2011 NORTHEASTERN NATURALIST 18(Monograph 8):1–128 Elucidating Marine Biogeography with Macrophytes: Quantitative Analysis of the North Atlantic Supports the Thermogeographic Model and Demonstrates a Distinct Subarctic Region in the Northwestern Atlantic Walter H. Adey1,* and Lee-Ann C. Hayek2 Abstract - Quantitative analysis of North Atlantic macrophyte (seaweed) abundance was used to test the thermogeographic model (TM) of marine biogeography. The TM uses coastal area and seawater temperature over Pleistocene time to reveal long-term climate/ area clusters that predict biogeographic regions. An earlier study with the TM predicted 20 of 24 classical biogeographic regions; the 4 omitted regions were either weak, disputed, or structurally (sandy/silty regions) inapplicable to the TM. The TM predicted a new western North Atlantic Subarctic Region, centered on the Strait of Belle Isle and lying between Newfoundland and Labrador. In contrast, Nova Scotia and the Gulf of Maine were found to be a Boreal/Subarctic transition zone. The predictions of the TM were earlier supported with coralline algal abundances, and here are supported by a test with macrophyte assemblages. Seaweed assemblages were studied at 51 primary SCUBA stations with 4–7 standard depth zones at each station from southern Labrador to Cape Elizabeth, ME. Meter-square quadrats were taken at each station, followed by at-sea sorting, identification, and weighing of biomass by species. A permanent record of these studies is provided by archived underwater photography and voucher herbarium samples, including tissue in silica gel to support future DNA analyses. Comparisons of European seaweed assemblages were accomplished by re-analyzing semi-quantitative data available in the literature. We demonstrate with the biomass data that a 3000-km stretch of coast (i.e., northern Gulf of St. Lawrence, northeastern Newfoundland, and southern Labrador) that is centered on the Strait of Belle Isle has a unique assemblage of seaweeds. In its location and marine climate, this coast closely matches the Subarctic Region predicted by the TM. Based on graphic demonstration, this Subarctic Region is radically different from that of Nova Scotia and the Gulf of Maine, and its dominant species derive from the North Pacific Ocean. Several different statistical approaches applied to the macrophyte data demonstrate the strength of these findings, while also finding that a subset of Subarctic species persists in deeper water to the south in the transition zone (i.e., southwestern Nova Scotia, Gulf of Maine). Only a very small proportion of the seaweed flora of the Subarctic Region is Arctic in origin (<4%). Earlier geographic analyses of biodiversity did not discover the Subarctic Region because rare species hide the strength of assemblages based on abundance. Many of the dominant seaweeds in the transitional region of the Gulf of Maine and Nova Scotia occur widely in Europe, and are Boreal in origin. Such Boreal species form 47–80% of the macrophyte biomass at depths shallower than 5 m in the 1National Museum of Natural History, Department of Botany, NHB-166, PO Box 37012, Smithsonian Institution, Washington, DC 20013-7012. 2National Museum of Natural History, Statistics and Mathematics, NHB-MRC-121, PO Box 37012, Smithsonian Institution, Washington, DC 20013-7012. *Corresponding author - adeyw@si.edu. 2 Northeastern Naturalist Vol. 18, Monograph No. 8 northwestern Atlantic transition zone southwest of the Subarctic Region. In deeper water (>5 m), however, European Boreal species are only 22–25% of the macrophyte biomass. Although many of the dominant, shallow sublittoral species of the European Boreal flora have made the North Atlantic passage, only about 10% (by biomass) of mid-depth species (2.5–5 m) have crossed and even fewer of the deeper European species are found in the northwestern Atlantic. A few European “Boreal” seaweeds reach further north into the Subarctic Region, but only at low levels of biomass and primarily at mid-depths (2.5–5 m). The Subarctic is inhospitable for establishment of such species in shallow water because of low winter temperatures and sea ice, and in deeper water by low summer temperatures (<5 ºC). Few ecologists have had the opportunity to work in the Subarctic Region; instead, most of the recent descriptive and experimental studies of northwestern Atlantic ecosystems are based on studies in the smaller transition zone in the Gulf of Maine and southwestern Nova Scotia. Consequently, we provide detailed photographic, tabular, and graphic descriptions of the seaweed assemblages within the Subarctic and their interactions with the major grazer, the sea urchin Strongylocentrotus droebachiensis. Three of the four most abundant seaweeds in the North Atlantic Subarctic Region (Agarum clathratum, Desmarestia viridis, and Ptilota serrata) occur in mid-deep water and are protected from grazing by secondary chemicals. The kelp Alaria esculenta often forms a monocultural canopy in the shallow subtidal zone, and it is the secondmost abundant seaweed in the Subarctic. Sea urchins form “fronts” that graze back the Alaria in spring and summer when wave action is moderate, but the wide Alaria zone is persistent on more exposed shores. Agarum clathratum, D. viridis, and P. serrata) in deeper water tend to form a savanna: dense patches of fleshy seaweeds separated by areas of coralline/urchin “barrens”. Unlike Arctic or estuarine barren zones, the coralline “barrens” of Subarctic rocky shores are calcitic, biostromal structures with a rich invertebrate infauna. These formations demonstrate great temporal stability, with longevities of decades to centuries, as shown through laser/ablation mass spectrograph analysis of Mg2+ content in the coralline Clathromorphum compactum. This research examines the relationship of the Subarctic to the eastern Atlantic Boreal and to the largely Boreal-dominated transition zone in the Gulf of Maine and Nova Scotia. By analyzing the TM and its relationship to the history of deglaciation after the Last Glacial Maximum (LGM), we hypothesize the western Atlantic rocky shore was occupied primarily by Subarctic glacial relict seaweeds in the early Holocene. During the mid-Holocene’s pre-rebound hypsithermal, due to a warmer and more open Canadian Northwest Passage, these species were likely supplemented by an injection of North Pacific Subarctic seaweeds. In the latest Holocene, and perhaps only in historical times, a surge of eastern North Atlantic Boreal seaweeds entered and came to dominate the southern, climatically transitional half of this coast; invasion and change continue today, and likely represent a major anthropogenic bridging of an oceanographic barrier. 2011 W.H. Adey and L.-A.C. Hayek 3 TABLE OF CONTENTS Introduction 4 The fog of marine biogeography 4 Thermogeography: A review and introduction to the thermogeographic 5 model (TM) Pleistocene glaciation and the formation of the Atlantic Subarctic 9 Goals of this investigation 10 Seaweed assemblages and plant-animal interactions in the northwestern 11 Atlantic rocky shore ecosystem Strongylocentrotus droebachiensis 14 The macroalgae 18 Alaria esculenta 19 Saccharina spp. 20 Laminaria digitata 24 Agarum clathratum 25 Ptilota serrata 27 Chondrus crispus 27 Crustose corallines 30 Physical ecological factors 33 Materials and Methods 35 Station locations 35 The research vessel 37 The stations and sampling procedures 37 Statistical methods 41 Rocky pinnacle sites 42 Taxonomy and identification 43 Results 44 Newfoundland/Labrador/Quebec (NLQ) 49 Rocky shore 49 Rocky pinnacles 57 The Gulf of Maine (GOM) 58 Southern Nova Scotia (SNS) 63 Discussion 71 The Subarctic is distinct from the Arctic 72 The sea-ice factor 72 Industrial fishing and Strongylocentrotus create a faux region? 77 Rocky pinnacle macroalgal assemblages 80 Seaweed assemblages of the North Atlantic Subarctic 82 The North Pacific Subarctic vs. the North Atlantic Subarctic 83 The history of the Subarctic vs. Boreal regions and their transition zone 86 Comparative analysis of seaweeds of the Subarctic and Boreal regions 87 Invertebrate populations 101 Conclusions 101 Literature Cited 104 Appendices A. Station data: (1) exposed, (2) intermediate, (3) protected, (4) rocky pinnacle 111 B. Station summary analysis 116 C. Station mean exposure status 116 D. Species abundance data by region 117 E. Species cited with taxonomic authority: (1) algae, (2) animals 126 4 Northeastern Naturalist Vol. 18, Monograph No. 8 Introduction The fog of marine biogeography Based on the presence or absence of fish and invertebrate species, especially “endemic” species, there is a large literature supporting the existence of coastal marine biogeographic provinces or regions (e.g., Briggs 1974, 1995; Ekman 1953; Vermeij 1978). Nevertheless, the field suffers from inconsistencies and disputes. Longhurst (1998) stated that, in a physically based, pelagic-centered biogeography, the “taxonomic biogeography of the sea belongs to the family of intractable scientific problems.” Rosenzweig (1995) referred to the “classical” biogeographic regions as “extremely leaky”. Algal phytogeography, a latecomer to biogeography, is also contentious (e.g., Garbury and South 1989). Van den Hoek (1975) redefined biogeographic regions based on the appearance and disappearance of species at regional margins, rather than by the zoologists’ percentage of endemics, while Lüning (1990) preferred summer/winter isotherms. An index numeral that is a ratio of red plus green algae to brown algae is sometimes applied in an attempt to separate biogeographic regions, but it produces only gradients, rather than distinct regions (e.g., Mathieson et al. 1991). Coastal benthic biogeography has been greatly limited by the lack of a theoretical base; until the end of the 20th century, it was an empirical science. Algal phytogeography, especially, has been constrained by a near total reliance on presence/absence data, species lists, and floras. Biogeographic regions are often characterized with endemic and rare species on the “tail of the curve”, where they are least abundant, and most variable, rather than with abundance data based in means and statistical analysis. Rarely have these classical approaches taken into account the blurring of boundaries by Pleistocene climatic cycling, the widespread transport of marine organisms by humans in the past several centuries, and by inconsistencies of approach. Marine organisms are distributed along coasts primarily in accordance with temperature, salinity, substrate, and other physical/chemical factors, and secondarily by organism interactions. At the large scale, the movement and shape of continents can establish oceans, seas, and straits that can provide for the isolated evolution of organisms. Occasionally, local geography in the form of the configuration of coastlines can also establish physical parameters and set barriers to the movement of organisms. More often, the distribution of coastal marine organisms is independent of the coastline configuration as it appears on maps and charts. While marine scientists may study organisms within the confines of their local bay, gulf, state, or country, the factors that control the distribution of organisms rarely relate to this aspect of geography. While maps and charts have always been the primary analytical and presentational tool of marine biogeographers, often maps confuse the relationship between physical factors and organism distribution. These difficulties can be overcome with models that utilize the principal factors controlling organismal distribution along coastlines, while avoiding maps and charts as primary analytical tools. 2011 W.H. Adey and L.-A.C. Hayek 5 Thermogeography: A review and introduction to the thermogeographic model (TM) Adey and Steneck (2001) developed a time-integrated thermogeographic model to demonstrate why benthic marine algal assemblages of coastal rocky marine, sublittoral zones will develop biogeographic patterns in their distribution and abundance. The TM is a predictive, abiotic model in which the maximum and minimum sea-surface temperatures (sea climate) are tabulated and plotted for each nautical mile of rocky coastline for both the present (1955) and the Last Glacial Maximum (LGM) (CLIMAP 1976; see Adey and Steneck 2001). Next, the density of nautical miles for each degree C square is contoured for the present and for the LGM (18K). These two alternate states (glacial and interglacial) characterize the principal climatic character of the global marine realm since late Pleistocene time (0.7–1.8 Ma); it is during this time that most living species have evolved (Briggs 1995). Some sea climates (specific thermal regimes) have a large amount of coast, others none. When the sea climate/coastal area contour diagrams for LGM (glacial) and present (interglacial) times are laid on top of each other (and mathematically integrated), another contour diagram results (Fig. 1.). This diagram is the thermogeographic model. It is seen that some sea climates with large coastal areas remain more or less constant through Pleistocene time, while other sea climates are absent or ephemeral (lacking significant coastlines over Pleistocene time). For example: the Tropical West Atlantic (rocky/coral shore) expands and contracts slightly during the Pleistocene, but it does not alter its essential character of a large coastal area with summer temperatures >25 ºC and winter temperatures >18 ºC. On the other hand, the Arctic rocky shore presents a very large, shallow coastal area of <-1.5 ºC during the winter and <5 ºC during summer through interglacial times. However, during glacial times it becomes diminishingly small, essentially being pushed southwards by continental glaciation to narrow transitional bands in the northern Atlantic and Pacific oceans. The lack of biodiversity in the Arctic (Adey et al. 2008), in spite of its present large area, reflects this instability through time. The resulting abiotic “thermogeographic” model (Fig. 1) defines 20 regions that correspond with the cores of 24 traditionally recognized biogeographic regions determined by published distributions of organisms (see Adey and Steneck 2001 for a geographic comparison). The four remaining classical regions were weak and disputed or lacked significant rocky shore. In the colder North Atlantic, the primary traditionally defined regions (e.g., Briggs 1974), the Western Atlantic Boreal and the Eastern Atlantic Boreal, are represented by equivalent regions in the TM (the Subarctic-west and Boreal or Celtic-east, respectively). However, the boundaries of those regions in the western Atlantic are markedly different between the classical organism-defined approach (Cape Hatteras to the Strait of Belle Isle; Briggs 1974), and the physical/time model (Newfoundland and northern Gulf of St. Lawrence to central Labrador; Adey and Steneck 2001). Thermogeographic regions (TRs), although clearly correlated with most classical biogeographic regions, have many shapes in area/sea climate space, from 6 Northeastern Naturalist Vol. 18, Monograph No. 8 elliptical to two-or three-lobed. In a few cases, two relatively strong TRs are conjoined by a narrow isthmus. Most striking, however, is that 10–20% of the world’s coastline does not appear to lie in any TR. These transitional zones, with no equivalent biogeographic regions, result from shifting Pleistocene climates and their contained coastlines. An important finding of the TM is that the rocky western North Atlantic Coast from Long Island Sound through the Gulf of Maine, and the southern Gulf of St. Lawrence comprises one of these transitional zones (Adey and Steneck 2001, see their Fig. 7). Adey and Steneck (2001) also proposed that biogeographic patterns should be determined by quantitatively analyzing community assemblages, rather than by the presence or absence of individual “key” species or percent endemism. They tested the efficacy of thermogeographic regions, as determined by the abiotic TM, with abundance-determined patterns in the distribution of crustose coralline red algae (Rhodophyta/Corallinales) in the colder part of the northern hemisphere. This group was chosen because a large, quantitative set of bottom-cover data Figure 1. Coastal thermogeographic regions as defined by the Adey and Steneck (2001) five-dimensional abiotic model (TM). The principal variables were mean minimum and maximum surface temperature, with coastal area over time (present and 18K) appearing as contours. Isolation by oceans and continents, i.e., northern and southern and Atlantic and Pacific/Indian Oceans, were introduced by separating the main diagram into quadrants, and then stretching some overlapping coasts. The strength of each region is represented by the number of contours of coastal area that is constant over Pleistocene time. 2011 W.H. Adey and L.