schooling behaviour of sticklebacks. EURASIP Journal on Image and Video Processing. 2013:61. DOI:10.1186/1687-5281-2013-61. Berger, W.H., and L.F. Parker. 1970. Diversity of planktonic Foraminifera in deep-sea sediments. Science 168:1345–1347. Brooks, D.A. 1992. Tides and tidal power in Passamaquoddy Bay: A numerical simulation. Continental Shelf Research 12:675–716. Brooks, D.A. 2004. Modeling tidal circulation and exchange in Cobscook Bay, Maine. Northeastern Naturalist 11 (Special issue 2):23–50. Bull, J.W., A. Gordon, E.A. Law, K.B. Suttle, and E.J. Milner-Gulland. 2014. Importance of baseline specification in evaluating conservation interventions and achieving no net loss of biodiversity. Conservation Biology 28:799–809. Chao, A. 1987. Estimating the population size for capture–recapture data with unequal catchability. Biometrics 43:783–791. Chao, A., and S.-M. Lee. 1992. Estimating the number of classes via sample coverage. Journal of the American Statistical Association 87:210–217. Chao, A., R.L. Chazdon, R.K. Colwell, and T.-J. Shen. 2005. A new statistical approach for assessing similarity of species composition with incidence and abundance data. Ecology Letters 8:148–159. Chen, K., G.G. Gawarkiewicz, S.J. Lentz, and J.M. Bane. 2014. Diagnosing the warming of the Northeastern US Coastal Ocean in 2012: A linkage between the atmospheric jet-stream variability and ocean response. Journal of Geophysical Research: Oceans 119:218–227. Chen, K., G. Gawarkiewicz1, Y.-O. Kwon, and W.G. Zhang. 2015. The role of atmospheric forcing versus ocean advection during the extreme warming of the Northeast US continental shelf in 2012. Journal of Geophysical Research: Oceans 1 20:4324–4339. Collette, B.B., and G. Klein-MacPhee. (Eds.). 2002. Bigelow and Schroeder’s Fishes of the Gulf of Maine, 3rd Edition. Smithsonian Institution, Washington, DC. 748 pp. Colwell, R.K. 2009. Biodiversity: Concepts, patterns, and measurement. Pp. 257–263, In S.A. Levin (Ed.). The Princeton Guide to Ecology. Princeton University Press, Princeton, NJ. 848 pp. Cooper, J.A., and M.J. Blanchard. 2016. Coastal biodiversity trawl of the Passamaquioddy Bay area: 2009 to 2014. Canadian Technical Report of Fisheries and Aquatic Sciences 3176. xi + 52 pp. Fisheries and Oceans Canada, Ottawa, ON, Cana da. Debertin, A.J., J.M. Hamson, and S.C. Courtenay. 2018. Linking zooplankton assemblages with oceanographic zones in an Atlantic coastal ecosystem. Canadian Journal of Fisheries and Aquatic Sciences. 75:868–882 DOI:10.1139/cjfas-2016-0342. Ebner, B.C., J.D. Thiem, D.M. Gilligan, M. Lintermans, I.J. Wooden, and S. Linke. 2008. Estimating species richness and catch per unit effort from boat electro-fishing in a lowland river in temperate Australia. Austral Ecology 33:891–901. Northeastern Naturalist 568 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 Edgar, G.J., R.H. Bustamante, J.-M. Fariña, M. Calvopiña, C. Martínez, and M.V. Toral- Granda. 2004. Bias in evaluating the effects of marine protected areas: The importance of baseline data for the Galapagos Marine Reserve. Environmental Conservation 31:212–218. Glass, C.W., and C.S. Wardle. 1989. Comparison of the reactions of fish to a trawl gear at high and low light intensities. Fisheries Research 7:249–266. Gotelli, N.J., and R.K. Colwell. 2001. Quantifying biodiversity: Procedures and pitfalls in the measurement and comparison of species richness. Ecology Let ters 4:379–391. Greenstreet, S.P.R., and S.J. Hall. 1996. Fishing and ground-fish assemblage structure in the North-Western North Sea: An analysis of long-term and spatial trends. Journal of Animal Ecology 65:577–598. Hare, J.A., W.E. Morrison, M.W. Nelson, M.M. Stachura, E.J. Teeters, R.B. Griffis, M.A. Alexander, J.D. Scott, L. Alade, R.J. Bell, A.S. Chute, K.L. Curti, T.H. Curtis, D. Kircheis, J.F. Kocik, S.M. Lucey, C.T. McCandless, L.M. Milke, D.E. Richardson, E. Robillard, H.J. Walsh, M. C. McManus, K.E. Marancik, and C.A. Griswold. 