-A.C. Hayek 7 covering much of the northern North Atlantic and part of the North Pacific had been published (see Adey and Steneck 2001, for review). Figure 2 combines several diagrams of Adey and Steneck (2001: figs. 3, 6, and 8) to demonstrate the relationship between the TM and one algal assemblage (corallines)—the background contours in this figure are the present-day coastal area within temperature regimes; the ellipses represent peaks of coastal area integrated over Pleistocene time (i.e., biogeographic regions); and the blue contours are Subarctic coralline relative abundance. Labeled and shown as red lines in the diagram are key coastlines, not in geographic space as latitude and longitude, but rather their TM temperature. Subarctic coralline species peak in abundance at greater than 90% within the region designated as Subarctic by the TM. Away from the center, towards both Arctic and Boreal, the abundance of Subarctic species is gradually reduced. Adey and Steneck (2001) present similar diagrams for Arctic and Boreal coralline species. Figure 2. The distribution of Subarctic crustose coralline algal cover (blue contours) on a background of present-day coastal area distribution as a function of summer–winter temperatures as determined in the Adey and Steneck (2001) TM. The background diagram shows the temperate to Arctic coastal area/temperature distribution (Adey and Steneck 2001: Fig 3) with the resultant thermogeographic regions (ellipses) superimposed (Adey and Steneck 2001: Fig. 6). The location of “core” Subarctic coast (northern Gulf of St. Lawrence, northeastern Newfoundland and Labrador) is shown as a red line within the Subarctic ellipse in the center of the diagram and extending nearly to the Arctic circle, because it includes northern Labrador. The Gulf of Maine and Nova Scotia are shown as red lines, with their intersection representing the Bay of Fundy. 8 Northeastern Naturalist Vol. 18, Monograph No. 8 In a section showing coralline abundance diagonally across the North Atlantic in TM hyperspace, the difference between species abundance (cover) and species number is strikingly illustrated (Fig. 3); species number is gradational, but “home” species are highly abundant within regions (note that the line traced in TM hyperspace passes across the intersection of the Gulf of Maine and Nova Scotia [i.e., the Bay of Fundy] at North Norway in the diagram). While the cores of regions, in terms of abundance, are highly dominated by species evolved to fit those regions, species from adjacent regions intermix in transitional areas. The TM was a first attempt at a theoretical basis for coastal biogeography. Since the 1980s, when the data were accumulated and the program run on a large main-frame computer, there has been a considerable increase in our knowledge of Pleistocene climate and the intricacies of sea-surface temperatures (Greene et al. 2008). A finer-tuned model could now be constructed. Nevertheless, the success of the TM in matching globally traditional coastal biogeography, as well as North Atlantic coralline abundance, is apparent. In hindsight, the nature of the primary problems in the traditional, descriptive biogeographic models becomes clearer. Figure 3. Abundance (cover) and species number of coralline regional groups plotted along a line on Figure 2 running from fringe Arctic through the Subarctic; continuing, this section runs diagonally along the N–S (North to South) Outer Norway line shown on Figure 2, thence diagonally through the middle of the North Atlantic Boreal to Galicia, Spain. 2011 W.H. Adey and L.-A.C. Hayek 9 First, in the light of the cyclicity of the Pleistocene climate, during which time most living species have evolved and migrated, it was highly unlikely that all coastal biogeographic regions would be contiguous without transition zones; it was also highly unlikely that biogeographic patterns of organismal distribution would be tightly linked to modern geography. Second, as we have noted, species presence/ absence information is likely to create transitory, illusory boundaries, even in cases in which short-distance transitions are created by time and geography. Pleistocene glaciation and the formation of the Atlantic Subarctic Adey et al. (2008) and Greene et al. (2008) outline the geological history of the colder northern hemisphere during the Pleistocene, with special emphasis on the opening of The Bering Strait (about 3.5 million years ago [Ma]) and the development of the Arctic as a biogeographic region. The Subarctic as a biogeographic region has an even more ancient history, having formed in the northern Pacific during the Pliocene (5.3–1.6 Ma) and provided the biological materials for the development of the Arctic biota. As Adey and Steneck (2001) show, the Atlantic Subarctic is a subregion of the larger, long-term “biogeographic engine” that is the North Pacific Subarctic (Briggs 2003). Not all glacial episodes were as severe as the late Wisconsinian glaciation (LGM), when the Laurentide ice sheet moved nearly to the margin of the northwestern Atlantic continental shelf (Shaw 2006, Stea 2004). Although during glacial episodes, finer sediments would have occurred on much of the outer shelf due to lowered sea level, rocky patches would likely have also existed, especially on the outer shore of the fore-glacial bulge (Shaw 2006, Shaw et al. 2002). Thus, limited glacial refugia for Arctic and Subarctic rocky shore biota would have been present, and would have formed the basis for a small endemic element in the rocky shore biota (e.g., Addison and Hart 2005). Maggs et al. (2008) discuss the possibility of distinct genetic signatures for such “refugium” species, and future work in this field may well assist in understanding the extent of refugia. Today, much of the northwestern Atlantic Subarctic biota is linked to the larger North Pacific Region through the filter of the Canadian Arctic Archipelago. This connection might have occurred during many interglacials, although it is also possible that exchange was limited to a few “super interglacials” when sea level through the Archipelago was higher (3.5 Ma, 0.8–1.2 Ma, 120 Ka [Eemian], 5 Ka [mid Holocene]; Greene et al. 2008). Subarctic species passing through the filter would have joined with the species that had survived glaciation locally in refugia to form a distinct subregion of the North Pacific Subarctic. During the latter half (<5000 years ago) of the Holocene interglacial, much of the southern part of the northwestern North Atlantic would have been climatically a transition zone between the eastern Atlantic Boreal and the Subarctic, and presumably this was true during many earlier interglacials. However, the depression southward of sea water isotherms during previous glacial times would have sharpened the coastal temperature gradient against a Gulf Stream-warmed shore (CLIMAP Project Members 1976). Perhaps more important for seaweeds, it would have moved Boreal species southwards to non-glaciated shores, highly dominated by unstable fine sedimentary material. That shift would have provided 10 Northeastern Naturalist Vol. 18, Monograph No. 8 limited glacial refugia for seaweeds that are adapted to rocky shores. Consequently, most Boreal seaweeds to enter this area after a glaciation would likely require passage across the North Atlantic. During the latest part of the Holocene, human activities would have greatly increased the rate of seaweed transfer; thus, the northwestern Atlantic rocky shore ecosystem, south of the “core” region of Newfoundland, Labrador, and northern Gulf of St. Lawrence, is currently in a very dynamic state (e.g., Johnson et al., in press). Goals of this investigation 1. To perform a second test of the TM using the biomass of species of fleshy macroalgae. We accomplish this test in part by determining statistically the significance of the difference in seaweed assemblages between the western Atlantic Subarctic of the TM and the transitional zones of the Gulf of Maine and Nova Scotia (the latter being the center of the Western Boreal Region of classical shore biogeography). We also employ multivariate Bray-Curtis and MDS analyses to test whether the same relationship, which exists in TM hyperspace, defines equivalent assemblages of seaweeds; 2. To describe the seaweed assemblages of the Subarctic Region in comparison to the transitional zones to the south; the difference is not generally recognized in the 20th-century rocky shore literature. This study covers the core of the Subarctic centered on the Strait of Belle Isle; we also cite published descriptions and experimental manipulations that exist for the western, southern, and eastern margins of the Region; 3. Using the same methodology, to describe the seaweed assemblages of the Gulf of Maine and Nova Scotia transitional zones and relate them to the Boreal flora of the eastern North Atlantic; and 4. To state a series of hypotheses based on the TM and our rapidly expanding understanding of late Pleistocene events, as well as our quantitative understanding of seaweed assemblages as presented in this paper and in supporting literature. In the remainder of this Introduction, we briefly review the literature on seaweed assemblages in the northwestern North Atlantic, followed by a review, partly with visual images, of the dominant seaweeds and their animal interactions of the Subarctic Region, Nova Scotia, and the Gulf of Maine. We do the species review for three reasons: (1) the field of coastal marine biogeography is dominated by zoogeographers, who have limited phycological backgrounds, (2) few marine biologists and ecologists have had the opportunity to work in the Subarctic rocky subtidal (as opposed to the Boreal transitional/area comprised of the Gulf of Maine and Nova Scotian coasts), and (3) the TM is a global coastline model and many readers from areas outside the North Atlantic will not know the flora analyzed in this report. In this paper, we will use the term (Atlantic) Subarctic Region as it is delimited in the TM: approximately the entire northern coast of the Gulf of St. Lawrence, including Anticosti, the Strait of Belle Isle area, the northeastern coast of Newfoundland, and the southern half of Labrador. Southwestern Greenland is 2011 W.H. Adey and L.-A.C. Hayek 11 also likely Subarctic, but we did not include this area in our study, nor is there significant literature on which we can draw. Subarctic species also extend, in decreasing abundance, northwards into the Arctic, southwards, into the Gulfs of St. Lawrence and Maine and throughout Nova Scotia, and to the east across the northern part of Iceland and Norway, with remnants extending to the British Isles. Most Subarctic genera and many species also occur in the North Pacific Subarctic Region. Occasionally, for emphasis, we refer to the “core” Subarctic, i.e., shorelines near the Strait of Belle Isle. Seaweed assemblages and plant-animal interactions in the Northwestern Atlantic rocky shore ecosystem A voluminous literature covers the community structure and ecology of the subtidal rocky benthos of the northwestern North Atlantic. However, much of it directly treats the mixed Subarctic/Boreal areas of the Gulf of Maine and Nova Scotia, (see e.g., the review and comparative works of: Mathieson et al. 1991; Steneck et al. 2002; and more recently—although directed primarily to the intertidal— Jenkins et al. 2008). Fewer studies deal with the Subarctic Region (the northern Gulf of St. Lawrence, northern Newfoundland, and southern Labrador), as defined by the TM. However, these studies are critical to our presentation (e.g., Gagnon et al. 2004; Himmelman 1985, 1991; Himmelman and Lavergne 1985; Himmelman et al. 1983; Keats 1986; Keats et al. 1985). This is by no means a small area of coast, being over twice as long by straight-line measure as the Gulf of Maine and Atlantic Nova Scotia combined. Kelps of the brown algal order Laminariales are the largest benthic organisms in the sea. When they occur extensively over wide areas, they are often referred to as forests. Although in many cases kelp forests are similar to their terrestrial counterparts— multistoried and diverse, with a wide array of associated invertebrates and fish—more often they are patchy and more like savannas. This more patchy character is particularly true of kelp communities in the northwestern North Atlantic, where the herbivorous Strongylocentrotus droebachiensis (Müller) (Green Sea Urchin) can be abundant and a critical ecological determinant (Scheibling 1986, Scheibling and Hatcher 2001, Scheibling and Stephenson 1984). However, the term “kelp forest” refers only to the middle/upper zone of a rocky sloping bottom bearing a three-or-more-tiered assemblage with depth. Typically, the larger kelps are the primary structuring elements, the canopy of a shallower zone in which an understory of small, mostly red, bushy algae overlies a crustose layer of mostly calcified coralline red algae. Both the kelps and the bushy red algae are often covered with brown and red, mostly filamentous, epiphytic algae. With depth, the canopy becomes patchy and then disappears, giving way to a low, bushy red zone. Finally, near the photic limit, if the bottom remains rocky, a broad zone of calcified crustose coralline algae encrusts much of the rocky surface, intermixed with sessile invertebrates. Shallower, there is typically a narrow sublittoral fringe zone in which the kelps give way to a mix of smaller red, brown, and green algae adapted to high wave energy, temperature extremes, and, in some places, moving ice in winter. In general, there is also a depth zonation of macroalgal species that is to some extent independent of assemblage morphology. 12 Northeastern Naturalist Vol. 18, Monograph No. 8 Mathieson et al. (1991) reviewed rocky shore ecology from Cape Cod to the Strait of Belle Isle, the traditional range of the western Atlantic Boreal. Most of the literature cited in that review refers to the Gulf of Maine and Nova Scotia, and some of the key papers that would have indicated the rather different community structure in Newfoundland and the northern Gulf of St. Lawrence were not cited or had yet to appear (e.g., Chapman and Johnson 1990; Himmelman 1985, 1991; Keats et al. 1985). A more recent review (Jenkins et al. 2008), cited a single paper of the key Subarctic literature, but failed to note the striking ecological differences of the Subarctic, apparently assuming that the Subarctic was simply a transition to the Arctic from a western Atlantic Boreal. The Corona Project (Coordinating Research on the North Atlantic), of which the Jenkins’ study was part, did not include a single scientist from the universities of the Subarctic Region (Cunningham 2008). Figure 4 is a schematic sketch from Mathiesen et al. (1991) showing the distribution of major rocky shore benthic organisms in the northwestern Atlantic Figure 4. Schematic diagram of the primary rocky bottom benthic macroalgae and key sedentary invertebrates as a function of wave exposure and depth in the northwestern North Atlantic (Gulf of Maine and Nova Scotia). Note: Only two species (Chondrus crispus [shallow] and Ptilota serrata [deep]) of the usually extensive subtidal bushy red algal zone have been included. Modified from Mathieson et al. (1991). 2011 W.H. Adey and L.-A.C. Hayek 13 in relation to exposure and depth. In the broadest sense, it is a graphical representation of the key elements of rocky shore community structure in Atlantic Nova Scotia and the Gulf of Maine in situations (and times) when the grazing S. droebachiensis is of moderate abundance; the lowest profile macroalga, coralline algae, are omitted for clarity. As described by Mathiesen et al. (1991), there is an alternate state of the subtidal, especially in the mid- to deeper zones. In this “overgrazed “state, patches (sometimes large areas) exist in which there are very few fleshy algae. Here, abundant sea urchins cover an expansive but “checkered” layer of crustose corallines. Figure 5 is a similar representation of benthic community structure around the western, southern, and southeastern shores of the island of Newfoundland, as published by Himmelman (1985). Another diagram, in perspective view, for a locality in the north-central Gulf of St. Lawrence is shown in Figure 6 (Himmelman 1991). In additional studies in the outer St. Lawrence estuary, Himmelman et al. (1983) and Gagnon et al. (2004), describe how when sea urchins are repeatedly removed from shallow, denuded areas of rocky bottom that have become sea urchin/coralline “barrens”, Alaria esculenta (L.) Greville, the dominant shallow-water kelp from the northern Gulf of St. Lawrence through Newfoundland establishes a “forest”. Keats et al. (1985) also showed that repeated periodic pack-ice destruction can lead to the temporary and usually partial removal of Alaria, allowing colonization Figure 5. Principal macroalgal distribution and Strongylocentrotus droebachiensis relative abundance as a function of wave exposure and depth on rocky shores on the west and south coasts of Newfoundland. (From Himmelman 1985). 14 Northeastern Naturalist Vol. 18, Monograph No. 8 Figure 6. Sketch of the distribution of dominant macroalgae and invertebrates on a rocky exposed shore of the Mingan Islands in the northcentral Gulf of St. Lawrence. (From Himmelman 1991). 2011 W.H. Adey and L.-A.C. Hayek 15 by other algal species. The replacement species are red and brown annuals, all common species on these shores. The only additional kelp species, Laminaria digitata (Hudson) J.V. Lamouroux, remained at about 1.5% of total biomass with or without ice scouring. Repeated ice destruction reduces the almost monocultural level of dominance of A. esculenta. However, seaward of protected bay shores, Alaria remains a key macroalgal species on these coasts, and given two to three years without ice scour, it typically becomes highly dominant in the upper 5–8 m of the rocky subtidal. On very exposed shores, it can occur as patches to as much as 20 m, especially where a rich savanna of Agarum is developed; below 8–10 m, the Alaria patches typically occur in the middle of larger Agarum patches, where it receives protection from the sea urchins (Gagnon et al. 2005). All of the diagrams that we have presented are idealized to the extent that rocky bottom is shown as occurring at all depths and all exposures. As we shall show, this can occur only on very steep slopes or areas of current scour; in general, the depth limit of rocky bottom is reduced, with less exposure due to the build-up of sediments that bury the rock surface. In this paper, we will describe stations from the northeastern Gulf of St. Lawrence, southern Labrador, and northeastern Newfoundland that were analyzed by harvest of macroalgal biomass in quadrats, and underwater digital images. These more northerly and colder water stations will be compared with similarly occupied stations in the Gulf of Maine and Nova Scotia. We will show that with minor variations, the primary seaweed assemblages depicted in Figures 5 and 6 are present over a broad area of coast, and statistically present highly significant differences from that of the Gulf of Maine and Nova Scotia (see Fig. 4). Our research shows that this extensive coastal area matches what we have shown in the TM to be the North Atlantic Subarctic. Previously, this region had been treated as an area of transition to the Arctic, without a distinctive character. The complex interactions within the northwestern Atlantic rocky shore ecosystem, including the algal assemblages, their invertebrate grazers and invertebrate predators, are well-studied (e.g., see the review articles of Bartsch et al. 2008, Chapman and Johnson 1990, Jenkins et al. 2008, Steneck et al. 2002, Vadas and Elner 1992). A hierarchy of seaweed competitive capabilities is partially established, and these are correlated with the effects of grazers and predators. As many authors have discussed—and as we show graphically in this paper—most seaweed species have characteristic depth distributions that are also a function of wave exposure. These basic patterns can be modified by the presence of other seaweed species, but it is the grazers that create the dominant structural effects, and in the subtidal of the rocky shores under discussion, the single dominant grazer is S. droebachiensis (see review by Scheibling and Hatcher 2001). Strongylocentrotus droebachiensis. The Green Sea Urchin is an often-abundant species on most shallow rocky shores in the northwestern North Atlantic (Fig.7). Strongylocentrotus droebachiensis is a member of the North Pacific family of echinoids, the Strongylocentrotidae, and is circumboreal in the North Atlantic and North Pacific. The species has been noted through the Canadian Arctic Archipelago, but apparently it is not common in the Arctic (Wilce 1994). It is the only 16 Northeastern Naturalist Vol. 18, Monograph No. 8 member of the family to occur outside the North Pacific. Quite likely, just as is true of the dominant seaweeds in the colder northwestern North Atlantic (e.g., Adey et al. 2008, Coyer et al. 2006), this grazer evolved from an ancestor in the North Pacific, having passed through the Arctic to the North Atlantic during the warmer periods of the Pleistocene. Recent large-scale genetic work by Addison and Hart (2005) suggests that some isolation in northwestern Atlantic refugia during glaciations does occur, but, on a Pleistocene time scale, both sporadic gene flows to the North Atlantic from the North Pacific as well as lesser back-crossing to the Pacific from the North Atlantic has characterized this species. Although Strongylocentrotus droebachiensis can be found to 300 m depth on many bottom types, it is most common on rocky bottoms at less than 50 m. This echinoid is typically a grazer of brown seaweeds in the northwestern Atlantic, including Alaria, Laminaria, Saccharina, and Chordaria, where they are available, but it will consume some red algae (e.g., Chondrus and Corallina) if the preferred foods are unavailable (Scheibling and Hatcher 2001). Invertebrates are often consumed incidentally as these echinoids graze their preferred algal foods; lacking algal foods, S. droebachiensis will consume larger invertebrates (e.g., anemones) and can even become cannibalistic. Nevertheless, S. droebachiensis will actively avoid some common algal species of northern Atlantic regions Figure 7. Moderate density Strongylocentrotus droebachiensis on a sea urchin/coralline patch at 14 m depth, at Cape Daumalen, near Canada Bay, NF (eastern side of Great Northern Peninsula). This is an open area, showing the underlying coralline, in a broader area that is greater than 50% Agarum (upper left). Crossaster papposus (Spiny Sunstar), is about 15 cm in diameter. Crossaster papposus feeds on both sea urchins and other sea stars and is a primary predator on S. droebachiensis (see Himmelman1991). The raised mound-like corallines are Clathromorphum compactum, with much of the intervening crust Clathromorphum circumscriptum; the branching corallines around the lower boulder margin are Lithothamnion glaciale Kjellman. 2011 W.H. Adey and L.-A.C. Hayek 17 (e.g., Agarum, Desmarestia, Ptilota, and Phycodrys), presumably because they are protected by distasteful or toxic chemical compounds or possess physically difficult characteristics (Himmelman and Nedelec 1990, Scheibling and Hatcher 2001). The basis of such avoidance is not always understood. For example Palmaria palmata (L.) (Dulse) is consumed by humans, but it contains kainic acid and is sometimes utilized to treat intestinal worms (www.Algaebase.org, accessed 20 September 2010). S. droebachiensis is not attracted to P. palmata, but will consume it readily when it is the only alga available (Himmelman and Nedelec 1990). Strongylocentrotus droebachiensis at moderate density, but with insufficient algal food, can survive for long periods with little growth; they can even become reproductive while on overgrazed coralline bottoms. As described by Himmelman (1986), high densities of young sea urchins, at 8–11 mm test diameters, can accumulate year classes and maintain themselves with little further growth on coralline substrate. These “reservoir” populations, partly protected (in crevices) from wave action and predators, find sufficient food on coralline surfaces (including settling diatoms and other microalgae). Given a window of opportunity (e.g., storm removal of large urchins at the front followed by rapid advancement downward of the Alaria belt), these urchins can become mobile and grow fast. Strongylocentrotus droebachiensis often aggregate in “fronts” or lines, and can intensively graze into the margins of kelp beds, reducing the kelps to stubble or “barren grounds”. Sea urchin grazing is inhibited by wave action in exposed, shallow waters. Exposed animals can be torn off the bottom and carried by wave action and currents to gravel bottoms, where they can be damaged, or buried in sand or silt. They are known to move to deeper waters in winter, thereby avoiding the more destructive wave action common at that time of the year. The feeding behavior of this species has evolved to seek preferred algal species, enhance nutrition, and optimize reproduction (Vadas 1977); however, this feeding behavior is modified by a seasonal rhythm to migrate. After three or more years, gonad development in the young S. droebachiensis begins in spring, and expands through the summer and fall; release of eggs and sperm into the water column occurs in coordinated pulses the following spring (Gaudette et al. 2006, Himmelman et al. 2008). If an urchin remains feeding on a coralline bottom, maturation may be delayed for several years. Fertilized in the water, the swimming larvae remain in the water column, feeding on plankton for one to five months, and become widely dispersed. Settlement occurs in summer, and recruitment appears to be favored on coralline rubble and “barren” bottoms, rather than into a kelp forest (Balch and Scheibling 2002). Larvae of S. droebachiensis are eaten by a wide variety of invertebrates and larval fish while in the water column. After settlement, they become cryptic to avoid predation, but crabs, lobsters, asteroids, and some fish all feed on small urchins. However, it is noteworthy that sea urchins are not typically found in cod stomachs; for example, Langton and Bowman (1978), showed that Gadus morhua L. (Atlantic Cod) ate mostly other fish (64%), crustaceans (21%), and mollusks (8%), with echinoderms (not just urchins) comprising only 1–2% of their diet. 18 Northeastern Naturalist Vol. 18, Monograph No. 8 Some studies found a greater array of small sea urchin predators in kelp beds than on coralline “barrens”, suggesting survival value in settling on “barrens” (McNaught 1999). Once reaching larger size (>30–50 mm), sea urchins may escape from many predators, although eider ducks and sea gulls likely remain serious predators in shallow water (Bustnes and Lonne 1995); asteroids, especially Crossaster papposus (L.) (Spiny Sunstar), also prey on S. droebachiensis (Himmelman 1991). Some authors have related declining fish stocks to lesser predation on S. droebachiensis, creating the widespread occurrence of “coralline barrens” in the northwest Atlantic (Steneck et al. 2002). Others feel that the evidence from cod stomachs suggests that this is unlikely (e.g., Langton and Bowman 1978; Miller 1985b). Some fish, such as Anarhichas lupus L. (Wolffish) (Keats et al. 1986) are well adapted to eat sea urchins, crabs, and molluscs. However, Wolffish have never been more than incidental by-catch in traditional inshore northwestern Atlantic fisheries. Cancrid crabs, radiating from the North Pacific, including Cancer irroratus (Rock Crab), occur throughout the core Subarctic and may be important urchin predators (Steneck et al. 2002). However, these crabs are only minor components of small, local fisheries. In Nova Scotia and Gulf of Maine waters, a pathogenic amoeba (Paramoeba invadens) can cause significant, periodic mortality of S. droebachiensis (Scheibling and Hatcher 2001). A number of such events were documented over the last several decades, and they were associated with periods of high water temperature and, perhaps, pulses of Gulf Stream water. In Nova Scotia and the Gulf of Maine, sea urchin population “explosions” and patchy, parasite-induced crashes have been endemic at least for several decades, but this process was modified by extensive commercial urchin harvest for roe during the 1990s (Steneck et al. 2008). The parasite cannot survive unless winter water temperatures are above 2–4 °C, and the disease in sea urchins is arrested at summer temperatures below 10–12 °C (no survival of sea urchins at 16 °C, 20% at 12 °C, and 100% survival at 8 °C; Scheibling and Stephenson 1984). Most of the Subarctic coast has summer and winter water temperatures that are below the survival range for P. invadens, and it is unlikely that water temperatures in the region during the last several centuries could support this parasite (e.g., Schiebling and Hatcher 2001, for Newfoundland). The macroalgae. On rocky, subtidal bottoms in the western North Atlantic, four primary seaweed zones can be distinguished with depth: (1) a shallow and narrow sublittoral fringe zone of mid-sized red and brown algae, including many annuals; (2) the “kelp forest”, often with rather distinct upper and lower subzones; (3) a deeper bushy red algal zone; and (4) the deepest, a crustose coralline zone. The bottom two zones extend upwards, under the kelp forest, although the species proportions shift considerably. The infralittoral zone, the band between low-water neap and spring tides and lying between the intertidal and the subtidal, can also be quite distinctive, although it typically grades into the sublittoral fringe (0 to 1–2 m depth). The height of the infralittoral zone relative to sea level can vary depending upon exposure and slope of the shore, but it is usually quite narrow as compared to the subtidal zones. 