2016. A vulnerability assessment of fish and invertebrates to climate change on the Northeast US continental shelf. PLoS ONE 11(2):e046756. DOI:10.1371/journal.pone.0146756. Horn, H.S. 1966. Measurement of overlap in comparative ecological studies. American Naturalist 100:419–424. Jouffre, D., and C.A. Inejih. 2005. Assessing the impact of fisheries on demersal fish assemblages of the Mauritanian continental shelf, 1987–1999, using dominance curves. ICES Journal of Marine Science 62:380–383. Jung S., and E.D. Houde. 2003. Spatial and temporal variabilities of pelagic-fish community structure and distribution in Chesapeake Bay, USA. Estuarine, Coastal, and Shelf Science 58:335–351. Kelley J.T., and A.R. Kelley. 2004. Controls on surficial materials distribution in a rockframed, glaciated, tidally dominated estuary: Cobscook Bay, Maine. Northeastern Naturalist 11 (Special Issue 2):51–74. Klein, E.S., S.L. Smith, and J.P. Kritzer. 2017. Effects of climate change on four New England groundfish species. Reviews in Fish Biology and Fisheries 2 7:317–338. Larsen, P.F. (Ed.). 2004. Ecosystem modeling in Cobscook Bay, Maine: A boreal, macrotidal estuary. Northeastern Naturalist 11 (Special Issue 2):1–440 pp. Larsen, P.F., S. Barker, J. Wright, and C.B. Erickson. 2004. Use of cost-effective remote sensing to map and measure marine intertidal habitats in support of ecosystem modeling efforts: Cobscook Bay, Maine. Northeastern Naturalist 11 (Special Issue 2):225–242. Lotze, H.K., and I. Milewski. 2004. Two centuries of multiple human impacts and successive changes in a North Atlantic food web. Ecological Applications 14:1428–1447. MacDonald, J.S., M.J. Dadswell, R.G. Appy, G.D. Melvin, and D.A. Methven. 1984. Fishes, fish assemblages, and their seasonal movements in the Lower Bay of Fundy and Passamaquoddy Bay, Canada. Fishery Bulletin 82:121–139. Mayo, R.K., and M. Terceiro (Eds.). 2005. Assessment of 19 Northeast groundfish stocks through 2004. 2005 Groundfish Assessment Review Meeting (2005 GARM). Northeast Fisheries Science Center, Woods Hole, MA. 15–19 August 2005. US Deprartment of Commerce Reference Document 05–13. Northeast Fisheries Science Center, Woods Hole, MA. 499 pp. McCune, B., J.B. Grace, and D.L. Urban. 2002. Analysis of Ecological Communities. MjM Software Design, Gleneden Beach, OR. 304 pp. Northeastern Naturalist Vol. 25, No. 4 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 569 Melvin, G.D., and N.A. Cochrane. 2012. A preliminary investigation of fish distributions near an in-stream tidal turbine in Minas Passage, Bay of Fundy. Canadian Technical Report of Fisheries and Aquatic Sciences 3006. Fisheries and Oceans Canada, Ottawa, ON, Canada. 43 pp. Mills, K.E., A.J. Pershing, C.J. Brown, Y. Chen, F.-S. Chiang, D.S. Holland, S. Lehuta, J.A. Nye, J.C. Sun, A.C. Thomas, and R.A. Wahle. 2013. Fisheries management in a changing climate: Lessons from the 2012 ocean heat wave in the Northwest Atlantic. Oceanography 26:191–195. Misund, O.A. 1993. Avoidance behaviour of herring (Clupea harengus) and mackerel (Scomber scombrus) in purse-seine–capture situations. Fisheries Research 16:179–194. Moore, M., and C. Folt 1993. Zooplankton body size and community structure: Effects of thermal and toxicant stress. Trends in Ecology and Evolution 8:178–183. Morisita, M. 1959. Measuring of the dispersion and analysis of distribution patterns. Memoires of the Faculty of Science, Kyushu University. Series E. Biology 2:215–235. National Oceanic and Atmospheric Association (NOAA) Fisheries. 2018. Species information system public portal. Available online at https://www.st.nmfs.noaa.gov/sisPortal/. Accessed 15 March 2018. NOAA National Geophysical Data Center. 