2011 W.H. Adey and L.-A.C. Hayek 19 The Laminariales, commonly called the kelps, provide the canopy and primary structuring elements of kelp forests. This is a diverse group of approximately 30 genera and 100 species, that are considered to comprise four families (Alariaceae, Laminariaceae, Lessoniaceae, and Costariaceae; see Lane et al. 2006). The Laminariales are most speciose in the North Pacific (Adey et al. 2008, Lane et al. 2006). Fewer than 10% of the total taxa have migrated through the Arctic to the North Atlantic since the opening of the Bering Strait (see Adey et al. 2008). In the northwestern North Atlantic, only five species, belonging to four genera, of this order occur: Alaria esculenta, Laminaria digitata, Saccharina latissima (L.) J.V. Lamouroux, Saccharina longicruris (Bachelot de la Pylaie) Kuntze, and Agarum clathratum Dumortier). A sixth kelp of somewhat uncertain relationship, Saccorhiza dermatodea (Bachelot de la Pylaie) Areschoug, is sometimes placed in the Phyllariaceae, a separate family of the Laminariales, and sometimes in the family Laminariaceae (no member of the Lessoniaceae occurs in the North Atlantic, and most species of that family are located in the South Pacific). Adult kelps are diploid sporophytes; the alternate haploid, sexual generation is typically a minute, filamentous gametophyte, resembling the smallest filamentous brown alga. The male gametophytes bear antheridia that produce flagellated sperm. The female gametophyte develops a sedentary oogonium that produces an egg, which in turn germinates in situ into a sporophyte when fertilized. Alaria esculenta — This species (including A. pylaii and A. grandifolia; see Kraan et al. 2000) is circumboreal in distribution, usually in shallow, wavebeaten situations, where locally it can be a virtual monoculture (Fig. 8a, b). All of the 11 known sister species occur in the North Pacific, thus following a welltrodden pattern for many Arctic and Subarctic species (Adey et al. 2008). Alaria esculenta is also uniquely characterized by an orange/brown color with a strong mid-rib and thin, ruffled wings. As with most members of the family, mature plants bear sporophylls, as special side blades, on the lower stipe (Fig. 8b). In the Canadian Arctic Archipelago, Alaria esculenta tends to be deeper and more patchy in its occurrence, as well as larger on average (Wilce 1994); it reaches south to the Kurile Islands in the western Pacific, and apparently does not reach northernmost British Columbia in the eastern Pacific. In the North Atlantic, it reaches south to the western Gulf of Maine in the west and nearly to the English Channel in the east. Clearly, the species extends further south in the North Atlantic, and to higher temperatures (approx. 17 °C summer) than in the North Pacific (approximately 8–12 °C). As we will show, on exposed shores in the western Atlantic Subarctic, the kelp forest becomes largely two-zoned: Alaria esculenta reaches from the infralittoral to 5–10 m, and Agarum clathratum extends from that level to 15–20 m, with scattered individuals occurring through the red algal zone to 25–30 m. As in the Arctic, Alaria esculenta can occur in patches in deeper water; below 10 m, it tends to be restricted to the centers of fields of Agarum clathratum, where it derives secondary protection from sea urchin grazing (see discussion below). In the Atlantic Subarctic, Alaria esculenta tends to reach maximum length (3–4 m) near the lower end of its depth band. These are likely older plants that have survived 20 Northeastern Naturalist Vol. 18, Monograph No. 8 the occasional ice scouring of the shallower subtidal. Shallower Alaria beds can be quite dense, in spite of occasional ice scour; Alaria appears to have the capability to recover quickly from such scour (Keats et al. 1985). In the Subarctic, the remaining five species of kelp occur as scattered individuals among the Alaria and Agarum, or occur in refugia, as we will describe later. To the south, in the Gulf of Maine and Nova Scotia, the picture changes; the shallow-water Alaria becomes more reduced in abundance and size, especially in Nova Scotia, and is replaced by two Saccharina species and Laminaria digitata. Saccharina species — Saccharina latissima (ex Laminaria saccharina) and Saccharina longicruris (ex Laminaria longicruris) (Figs. 9 and 10) are archetypical kelps that are abundant in the upper half of the kelp forest on moderate energy shores in both Nova Scotia and the Gulf of Maine. Saccharina latissima is a cosmopolitan temperate/boreal species that reaches south at least to Long Island Sound in the western North Atlantic and to Mauritania and the Mediterranean in the eastern North Atlantic. It apparently extends through the Arctic, reaching Japan and Korea in the western Pacific, and in the east has been reported at least to California (www.Algaebase.org, accessed September 2010). Saccharina longicruris is considerably more restricted in its distribution, likely occurring only in the western Atlantic and Canadian Arctic (Bartsch et al. 2008). Some workers have considered S. longicruris to be an ecological variant of S. latissima (Lane et al. 2006, Lindstrom 2001). They could be ecotypes, exhibiting Figure 8a. Dense Alaria esculenta forest at 2.5–4 m depth. The area shown is part of a virtual monoculture of this species from low-water spring tide to about 7 m depth at Partridge Point, 10 km south of Cape Bauld, northernmost Newfoundland. 2011 W.H. Adey and L.-A.C. Hayek 21 different morphology in different environments. However, mature, well-developed adults are quite different in morphology. Mature S. longicruris tends to be stipitate, with a long, hollow, upright, and stiff stipe, and a flattened, sometimes very broad trailing blade, whereas S. latissima tends to have a short, more flexible, and solid Figure 8b. 3-m long Alaria esculenta plant taken from 5–7 m depth at the above station. There is a band of these large, presumably multiyear plants at 5–7 m that is not reached by ice scour during most winters. Note the sporophylls attached to the stipe. 22 Northeastern Naturalist Vol. 18, Monograph No. 8 stipe with a ruffled blade. The two species can occur together and be quite distinct, although young plants are morphologically indistinguishable. Critically, S. longicruris is only known from the western Atlantic. This endemicity suggests that S. longicruris is currently speciating or has evolved relatively recently, during the later Pleistocene, as a result of periodic isolation from its likely parent, S. latissima. Probably, the two species can hybridize in the wild—this has been accomplished in the laboratory—resulting in spore-producing offspring that further confuse identification (Bartsch et al. 2008, Lüning 1990). In the Subarctic of Newfoundland, Labrador, and Quebec, the two Saccharina spp.are only occasionally seen in the kelp forests of exposed rocky shores; here, Figure 9. Saccharina latissima forest with scattered Laminaria digitata at 5–10 m depth in outermost Lunenburg Bay on the southern Atlantic Coast of Nova Scotia. The tufted species in the lower foreground is Ceramium rubrum. Figure 10 (opposite page). a. Saccharina longicruris growth on low cobble/boulder mound at 10–12 m depth in the inner end of Goshen Arm, New World Island, Newfoundland. This rocky mound lies in a silty basin, about 0.5–1 km from a rocky shore with abundant sea urchins, and yet is virtually free of those grazers. b. Mixed Saccharina latissima and S. longicruris patch on cobbles and boulders at 1–2 m depth in the outer part of Englee Harbour in northeastern Newfoundland. This patch receives reduced ocean swell but no significant sea from the open ocean, and it is surrounded by an extensive mobile coralline-covered shell/pebble bottom. Although only a few hundred meters from a rocky, sea urchin-dense shore, it is essentially free of Strongylocentrotus droebachiensis. The coralline on pebbles and Mytilis edulis shells in the foreground is mostly Clathromorphum circumscriptum, with small amounts of Lithothamnion glaciale. 2011 W.H. Adey and L.-A.C. Hayek 23 they are typically buried as individuals in the thick Alaria canopy, or occasionally deeper in the Agarum zone. At exposed stations, when the Alaria canopy is removed by ice, the Saccharina spp. are not among the short-term replacements (Keats et al. 1985). However, they can form patchy dense canopies in more protected waters with less competition from Alaria, but only when they are isolated 24 Northeastern Naturalist Vol. 18, Monograph No. 8 by mobile or soft substrate from Strongylocentrotus grazing fronts, (Fig. 10). On the other hand, in Nova Scotia and the Gulf of Maine, below the narrow, mixed Laminaria digitata/Alaria esculenta shallow zone, the two species often form dense beds both as virtual monocultures (Fig. 9) and as mixed populations. The stipitate morphology in Saccharina longicruris keeps the blade off the bottom, which could be protective on muddy bottoms (where the holdfasts are attached to boulders or other rock surfaces projecting from the muddy bottom). The same relationship could be selective on mobile pebble/shell bottoms with boulders. Laminaria digitata — As a short, thick-bodied, and digitate kelp, Laminaria digitata is very tough and strongly attached to the substrate. It dominates the shallow waters of most exposed shores in Nova Scotia (Fig. 11) and, to a lesser extent, the Gulf of Maine. It is more limited in its occurrence and is often replaced by Alaria in the core Subarctic; it does not reach the Arctic. Laminaria digitata is particularly abundant in the northern British Isles, where it competes with Laminaria hyperborea (Gunnerus) Foslie in the shallow sublittoral (Connor et al. 1997, Lüning 1990). It extends southward along the eastern Atlantic Coast at least to West Africa and the Canary Islands, and reaches north to Spitsbergen. Laminaria digitata is a North Atlantic species. However, its nearest relative is the Arctic Laminaria solidungula J. Agardh, and as Adey et al. (2008) show, the two species derive from North Pacific ancestors. Laminaria species are sprinkled down the colder and/or deeper parts of the Atlantic Ocean, essentially one species per biogeographic region; given appropriate temperature, nutrient, and light Figure 11. Large Laminaria digitata plants (2 m long) at 2.5 m depth in exposed, outer Lunenburg Bay, southeastern Nova Scotia. At this locality, L. digitata, dominant at 0.5 and 2.5 m, is replaced by Saccharina latissima by 5 and 10 m depth (Fig. 9). 2011 W.H. Adey and L.-A.C. Hayek 25 conditions at depths, kelps can move into tropical environments (Graham et al. 2007). This is unlike the situation in the North Pacific, where numerous species overlap geographically and are presumably specialized to narrow ecological niches. This pattern suggests rapid evolution of new Laminaria species once the transit of the North Pacific through the Arctic to the North Atlantic was achieved. Only in the case of the eastern Atlantic Boreal/Lusitanean, where biogeographic regions are close, is there considerable species overlap (L. digitata, L. hyperborea, L. ochroleuca, north to south; Lüning 1990). Agarum clathratum — One of the most widespread minor kelp species in Nova Scotia and the Gulf of Maine, Agarum clathratum tends to occur as scattered plants in the lower half of the kelp forest and deeper, in the red algal zone (Fig. 12). However, in the lower half of the kelp forest in the Subarctic, it often becomes the dominant seaweed (see Figs. 6 and 13). While sometimes extensive, densely covering large areas of rocky bottom (a kelp forest), more typically it is patchy, separated by open coralline patches, and forming more of a kelp savanna. Placed in the family Costariaceae (Lane et al. 2006), which is entirely North Pacific in its distribution except for Agarum clathratum, this species occurs across the North Pacific from Hokkaido and the Okhotsk Sea to Alaska and British Columbia. It also occurs through the Canadian Arctic and down the North American Atlantic Coast to the Gulf of Maine. Agarum clathratum does not extend to Europe, including Iceland, nor does it occur on the northern Russian Figure 12. Lone Agarum clathratum plant at 22 m on West Cod Ledge off Cape Elizabeth, ME. At this exposed station, this species is rapidly reduced in abundance below 20 m depth, occurring as scattered plants among bushy small reds (Ptilota serrata right; and Phyllophora pseudoceranoides left) and encrusting invertebrates and corallines. The depth limit of seaweeds at this station is about 25 m; rocky bottom continues to depths greater than 30 m. 26 Northeastern Naturalist Vol. 18, Monograph No. 8 coast. The species may have co-evolved with S. droebachiensis, and one could make the argument that it could only survive, at least as a kelp forest, where S. droebachiensis is present. The S. droebachiensis does extend well beyond the range of A. clathratum in the eastern Atlantic, although at low abundance. A rather small kelp with a distinctive mid-rib and perforated blade (Fig. 12), Agarum clathratum is characterized by the ability to survive with abundant S. droebachiensis, apparently without the assistance of physical supporting factors (e.g., wave action). It has been suggested that A. clathratum can only survive with Laminaria (Saccharina) plants available to provide preferred food for sea urchins, but this would not seem to be the case in the Subarctic (Keats et al. 1982). As Gagnon et al. (2005) show, stands of Agarum clathratum are able to not only survive and expand against grazing “fronts” of urchins, but provide shelter and survival to young, poorly protected A. clathratum and sometimes to deeper patches of Alaria esculenta. Adult A. clathratum have a very tough and leathery cortex, but they also are protected by toxic chemicals, perhaps phenolics (Gagnon et al. 2005, Himmelman and Nedelec 1990). Winter sporulation and settlement leads to the formation of small filamentous gametophytes; these in turn undergo sexual reproduction, with the oogonia remaining attached to the gametophytic filament, resulting in the formation of juvenile plants in situ in the early spring. This cycle is likely timed to reduce predation on the most sensitive stages of the life cycle; as we noted above, S. droebachiensis retreats to deeper water or becomes highly cryptic in winter. Figure 13. Dense and extensive bed of Agarum clathratum at 20 m depth, on exposed Partridge Point, south of Cape Bauld in northernmost Newfoundland. 2011 W.H. Adey and L.-A.C. Hayek 27 Studies to date suggest that Agarum clathratum invests considerable energy into constructing a tough, but weakly productive thallus (Mann 1973, Vadas 1968). The toxic compounds that reduce its competitiveness (i.e., growth rate) compared to other kelps allow it to coexist in adult stands with abundant Strongylocentrotus droebachiensis. Physical factors, waves, and mobile substrate can reduce cover, and while unprotected juveniles rarely escape grazing when sea urchins are abundant, there is evidence that the large, bushy, and chemically protected brown alga Desmarestia viridis (O.F. Müller) J.V. Lamouroux can provide shelter to young Agarum clathratum plants and lead to new kelp bed formation (Gagnon et al. 2003). Adult stands of A. clathratum can slowly expand and coalesce, in the face of intensive sea urchin grazing, eventually occupying large areas (Figs. 6 and 13). As Gagnon et al. (2004) have demonstrated, stands of Agarum clathratum can limit shoreward movement of sea urchin fronts. The lower boundary of the deep kelp bed is not sharp, but rather grades into the bushy red algal band with only scattered individuals of A. clathratum. Ptilota serrata — Many species of small, bushy red algae can be found in the understory beneath the dominating Agarum of the lower half of the northwestern Atlantic kelp forest. However, when urchin grazing is intense, only a single species, Ptilota serrata Kützing, is abundant. It is avoided by S. droebachiensis (Gagnon et al. 2005, Himmelman and Nedelec 1990), and probably harbors toxic or distasteful compounds. Ptilota serrata is typically a dark red, coarsely bushy alga of less than 20 cm in length. Often abundantly branched, the branches tend to be flattened, basically pinnate in structure, with the lateral branchlets having incurved tips and being serrate on the lower margins (Fig. 14). It is easily recognized while diving and in digital photographs. Ptilota serrata is circumboreal, but rather restricted to colder waters, reaching south to northernmost Japan and the Okhotsk Sea (Lüning 1990) and passing through the Canadian Arctic (Wilce 1994) to the Gulf of Maine, where it can “form an almost uniform, monospecific turf at the extinction depth of foliose algae” (Sears 2002). In the eastern North Atlantic, it occurs in Iceland, the Faeroes, and northern Norway, but not in the British Isles (www.Algaebase.org, accessed 20 September 2010). Reproducing year round, even in its colder water range, Ptilota serrata tends to be a plant of deeper water (Wilce 1994), and as reported by Norall et al. (1981), it reaches its maximum size and reproductive effort below 18–24 m. Based on Choi et al. (2008), and the species listing of www.Algaebase.org (accessed 20 September 2010) of four listed close relatives of P. serrata, three occur in the Pacific Ocean and one in the eastern North Atlantic. As Adey et al. (2008) showed for Arctic endemic seaweeds, this colder Subarctic genus also appears to be based in the North Pacific. Chondrus crispus — In parts of its range, especially eastern Nova Scotia and to a lesser extent the northeastern coast of Maine, Chondrus crispus Stackhouse (Irish Moss) can be a virtual monoculture in its narrow, lowest intertidal and infralittoral to fringe sublittoral zone (Fig. 15). Although the species can extend to deeper water as isolated individuals, in the large tidal range of the eastern Gulf of Maine and southwestern-most Nova Scotia, particularly on large expanses of smooth 28 Northeastern Naturalist Vol. 18, Monograph No. 8 granitic rock, it can form extensive beds distinctly visible at low water spring tides (Scrosati et al. 1994). On steeper rock faces, especially at the upper margin of the Chondrus zone, C. crispus can be intermixed with or even dominated by Mastocarpus stellatus (Stackhouse) Guiry (Dudgeon et al. 1990, 1999). The mature, individual C. crispus plant has a cartilaginous brownish-red to sometimes greenish-white thallus (depending upon temperature and nutrient availability), is less than 20 cm in length and more typically 10–15 cm, with a 2–5-mm-diameter rounded stalk making up half of the total height of most plants. The crown of a C. crispus plant is repeatedly expanded by dichotomous branching, producing a flattened, irregular blade-like morphology. Most Rhodophyta have an anatomical structure that is based on filamentous building blocks, but unlike Ptilota, or which a dissecting microscope can easily distinguish this pattern, Chondrus is pseudoparenchymatous in structure. Chondrus crispus has morphologically similar alternating generations of asexual plants (spore-forming) and sexual plants (producing eggs [carpogonia] and sperm [spermatia]). Sexual reproduction occurs in summer, while tetraspore formation and release occurs in fall and winter. Chondrus crispus extends north/south over the entire Boreal/Lusitanean European Coast (Lüning 1990); it reaches from south Iceland and the entire Figure 14. Patch of Ptilota serrata at 25 m among Agarum clathratum drift fragments caught by Strongylocentrotus droebachiensis. There is a small, attached young plant of A. clathratum in the upper right and a patch of Phycodrys riggi in the lower right. Note the brittle stars (Ophiopholis spp.) hiding under the Ptilota. Noddy Point, northern Peninsula of Newfoundland in the Strait of Belle Isle. 2011 W.H. Adey and L.-A.C. Hayek 29 Norwegian coast, ranging south to Portugal, the Canary Islands, and the Azores. It is also quite abundant on rocky shores in the warmer western Atlantic from Nova Scotia and Prince Edward Island to Long Island, where, it can form a distinct and dense band, several meters wide, near low-water spring tides. However, Wilce (1994) and Lee (1980) do not report it from the Canadian Arctic, and in their extensive collection effort at Makkovik Bay in mid-Labrador, near the Subarctic/ Arctic transition, Hooper and Whittick (1984) found neither this species nor its associate Mastocarpus stellatus. In their study of the western North Atlantic rocky intertidal, Adey and Hayek (2005) occupied 50 stations that included an examination of the critical infralittoral zone in the Newfoundland, Quebec, and southern Labrador region. Re-analyzing the C. crispus data from Adey and Hayek (2005) produces the following: (1) For 10 stations from SW to SE Newfoundland, only two stations had as much as 50% biomass of C. crispus in the lower intertidal and infralittoral; two stations had no C. crispus, and the remaining stations had only a few individuals or scattered patches with 5–10% of the total biomass in quadrats; (2) In northeastern Newfoundland, from Trinity Bay to Canada Bay, of 14 stations, 5 failed to provide any C. crispus, 7 provided only a few scattered individuals, and 2 stations provided small patches but no more than 15% of quadrat biomass; (3) Finally, for the roughly 2000-km stretch from the outer St. Lawrence Estuary along the north shore of the Gulf of St. Lawrence to Battle Harbour, Labrador, Figure 15. Chondrus crispus on exposed Bates Island ledges off Casco Bay, ME. In both the Gulf of Maine and southeastern Nova Scotia, flat-lying ledges near mean low-water spring tides usually develop dense beds of Chondrus crispus. 30 Northeastern Naturalist Vol. 18, Monograph No. 8 including the Northern Peninsula of Newfoundland from Englee to Quirpon, 26 stations produced not a single C. crispus plant (or its associate Mastocarpus stellatus) either in the quadrats or in the search and description reconnaissance that preceded the quadrat harvests. Chondrus crispus was recorded to occur everywhere on the coasts of Newfoundland (Lüning 1990, South and Hooper 1980). However, there is a sharp boundary between moderate abundance and virtual absence that extends roughly east/west across the Island of Newfoundland and the middle of the Gulf of St. Lawrence (with a diversion northward on the west coast of NF to the Strait of Belle Isle). This boundary corresponds closely with the boundary of the Subarctic Thermogeographic Region, as shown in Figure 2, and the Subarctic/Boreal transition zone. Chondrus crispus is a member of a large genus (10 species), most of the members of which are located in the boreal to tropical North Pacific (www.Algaebase. org., accessed 20 September 2010). A molecular analysis by Hommersand et al. (1994) placed four Japanese Chondrus species in a single clade, with C. crispus (specimen from Ireland) as the next outlier. This Chondrus clade is most closely related to the genus Sarcothelia (2 spp.) from South Africa and New Zealand and Mazzaella (14 spp.), widespread from the west coast of the United States south to Chile and South Africa. Hommersand et al. (2004) suggest that C. crispus migrated through the Arctic prior to the Pliocene/Pleistocene cooling. However, this would have required transit of the Canadian Archipelago during the early opening of the Strait from approximately 4.8–7.4 Ma, where there is no other evidence for Pacific to Atlantic transit (Greene et al. 2008) and some evidence for a reverse passage. Another scenario is that C. crispus passed through the Arctic during one of the later Pleistocene’s “super” interglacials. Considering the failure of C. crispus to penetrate significantly into the core western Atlantic Subarctic Region, that also seems unlikely. The presence of Chondrus crispus in the North Atlantic is a puzzle, and potential human transport from Asia, perhaps via Europe in the 15th and 16th centuries, needs to be carefully examined by genetic methods. Chondrus crispus is a strong candidate for ballast stone transport. Crustose corallines — Underlying kelp forest, kelp savanna, and the deeper, bushy, red algal assemblage, much of the shallow rock bottom in the northwestern North Atlantic is coated with mostly living crustose coralline carbonate. This deposit forms a reef-like biostrome, typically 5–15 mm thick, but sometimes to a thickness of 10–20 cm (Fig. 16). Due to the boring of clams and worms, these coralline biostromes are often porous and provide habitat for a wide variety of invertebrate infauna. Nearing the photic limit of Agarum clathratum, Ptilota serrata, and other small red seaweeds, and within the open spaces of kelp savanna, the corallines typically provide the dominant photosynthetic surface. Even though it is accepted that Strogylocentrotus droebachiensis can survive on coralline bottom for long periods (Himmelman 1986) and may recruit to it, the coralline “barrens” have been widely accepted as the overgrazed or alternate state of a more “normal” kelp forest stage (Mathiesen et al. 1991, Steneck et al. 2002). However, as measured by analysis of the temperature-controlled, yearly cycling of Mg2+ in the high-Mg Calcite 2011 W.H. Adey and L.-A.C. Hayek 31 Figure 16. a. Boulder with a 10–30 mm thick coralline biostrome (cor-strome); 20 m at Three Island, off Shoal Bay, in southern Labrador. This cor-strome is mostly a mixture of Clathromorphm compactum (lower and middle right) and Lithothamnion lemoineae (middle left). The large C. compactum area in the middle/right has been patchily grazed by Strongylocentrotus droebachiensis. A small patch of Lithothamnion glaciale occurs in the lower middle of the image. b. Junction between cobble/boulder slope and a silty bay bottom at 5 m in Port Marnham of St. Lewis Sound, southern Labrador. The “heads” of Lithothamnium glaciale attached to the cobbles break off due to the boring of worms and clams. They fall to the silty/shelly bottom where they are kept in motion by bioturbation and perhaps occasional strong waves, becoming rounded balls (maerl or rhodolites). The sea star Leptasterias polaris is an abundant predator of the bivalves Mya, Mytilis, and Hiatella on these bottoms (Himmelman 1991). 32 Northeastern Naturalist Vol. 18, Monograph No. 8 of Clathromorphum compactum (Kjellman) Foslie, one of the most abundant corallines on coralline bottoms, these crusts show typically 100 years, and in many cases over 200 years, of continuous growth (Halfar et al. 2011). These mostly Agarum kelp savannas, with their open coralline patches, are likely quite stable in time. Figure 17 provides a geographic re-plotting of the coralline cover information of Adey and Steneck (2001) from their Figures 8–10. The zones of greater than 89% Subarctic (blue hatching) and greater than 90% Boreal (Celtic) (red hatching) are shown along with the maximum southerly extent of Arctic and northerly extent of Boreal coralline species. In the Subarctic northwestern Atlantic, on more exposed shores, Clathromorphum circumscriptum (Strömfelt) Foslie and Lithothamnion lemoineae Adey dominates at less than 5 m, with Clathromorphum compactum at mid-depths and Leptophytum leave (Strömfelt) Adey and Lithothamnion glaciale at greater than 25 m, especially on pebble bottoms. In more protected waters, Lithothamnion glaciale is often the most abundant single coralline over an entire transect from low tide to the photic limit; however, it typically has a very strong peak from 15–25 m. These are all Subarctic species; except for the northwestern Atlantic endemic Lithothamnion lemoineae, they all reach through the Arctic and into the North Pacific Subarctic (Adey and Steneck 2001). Figure 17. North Atlantic coasts with greater than 90% Subarctic coralline species (blue hachure) and 90% Boreal coralline species (red hachure). The blue line is southern limit of Subarctic corallines. The black line is the approximate position of the southern limit of Arctic corallines; these plants occur mostly below 20 m depth. · = scuba station. 