2011. Coastline extracted (e.g., WVS, GSHHG), 5/1/2011. Available online at http://www.ngdc.noaa.gov/mgg/shorelines/gshhs.html. Accessed 19 September 2017. Northeast Fisheries Science Center (NEFSC). 2012. 54th Northeast Regional Stock Assessment Workshop (54th SAW) Assessment Report. Stock Assessment Report 12-18. US Department of Commerce, Northeast Fisheries Science Center, Woods Hole, MA. 600 pp. NEFSC. 2014. 59th Northeast Regional Stock Assessment Workshop (59th SAW) Assessment Report14-09. Stock Assessment Report. US Department of Commerce, Northeast Fisheries Science Center, Woods Hole, MA. 782 pp. Northeastern Regional Association of Coastal and Ocean Observing Systems (NERACOOS). 2014. Ocean and Weather Climate Display. Available online at http://neracoos.org/ datatools/climatologies. Accessed 19 September 2017. Ojeda, F.P, and J.H. Dearborn. 1990. Diversity, abundance, and spatial distribution of fishes and crustaceans in the rocky subtidal zone of the Gulf of Maine. Fishery Bulletin 88:403–410. Oksanen, J., F.G. Blanchet, R. Kindt, P. Legendre, P.R. Minchin, R.B. O’Hara, G.L. Simpson, P. Solymos, M.H.H. Stevens, and H. Wagner. 2015. vegan: Community ecology package. Available online at https://cran.r-project.org/web/packages/vegan/index.html. Accessed 19 September 2017. Pauly, D., and W.W.L. Cheung. 2017. Sound physiological knowledge and principles in modeling shrinking of fishes under climate change. Global Change Biology 2017:1–12. Peet, R.K. 1974. The measurement of species diversity. Annual Review of Ecology and Systematics 5:285–307. Pershing, A.J., M.A. Alexander, C.M Hernandez, L.A. Kerr, A. Le Bris, K.E. Mills, J.A. Nye, N.R. Record, H.A. Scannell, J.D. Scott, G.D. Sherwood, and A.C. Thomas. 2015. Slow adaptation in the face of rapid warming leads to collapse of the Gulf of Maine cod fishery. Science 350:809–812. Pyle, J. 2012. Studying the Delaware Bay: 2011 report. Special Report. New Jersey Department of Environmental Protection, Trenton, NJ. Available online at http://www.state. nj.us/dep/fgw/artdelbaystudy12.htm. Accessed 19 September 2017. Northeastern Naturalist 570 J.D. Vieser, G. Barbin Zydlewski, and J.D. McCleave 2018 Vol. 25, No. 4 R Core Team. 2015. The R project for statistical computing. Vienna, Austria. Available online at http://www.r-project.org/. Accessed 19 September 2017. Scott, W.B., and M.G. Scott. 1988. Atlantic Fishes of Canada. Canadian Bulletin of Fisheries and Aquatic Sciences 219:731 pp. Sherman, S.A., K.L. Stepanek, C.A. King, R. Tetrault, and R. Eckert. 2012. Annual Report on the Maine–New Hampshire Inshore Trawl Survey. Annual Report. Maine Department of Marine Resources, Agusta, ME. Available online at http://www.maine.gov/ dmr/science-research/projects/trawlsurvey/reports/documents/2012.pdf. Accessed 19 September 2017. Shields, M.A., L.J. Dillon, D.K. Woolf, and A.T. Ford. 2009. Strategic priorities for assessing ecological impacts of marine renewable energy devices in the Pentland Firth (Scotland, UK). Marine Policy 33:635–642. Shields, M.A., D.K. Woolf, E.P.M. Grist, S.A. Kerr, A.C. Jackson, R.E. Harris, M.C. Bell, R. Beharie, A. Want, E. Osalusi, S.W. Gibb, and J. Side. 2011. Marine renewable energy: The ecological implications of altering the hydrodynamics of the marine environment. Ocean and Coastal Management 54:2–9. Simpson, E.H. 1949. Measurement of diversity. Nature 163:688. Tyler, A.V. 1971. Periodic and resident components in communities of Atlantic fishes. Journal of the Fisheries Research Board of Canada 28:935–946. Viehman, H.A., G.B. Zydlewski, J.D. McCleave, and G.J. Staines. 2015. Using hydroacoustics to understand fish presence and vertical distribution in a tidally dynamic region targeted for energy extraction. Estuaries and Coasts 38 (Supplement 1):s215–s226 .