2011 W.H. Adey and L.-A.C. Hayek 33 In the 1960s, the sublittoral Boreal corallines that extended to the western Atlantic, primarily Phymatolithon lamii (= P. rugulosum), Phymatolithon laevigatum, and Lithophyllum crouani (= L. orbiculatum) provided roughly 25–35% of the coralline cover in the Gulf of Maine and Nova Scotia, where they mostly occurred under the kelp forest (Adey 1970a). These species did reach into the core Subarctic, as figures 3 and 17 show, but at a cover of less than 5%. The Arctic coralline species were also present in the Subarctic Core and occur nearly to the Subarctic limit in the western Atlantic (Adey and Steneck 2001, Adey et al. 2005). However, these are deeper-water species that are “plunging under” the overlying warmer layers. Only one of the Arctic species, the deeper water Leptophytum foecundum (Kjellman) Adey was found significantly beyond the Subarctic core, and cover percentages were generally less than 2–3%. The two deeper-water Subarctic species Leptophytum laeve and Lithothamnion glaciale occur in the northeastern Atlantic, where they decrease in cover from 13% in the Shetland Islands to zero at the English Channel (Adey and Adey 1973). Arctic species reach their maximum southerly extent in northeasternmost Iceland and North Norway. This range limit is to be expected; unlike in the western Atlantic, strong, shallow thermoclines with underlying cold water, are not present. The Norwegian fjords (and probably some Scottish sea lochs as well) can have a sharp thermoclines, and that is responsible for the extension of both Subarctic and Arctic species southward within the more northerly fjords (Adey 1971). However, even in this case, at least in southern Norway, surface temperatures are still relatively high (15–20 °C), and the thermocline does not drop to Arctic levels within the limited photic zone of the fjords. Coastal waters in the British Isles range from 12–16 °C in summer and 7–12 °C in winter at the surface, and typically only decrease or rise a few degrees to 30–40 m depth. The calcified, crustose corallines cover a large percentage of rocky and larger pebble/shell bottoms in the western North Atlantic. Some Lithothamnion and Clathromorphum species growing on rock can become many centimeters thick; eventually, they are “bored-out” by molluscs and worms, and the crusts break off and fall to deeper water. At a critical wave energy—which is reached at moderate depths (2–10 m) in mid-energy zones and deeper (20–40 m) on exposed coasts— pebbles, shells, and coralline fragments occasionally turn over and thus can become coated with coralline around their entire circumference. Free coralline fragments or coralline-coated objects can form extensive fields of branched maerl or spherical rhodolith bottoms. Under suitable conditions (a balance between stability and constant motion), the maerl, either the Subarctic Lithothamnion glaciale (Fig. 16b) or the Arctic Lithothamnion tophiforme in deeper water (Adey et al. 2005), or coralline pebble bottom with Clathromorphum spp. (Fig.10b) will lie at that rock/sediment interface, with its extent depending upon the slope of the bottom. Although we will discuss maerl/pebble bottoms again later when we consider refugia for kelp, our focus in this paper is on rocky bottoms. Physical ecological factors Almost all seaweeds in colder waters are limited to rocky bottoms. At a depth that is a function of wave (or tidal current) energy and local sediment availability, 34 Northeastern Naturalist Vol. 18, Monograph No. 8 rocky bottoms transition into sedimentary bottoms (Fig. 18). In bay environments, this transition typically occurs at a few meters to 10–20 m, although at bay heads, rocky bottoms may be limited to the uppermost intertidal, with mud “flats” rising from the sublittoral and well into the intertidal. On the open coast, this transition usually lies in the 20–40 m range unless a particularly large quantity of glacial sand is locally available. Figure 18. Sublittoral bottom types (rock, sand/shell/maerl, mud) as a function of depth and exposure on the Maine Coast. In Newfoundland/Labrador and in southern Nova Scotia, the exposed side of the diagram is typically stretched 5–10 m as the supply of gravel, sand, and silt from till deposits is considerably reduced from that in Maine. After Adey and Loveland (1998). 2011 W.H. Adey and L.-A.C. Hayek 35 In situations where steep rock faces and wave or current energy keep rock surfaces clear of sediment, light can be depth-limiting for seaweeds. Vadas and Steneck (1988) describe an isolated rock pinnacle of very high water clarity in the middle of the Gulf of Maine. At that site, photic limits are approximately 30 m for Laminaria (Saccharina) kelps, 40 m for Agarum and bushy red algae, and 55 m for coralline crusts. The equivalent light limits on exposed parts of the central Maine mainland shore are 12 m, 18 m, and 25 m, respectively (W.H. Adey, pers. observ.). Water clarity provides photic limits on exposed shores in Atlantic Nova Scotia and Newfoundland/Labrador that are intermediate between these two situations (W.H. Adey, pers. observ.). As we have shown above, rocky shore assemblages of seaweeds are also different depending on exposure to wave action (e.g., Figs. 4, 5; Mathiesen et al. 1991, Himmelman 1985). Some seaweeds are poorly adapted to withstanding high levels of wave action. Other species invest heavily in structures and morphology that enable them to occupy rocky zones of intensive wave stress. The primary seaweed grazer in the northwestern Atlantic (Strongylocentrotus droebachiensis) has a limited ability to withstand intensive wave action and is thus restricted to deeper water (>5–10 m) on exposed shores (W.H. Adey, pers. observ.). Materials and Methods Station locations During summers from 2000 to 2002, the rocky shore from Cape Cod (in the southern Gulf of Maine) to Lewis Sound (north of Battle Harbour) in southern Labrador and across the northern shore of the Gulf of St. Lawrence, was surveyed to collect intertidal data on seaweed biomass (Fig. 19; Adey and Hayek 2005). Based on the information obtained from the visible infralittoral and subtidal fringe during that survey, and the published subtidal macrophyte and Strongylocentrotus droebachiensis literature (see Figs. 4, 5, 6; and especially Himmelman 1985, 1991; Himmelman and Lavergne 1985; Himmelman and Nedelec 1990; Himmelman et al. 1983; Hooper and Whittick 1984; Keats 1986; Keats et al. 1982, 1985), it was decided to concentrate our subtidal quantitative surveys of seaweed biomass on southern Labrador, the Quebec shore of the northeastern Gulf of St. Lawrence, and northeastern Newfoundland from Cape Bauld to Trinity Bay. These investigations were carried out from 2003 to 2006. Similar unpublished studies had been carried out W.H. Adey from Cape Elizabeth to Machias Bay in the Gulf of Maine during the 1990s, and based on the results of the Newfoundland/Labrador work, additional stations were added in the Gulf of Maine and southwestern Nova Scotia in 2005 and 2006. Also, a set of subtidal fringe (infralittoral and 0.5 m below mean spring tide level) stations were occupied during the summer of 2004 along the entire Atlantic Coast of Nova Scotia. Two bay/offshore complexes were examined and collected extensively: Lunaire Road/Canada Bay in the northeastern part of the northern Peninsula of Newfoundland (2005), and Gouldsboro and Dyer bays in eastern Maine (1982–1983; 2009) (see Fig. 19). 36 Northeastern Naturalist Vol. 18, Monograph No. 8 Figure 19. Stations occupied in this project in the northwestern North Atlantic. The solid red dots are the intertidal stations occupied in the initial survey and reported by Adey and Hayek (2005). The red Xs are infralittoral/0.5-m stations occupied primarily to compare Atlantic Nova Scotia, sea-ice-affected shores with non-sea-ice shores. The solid blue dots are full dive stations and the two long red dashes represent areas of numerous local stations and bottom mapping; the red dash on the east side of the northern peninsula of NF represents two separate areas: Lunaire Road to the north and Canada Bay/Englee to the south. The asterisks with arrows are rocky pinnacle stations; the naked arrows are sites where other researchers have carried out extensive algal biomass, community structure, and/or sea urchin-removal studies (see text). 2011 W.H. Adey and L.-A.C. Hayek 37 With the intention of comparing seaweed populations in the Subarctic Region (as determined by the thermogeographic model) with those in the Subarctic/Boreal transition coasts of Nova Scotia and the Gulf of Maine, sets of sampling stations ranging from exposed shores through those of intermediate exposure to highly protected sites were occupied every 50–100 km along each of those coasts. In the analyses and diagrams that follow, we will use the acronyms: NLQ for the Newfoundland, Labrador, Quebec data set; GOM for the Gulf of Maine (Maine Coast) data set, and SNS for the southern (Atlantic) Nova Scotia data set. ENS refers to the Atlantic Coast of eastern Nova Scotia. GSL is used for the Gulf of St. Lawrence. Specific locations in these coastal areas were selected based on the availability of a harbor for the research vessel within acceptable cruising range for a diving skiff to potential dive sites. In a few cases, the research vessel was moved inland, along extended bays, to attain a variety of protected sites, although sites of substantial freshwater input were avoided. We have assumed that this generalized pattern of station selection is independent of seaweed assemblages. The stations and their locations determined by GPS, the dates of station occupation, exposure factors, and the extent of rocky bottom with depth are given in Appendix A. Once a harbor base was located and the research vessel anchored, individual rocky shore sites were selected from the charts. These dive sites were checked, using a sounder mounted on the diving skiff, for a slope that would allow several depth zones of collection to be taken within a single dive without having to board and move divers; given the general difficulty of boarding dry-suited, heavily weighted divers in often choppy seas, this was a practical choice made to reduce overall station time. Typically, two to three dives were required to place, clear, and collect at least one and often two or more square-meter quadrats from each of 5–7 standard depth zones to the depth limit of significant algal cover (including corallines) or available rock substrate, whichever came first. Ideal bottom slopes for this work ranged from about 30–45°; thus, localities with nearly vertical slopes or long, gradually sloping rocky platforms that ranged from 0 to 20–30 m were not included in this comparative analysis of seaweed standing crop and community structure. Most station localities included localized vertical rock faces and horizontal flat surfaces within the dive transect. When near-vertical faces were a conspicuous part of a transect, efforts were made to lay a quadrat on those surfaces with an unconstrained random process. As we discussed in the Introduction, our station/depth limits were generally circumscribed by the depth limitation of rocky substrate rather than available light for algal development. The research vessel The R/V Alca i, a 20-m long, three-masted motor sailor equipped with modern navigational and operational equipment, was used for the project. Also equipped with full, cold-water (dry suit) SCUBA capability, the vessel carried a crew of six: a scientist/captain (W.H. Adey), a mate/engineer, two divers, a photographer, and a cook. The Alca i had laboratory facilities, with a large seawater sink for sorting algae and dissecting and compound microscope facilities for identification. It 38 Northeastern Naturalist Vol. 18, Monograph No. 8 also had drying facilities for herbarium press mounts of voucher algal samples. The Alca i carried a 14-foot, hard-bottom, inflatable dive boat from which all stations were occupied. The dive boat was taken aboard the research vessel for passages between station sets. This small-vessel, “floating laboratory” approach was necessary to provide a wide range of station access with diving equipment and personnel for working on rocky shores often lacking road access. More important, since stations had to be located on shores lacking marine laboratories, the vessel laboratory allowed sorting, identification, weighing, and drying (for vouchers) shortly after quadrat harvest. With typical stations providing 50–150 kg of wet (after draining) algal biomass that was prone to rapid degradation, continuous at-sea processing was essential to success in using an algal biomass parameter. The stations and sampling procedures At most stations, one or two one-meter-square PVC quadrats were dropped by SCUBA at each of six depth zones (0.5, 2.5, 5, 10, 20, and 30 m), depending mostly on substrate availability (see Fig. 18) or occasionally the apparent photic limit (especially for GOM stations). In some cases, three or more replicate quadrats were taken at some depth-zones when time was available. The stations were generally occupied at low water, based on tide tables (International Marine, 2003–2007) as well as an assessment of community zonation at each site. In some cases where the official tide tables appeared inapplicable to the station locality, “meter sticks” were established on the local shore and followed for several tide cycles to determine time of local low water. These procedures were followed to ensure precision of depth location in the shallow zones, especially the narrower subtidal fringe zones; the deeper zones were less critical to depth accuracy and were sometimes occupied at times other than low tide (with due regard to tidal height). The infralittoral zone (i.e., the area between low-water spring tides and low-water neap tides) was also occupied at most stations, but a 1/10th-m2 quadrat was used because of the frequent narrowness of the zone. The reference level for this sample was mean low-water spring tide, from the tide tables (sometimes cross-checked, as noted above), with the 31.6-cm-sided quadrat laid within several cm above that line. On exposed shores, where zones appeared to be elevated, the amount of elevation was considered and the quadrat raised accordingly; however, effectively, it was the lower part of the infralittoral zone that was sampled. On open shores, this was difficult to achieve with precision because of sea or swell; shallow-water sampling (infralittoral, 0.5, 2.5 m) was often delayed for several days, or the station was re-occupied at a later date to achieve greater precision in sampling. Typically, sampling at a station was worked on three successive days, with single quadrats each day for each depth zone for the upper zones on the first two days, and then the deeper zones on the third day. However, there were many variations in this pattern, partly due to weather, especially sea and swell (the exposed shallow zones could not be occupied during large swell periods), and partly due to the time required to sort and process the collections. Exposed stations required low to moderate wind/sea conditions, and on some occasions, such stations were occupied intermittently, with the more protected and intermediate 2011 W.H. Adey and L.-A.C. Hayek 39 exposure stations worked while awaiting a break in the weather for the exposed station. Station locations are shown in Figure 19, and the specific locations, dates, geographic coordinates, and exposure factors are given by region and exposure in Appendix A. This information is summarized by region in Appendix B. Note that while latitude and longitude were taken from marine charts at the time of station occupation, it is Google Earth data that are cited here; while the charts and Google Earth were essentially in agreement for latitude, longitude often differed by as much as a nautical mile. Quadrat location at each station-depth zone was haphazard, as typically the PVC square meter was dropped when the bottom came in view to the descending diver. In some cases in Newfoundland/Labrador where sea urchin/coralline barrens were extensive and larger seaweed patches more scattered, the sampling quadrat was dropped on a patch. Neither diver was a biologist or had training in seaweed identification. The support diver took digital photographs and video on most dives, so that a post-dive view of the bottom could be obtained to assist in efficiency of station completion and for future reference. The partial stations in GOM occupied in the 1990s and 1982–1983 were mostly collected by W.H. Adey and his students using the same basic procedure, although from different research vessels. Additional quadrats, depth zones, and some new stations were taken in Maine during 2006 and 2009, using the divers and procedures outlined above. After the quadrat was set on the bottom, the chief diver removed all macroscopic fleshy algae with a dive knife and placed them in a small-mesh plastic dive bag. A suction device was not employed and inevitably some very small algae or algal fragments were lost. Since the purpose of this study was not to measure biodiversity, but rather to determine species biomass, such losses were not likely significant to the intent of the project. In the subsequent graphic and tabular analyses, species with biomass values less than 1% of biomass are tabulated (at <1%, <0.1%, and <0.01%), but are not included in the essential analyses. We did not consider this a critical issue, as in virtually all field studies, a line is drawn for precision of coverage (usually at microscopic vs macroscopic) but sometimes it is phyletic (corallines and fleshy crusts, diatoms, and cyanobacteria often being omitted, even when they are macroscopic in dimension). With reference to earlier work by one of the authors (W.H. Adey, see Literature Cited), crustose corallines were included by bringing rock substrate to a surface floating laboratory where cover could be measured as a proxy for biomass. Also, many seaweed species pass through a young microscopic phase and are omitted from all subtidal benthic studies except where the bottom is removed for surface laboratory analysis. Time, weather, and cost are always critical factors in field sampling intensity. Increased intensity will inevitably lead to reduced geographic coverage; in this case, the latter parameter was the primary intent of this study. Within each region, stations were selected by wave-exposure. Wave exposure is a critical factor in determining macroalgal assemblages, and regional comparisons need to be based on similar exposure characteristics. All exposed stations occupied in this study occur on the open coast where they are subject to sea and swell at very large fetch. As shown in Appendix A, the angle of exposure to the open sea at each station varied widely, from 23° to 204°. However, the 40 Northeastern Naturalist Vol. 18, Monograph No. 8 mean exposures for NLQ, GOM, and SNS are 103°, 117°, and 106°, respectively, indicating that as comparative sets of stations, the exposure factor is similar. Most exposed stations in all three regions face northeast to southeast into the North Atlantic, Gulf of Maine, and/or Labrador Sea. A single station in NLQ occurs in the northern Gulf of St. Lawrence, and it faces seaward (i.e., into the Gulf of St. Lawrence) at 168° from NE to SW. In this case, the NE fetch is only 100 km; however, the fetch to the south is 500–600 km, and thus the station can be subject to very large waves. Likewise, the Salvage/Skerwink station off Trinity, NF, has a narrow exposure (23°) to the Labrador Sea, but has 30–90 km exposure to the east and south in large Trinity Bay. An important factor in wave and swell exposure is the character of the seaward slope of bottom off each station. A very low-sloping bottom or off-lying skerries might reduce wave energy in shallow water. However, as we discussed above, station sites were selected for moderate bottom slope (30–45°) to facilitate collection. As noted in station characteristics (Appendix A), the mean depth of the rocky/sediment boundary for exposed stations was about 20 m in each region. Thus, rocky bottom at 30 m (and occasionally 20 m) was sometimes replaced by sand or gravel, and sometimes this necessitated moving a short distance in the same area, e.g., West Cod Ledge off Cape Elizabeth, to obtain deeper quadrat samples. For the intermediate stations, open-water exposure is more limited, with a mean of 11–25° as compared to 103–117° for the exposed shores; however, such stations are set back from the outer shore 1.0, 2.9, and 5.9 km for NLQ, GOM, and SNS, respectively. The more finely dissected shore of the largely sedimentary rock coast of NF as compared to the granitic or metamorphic shores of GOM and SNS provides narrow channels and short distances to protected waters. The more open bays of GOM and SNS require greater distances to limit wave action while still providing the temperature/salinity amelioration of the open ocean. For these intermediate stations, the rock substrate limit occurred at 15.5, 12, and 9 m (respectively, for NLQ, GOM, and SNS), less than that in higher energy shores; the generally steeper NLQ shores also bear less fine glacial sediment, providing lesser depth limitation. In summary, as seen in Appendix C, open-ocean exposure angle drops from a mean of 109° for exposed stations to 17° for intermediate stations to 2° for protected stations, while the exposure distance changes respectively from 0 to 3 km to 7 km. There is no significant difference between NLQ, GOM, and SNS regions for exposed stations. For intermediate and protected stations, there is a tendency for GOM and SNS to have larger exposure angles, but this trend is compensated by greater distances from the open ocean. Note that the shorter distance to open water in NLQ results from generally narrower channels and zero exposure to large fetch as compared to 2.2° for GOM and 3.5° for SNS. We generally maintained a requirement for rocky bottom to at least 5 m for protected stations, although a single station in NLQ (St. Lunaire Road, NF) had rocky bottom to only 3 m depth. Following each dive, the sample bags (2–5) were returned to the Alca i and hung over the side where the movement of the boat, with wind, sea, and tide, kept them in good condition for the typically 12–24-hour period required to fully 2011 W.H. Adey and L.-A.C. Hayek 41 process a day’s dive. Following identification and sorting in wet trays, or for kelps or other large algae left to drain in the sink, each species bundle was drained in a one-cm mesh bag, in the vessel’s laboratory (blotted in a few “spongy” cases), weighed wet (typically two-hour drain weight), and logged. Every effort was made to standardize the procedure, and only a few species proved difficult (most notably Codium fragile (Suringar) Hariot). Before draining, representative specimens of key species were selected for herbarium mounts, with smaller sub-samples placed in plastic bags with silica dehydrant. The latter samples were taken to make later molecular analysis possible. The herbarium sheets were mounted in a press and placed in the warm and dry engine room for adequate drying away from the generally humid environment in the vessel’s laboratory. The herbarium sheets, silica-gel dried subsamples, hard copy, and digital-based station/species data are stored at the US National Herbarium (US) in Washington, DC, as voucher specimens. At present, these materials are available by contacting author W.H. Adey; some are currently on loan to outside researchers. Eventually, all will be placed in US and supplied with accession numbers. Statistical methods Where appropriate, we fit general linear models (GLM) to our data set using factorial models, based upon both biomass and species counts; we sought to answer questions of exposure, depth, region, and interaction significance over time and space. The data consisted of the more abundant and common species. As we describe below, we eliminated the rare species that did not contribute to the difference or dissimilarity between factors. Variance assumptions were first tested with Hartley's F-max test and Levene’s test within this framework. Missing data were eliminated. We used partial eta squared as a measure of observed effect size for major hypotheses. Before analysis, data were transformed to natural logarithms with unity added to stabilize the variance. However, due to the nature of this data, equivalent significance levels were found for both transformed and untransformed data. Null hypotheses for differences over time and space with their statistical interactions were tested for the three geographic areas and three exposure levels. To answer questions on the comparison of three or more groupings, we used one-degree-of-freedom Sheffé contrasts, which require no error adjustments. Computations for both descriptive and inferential statistical analyses were completed with SPSS (SPSS, Inc., ver 17, 2008). Tests for linear trends in proportions over the data from NLQ to SNS were performed, but were not significant and are not reported. Within each geographic area, when results were examined for this single locality or in comparison to the other localities, we also performed two one-degree- of-freedom tests. To test the significance of the difference between species proportions, we used two approaches. The first was to transform the relative abundances by the variance-stabilizing arcsine transformation (e.g., Sokal and Rohlf 1969). We chose to model the comparisons with log likelihood ratio and chi square tests. In all cases, significance levels of both tests were equivalent. For any distribution of frequencies, we hypothesized that sampling was from a population in which the species frequencies are estimates of the parametric frequencies 42 Northeastern Naturalist Vol. 18, Monograph No. 8 expected. The distribution used for comparison in the goodness-of-fit tests in each section became the basis for our comparison; we used as expected: NLQ, GOM, and SNS, and asked how well distributed comparative observations were in the adjacent region. Finally, these independent tests were performed with multiples of the most abundant species at: infralittoral, 0.5, 2.5, 5, 10, 15, 20, and 30 m. In addition to the statistical hypothesis testing above, we used the program Primer (Primer 6, 2009 Primer-ELtd. Lutton, Ivybridge, UK) to complement these tests with graphical and exploratory multi-species results for each geographic area, exposure level, and all depths over the three zones (regions). The PRIMER analysis of similarity routine (calculated with its global R measure based upon the full-dimensional similarity matrix of untransformed Bray–Curtis coefficients) was used to quantify a picture of the zonal separation. We used the similarity matrix for the plot in Figure 20, and it clearly reveals many features predicted by the TM. Figure 20. Bray-Curtis multivariate similarity ordination plot showing strong separation of the NLQ (Subarctic; blue lines) stations from the Gulf of Maine (GOM; green lines) and southern Nova Scotia (SNS; red lines) stations with the seaweed abundance by biomass determined at various degrees of wave-exposure and depth (e.g., infralittoral, I; 0.5 m, 2.5 m, etc.) in all three sampling regions. The deepest depth zones at exposed stations in GOM and SNS crossover into the range of the Subarctic; this result is due to the dominance of Subarctic species related to a strong thermocline, and colder temperatures at depth. 2011 W.H. Adey and L.-A.C. Hayek 43 For the entire data set with all three geographic areas by exposure level, we used the program SIMPER with the same similarity coefficient to specify species’ contributions to the observed similarities/dissimilarities. Finally MDS, or non-metric multidimensional scaling, was used as a descriptive method for examining and plotting the multispecies assemblage in Figure 20. More than one decision enters into the selection of an appropriate dimensionality for MDS, for example, stability, ease of use, and interpretability, so we considered these factors along with the most widely used measure of goodness of fit or stress, which is the square root of the normalized “residual sum of squares”. In all cases, the two-dimensional representation was sufficient. Rocky pinnacle sites In the Newfoundland Labrador region, localized kelp beds sometimes occur at isolated sites in protected bays and harbors. These are sites where rocky pinnacles, rock ledge, or boulders and cobbles project up through soft sediment (Fig. 10a) or mobile gravel and shell beds (Fig. 10b), and where considerable but very localized development of Saccharina spp. kelp forest can occur. We also located several isolated wood and stone docks and breakwater rip-rap that rose above an expansive sediment floor, and similarly supported abundant kelp. We have investigated six of these bay/harbor sites and carried out a series of quadrats. Few, if any S. droebachiensis were found at these sites. The operational/collecting procedures were generally the same for rocky pinnacles as for rocky shore stations. Unfortunately, we could not devote the time to carry out systematic searches for rocky pinnacle sites, and they are likely more numerous than these data would suggest. The six sites that we worked were discovered while establishing rocky shore stations. Nevertheless, it is unlikely that the amount of rocky pinnacle area is significant when compared to the rocky shore area that is the focus of this paper. Taxonomy and Identification Using a dissecting microscope and hand-sectioning, we consulted Sears (2002) as our primary reference for routine identification with a compound microscope. Taylor (1957) was used as an essential back-up, and, as noted above, herbarium sheet vouchers and silica gel subsamples were routinely taken. Although not used in the field, South and Hooper (1980) and Mathieson et al. (2010) were used for later reference. Nomenclature including authority of all species mentioned is provided in Appendix E. At the scale of the biomass questions being asked, identification problems were minimal. At many GOM subtidal stations, a Porphyra spp.was moderately abundant (although always below 1% of biomass; see Table 3). This was tentatively identified as P. purpurea and is so labeled in the voucher collection in the US National Herbarium, although its identity remains uncertain. Since very little subtidal Porphyra was found in the NLQ and SNS stations, and those few in NLQ were regarded as a separate but unidentified species, any quantitative effects were negligible. Based on discussions with Max Hommersand (University of North Carolina, Chapel, Hill, NC, personal comm.), the Newfoundland/ Labrador Phycodrys species were probably P. riggii, and those in GOM and Nova 44 Northeastern Naturalist Vol. 18, Monograph No. 8 Table 1. Species total mean biomass in region (g/m2). Newfoundland/Labrador/Quebec (NLQ) Gulf of Maine (GOM) Southern Nova Scotia (SNS) Agarum clathratum 12,717 Chondrus crispus 13,796 Laminaria digitata 35,292 Alaria esculenta 6220 Saccharina latissima 13,011 Chondrus crispus 19,902 Chordaria flagelliformis 5921 Saccharina longicruris 10,437 Saccharina latissima 18,955 Desmarestia viridis 5101 Laminaria digitata 6408 Corallina officinalis 6599 Fucus distichus 2935 Alaria esculenta 4373 Fucus serratus 5204 Cumulative subtotals 32,894 (79%) 48,025 (80%) 85,955 (82%) Laminaria digitata 1550 Agarum clathratum 1969 Codium fragile 3908 Dictyosiphon foeniculaceus 1191 Cystoclonium purpureum 1783 Saccharina longicruris 2996 Devaleraea ramentacea 1046 Ceramium rubrum 1205 Fucus distichus 2398 Saccorhiza dermatodea 954 Neosiphonia harveyii 941 Ascophyllum nodosum 1549 Ptilota serrata 730 Phyllophora pseudoceranoides 751 Agarum clathratum 1503 Cumulative subtotals 38,365 (92%) 54,674 (91%) 96,400 (92%) Acrosiphonia arcta 552 Desmarestia viridis 635 Phyllophora pseudoceranoides 1087 Pylaiella littoralis 212 Corallina officinalis 616 Cystoclonium purpureum 860 Rhodomela confervoides 118 Phycodrys rubens 614 Alaria esculenta 657 Palmaria palmata 112 Fucus distichus 573 Palmaria palmata 488 Saccharina latissima 66 Acrosiphonia arcta 523 Antithamnion sp. 467 Cumulative subtotals 39,425 (94%) 57,635 (96%) 99,300 (95%) Turnerella pennyi 65 Callophyllis cristata 461 Desmarestia viridis 428 Petalonia fascia 63 Porphyra purpurea 375 Ceramium rubrum 386 Scytosiphon lomentaria 45 Devaleraea ramentacea 364 Saccorhiza dermatodea 313 Ectocarpus sp. 45 Desmarestia aculeata 216 Rhodomela confervoides 277 Chorda filum 26 Monostroma sp. 208 Neosiphonia harveyii 210 Cumulative subtotals 39,669 (95%) 59,259 (99%) 100,635 (96%) Totals 41,734 59,882 104,564 22 additional species, see Table 2 26 additional species, see Table 3 38 additional species, see Table 4 42 spp. (total) 46 spp. (total) 58 spp. (total) 2011 W.H. Adey and L.-A.C. Hayek 45 Scotia were probably P. rubens. This matter can be resolved with the voucher collections; because it would have little effect on the conclusions of this study, this issue is left for future study. Finally, in this study, no differentiation was made between component subspecies of Fucus distichus (Fucus distichus ssp. edentatus and Fucus distichus ssp. evanescens), although a large, desiccated collection was taken to allow future molecular systematic research. Results Table 1 provides a summary list of the dominant species present in each of the study regions (NLQ, GOM, SNS); the values given are mean biomass at all depths and exposures. Tables 2–4 list the lesser abundant species in each region. This type of presentation emphasizes the regional, rocky shore photosynthetic potential of benthic macroalgae because, as we described earlier, the depth zones are approximately logarithmic with depth. Assuming that the lengths of exposed, intermediate, and protected shorelines are approximately equal on these dissected headland/bay coasts, the listing provides a regional view of each species’ macroalgal contribution to each region’s rocky shore ecosystem. Table 2. Newfoundland/Labrador/Quebec species less than 1% of the total biomass. Species listed in columns, within station categories, by order of relative abundance. 0.1% < spp. < 1% 0.01% < spp. < 0.1% spp. < 0.01% Exposed stations Saccharina latissima Pilayella littoralis Ectocarpus sp. Phycodrys riggii Petalonia fascia Desmarestia aculeata Palmaria palmata Euthora cristata Corallina officinalis Scagelia pylaisii Scytosiphon lomentaria Rhodomela confervoides Polysiphonia urceolata Polysiphonia flexicaulis Membranoptera alata Intermediate stations Saccharina latissima Phycodrys riggii Laminaria digitata Pilayella littoralis Scytosiphon lomentaria Corallina officinalis Petalonia fascia Scagelia pylaisii Ectocarpus sp. Polysiphonia urceolata Chorda filum Elachista fucicola Monostroma sp. Euthora cristata Ceramium sp. Porphyra sp. Polysiphonia flexicaulis Membranoptera alata Chaetomorpha spp. Neodilsea integra Rhodomela confervoides Chondrus crispus Protected stations Rhodomela confervoides Pilayella littoralis Scagelia pylaisii Polysiphonia flexicaulis Phycodrys riggii Corallina officinalis Monostroma sp. Petalonia fascia Polysiphonia urceolata Scytosiphon lomentaria Neodilsea integra Euthora cristata Ectocarpus sp. 46 Northeastern Naturalist Vol. 18, Monograph No. 8 Figure 20 presents the results of a Bray Curtis similarity analysis of station/ depth species biomass using PRIMER software. Grouping station/depths by similarity of species abundance in the core Subarctic region (NLQ) shows Table 3. Gulf of Maine species whose biomass is less than 1% of total biomass. Species listed in columns, within station categories, by order of relative abundance. 0.1% < spp. < 1% 0.01% < spp. < 0.1% spp. < 0.01% Exposed stations Phyllophora pseudoceranoides Corallina officinalis Fimbriofolium dichotomum Neosiphonia harveyii Saccorhiza dermatodea Polysiphonia sp. Rhodomela confervoides Dumontia incrassata Polyides caprinus Phycodrys rubens Porphyra spp. (purpurea?) Bonnemaisonia hamifera Palmaria palmata Antithamnion sp. Monostroma sp. Membranoptera alata Devaleraea ramentacea Ceramium rubrum Euthora cristata Mastocarpus stellata Desmarestia viridis Desmarestia aculeata Fucus distichus Ptilota serrata Intermediate stations Ceramium rubrum Monostroma spp. Ahnfeltia plicata Phyllophora pseudoceranoides Dumontia incrassata Polysiphonia sp. Neosiphonia harveyii Rhodomela confervoides Polyides caprinus Corallina officinalis Alaria esculenta Phycodrys rubens Mastocarpus stellatus Euthora cristata Porphyra spp. (purpurea?) Chaetomorpha spp. Fimbriofolium dichotomum Acrosiphonia arcta Bonnemaisonia hamifera Fucus distichus Elachista fucicola Saccorhiza dermatodea Palmaria palmata Desmarestia viridis Chordaria flagelliformis Ptilota serrata Devaleraea ramentacea Desmarestia aculeata Cladophora spp. Protected stations Cystoclonium purpureum Rhodomela confervoides Ptilota serrata Neosiphonia harveyii Euthora cristata Elachista fucicola Fucus distichus Palmaria palmata Monostroma spp. Porphyra spp. (purpurea) Mastocarpus stellatus Fimbriofolium dichotomum Corallina officinalis Bonnemaisonia hamifera Membranoptera alata Desmarestia viridis Phycodrys rubens Dumontia incrassata Ascophyllum nodosum Phyllophora pseudoceranoides Scytosiphon lomentaria Polysiphonia sp. Antithamnion sp. Ceramium rubrum Ceramium circinnatum Laminaria digitata Alaria esculenta Fucus vesiculosus Acrosiphonia arcta Desmarestia aculeata Devaleraea ramentacea 2011 W.H. Adey and L.-A.C. Hayek 47 that the dominant colder water species tend to change with depth, more or less uniformly, especially in shallow to mid-depths. Similar changes occur in the Boreal transition coasts of SNS and GOM, although the species involved are mostly different, warmer-water species. However, on more exposed shores in the SNS and GOM, considerably warmer surface layers are underlain in the deepest photic zones by waters with summer temperatures in the Subarctic range. Correspondingly, in the deepest parts of exposed stations, species/biomass distributions change from the warmer SNS and GOM species to colder species, thus having similarities in the NLQ range. Figure 20 also shows the shift of species composition, all-depths and all exposures considered, from Table 4. Southern Nova Scotia species less than 1% biomass. Species listed in columns, within station categories, by order of relative abundance. 0.1% < spp. < 1% 0.01% < spp. < 0.1% spp. < 0.01% Exposed stations Corallina officinalis Palmaria palmata Ahnfeltia plicata Alaria esculenta Ceramium rubrum Euthora cristata Cystoclonium purpureum Leathesia difformis Elachista fucicola Ptilota serrata Neosiphonia harveyii Antithamnion spp. Fucus distichus Dumontia incrassata Rhodomela confervoides Monostroma spp. Desmarestia aculeata Scagelia pylaisii Nemalion multifidum Pilayella littoralis Phyllophora pseudoceranoides Fimbriofolium dichotomum Devaleraea ramentacea Chordaria flagelliformis Desmarestia viridis Acrothrix nova-angliae Phycodrys rubens Turnerella pennyii Callithamnion tetragonum Membranoptera denticulata Intermediate stations Cystoclonium purpureum Rhodomela confervoides Polysiphonia spp. Antithamnion spp. Neosiphonia harveyii Ahnfeltia plicata Ceramium rubrum Acrosiphonia arcta Cladophora spp.. Palmaria palmata Polyides caprinus Pilayella littoralis Desmarestia viridis Fucus vesiculosus Monostroma spp Chorda filum Chordaria flagelliformis Dumontia incrassata Saccorhiza dermatodea Phycodrys rubens Ulva lactuca Desmarestia aculeata UID filiform brown Alaria esculenta Ectocarpus spp. Euthora cristata Mastophora stellata Chaetomorpha spp. Codium fragile Pantoneura fabriciana Protected stations Codium fragile Ceramium rubrum Palmaria palmata Phyllophora pseudoceranoides Ahnfeltia plicata Phycodrys rubens Cystoclonium purpureum Cladophora spp. Euthora cristata Polysiphonia harveyii Dictyosiphon foeniculaceus Dumontia incrassata Rhodomela confervoides Monostroma spp. Chaetomorpha spp. Pilayella littoralis Antithamnion spp. Ascophyllum nodosum Saccorhiza dermatodea Ulva lactuca Chordaria flagelliformis Polysiphonia spp. Polyides caprinus Desmarestia viridis 48 Northeastern Naturalist Vol. 18, Monograph No. 8 the core Subarctic NLQ to the transitional regions, placing SNS furthest (least similar) from NLQ, with GOM in an intermediate position, but closer to SNS. Protected and intermediate stations with rocky bottoms in SNS and GOM reach relatively shallow depths with greater tidal mixing; they only show a small cold shift at depth, with warm surface species changing to warm deeper species. The biomass of all collected species, listed as a mean of all quadrats, by region, depth, and exposure is given in Appendix D. As shown in Table 1, the first five species by harvested biomass for the Newfoundland/ Labrador/Quebec NLQ Region provide approximately 80% of the entire rocky shore macroalgal biomass of the region and fall into three categories: (1) a sublittoral fringe (Chordaria flagelliformis and Fucus distichus); (2) an upper kelp forest (Alaria esculenta); and (3) a deeper, savanna-like kelp zone (Agarum clathratum and Desmarestia viridis). The next group of five, bringing the summed biomass to over 90% of the total, adds Devaleraea ramentacea (L.) Guiry and Dictyosiphon foeniculaceus (Hudson) Greville to the sublittoral fringe, Laminaria digitata and Saccorhiza dermatodea to the upper kelp forest, and Ptilota serrata to the lower bushy red algal zone. In this analysis, the unique character of the NLQ Subarctic Region is apparent in several ways. (1) The first five species in the NLQ list that comprise 79% of the total biomass provide only a total of 13% and 7.5% of the total biomass in GOM and SNS transitional areas, respectively. A MANOVA model showed that geographic area (Wilks criterion = 0.143, P = 0.0001; hypothesis degrees of freedom: 20, 64; Effect size: 0.542; power = 1.0) and exposure (Wilks criterion = 0.096; P = 0.0001; hypothesis degrees of freedom: 20, 64; Effect size: 0.616; power = 1.0) were highly significant factors. One-degree-of- freedom contrasts at a test level of 0.001 found that each of the two areas of GOM and SNS are distinct as well (P = 0.0001). (2) In NLQ, the kelp species Agarum clathratum, a North Pacific and northwestern Atlantic Subarctic endemic, provides twice the biomass of the next most abundant species and 31% of total biomass (as compared to providing less than 3% of the biomass in each of the transitional areas of GOM and SNS). (3) Two of the five most abundant species in NLQ (Chordaria flagelliformis and Desmarestia viridis) do not appear on the lists of the 20 most abundant species for the transitional areas (C. flagelliformis: <0.1%, D. viridis: <1%; Tables 3, 4). As shown in Table 1, about half of the top 20 macroalgal biomass producers on Gulf of Maine and Nova Scotia rocky shores do not appear in the NLQ tabulation; most of these species did not occur, even as trace elements, in the sampling of this study. Of the first five ranked species by harvested biomass in GOM, one, the topranked species Chondrus crispus, is rare to occasional only on the warmer side of NLQ, and two additional species (the Saccharina species) are not common on NLQ rocky shores. The next group of five includes Agarum clathratum, with 30% of the total biomass in NLQ but only 3% in GOM. For this species, it is not just relative weight that is lost; in total biomass, A. clathratum is 15% of that in NLQ. Its companions in the second-ranked group are relatively small filamentous or bushy reds that are uncommon in NLQ. Nova Scotia is similar in species ranking, although Laminaria digitata becomes highest ranked, with Chondrus crispus in 2011 W.H. Adey and L.-A.C. Hayek 49 the second-ranking position. A species of moderate occurrence in GOM, and only an occasional in NLQ, the articulate coralline Corallina officinalis L., becomes fourth-ranked in SNS; the 5th and 6th slots are occupied by the invasive species Fucus serratus L. and Codium fragile, neither of which were found at any of our NLQ stations. In summary, of the top 10 ranked species in NLQ, nine drop in rank from an average position of 4.3 (NLQ) to 17 in GOM and 18 in SNS. In reverse, of the top 10 ranked species in Nova Scotia, eight drop in abundance from a mean rank of 4.6 (SNS) to 16 in GOM and 34 in NLQ. In addition, as we will further demonstrate, the depth zonation is strikingly different between NLQ and the two southern regions. The distinctive subtidal fringe zone of NLQ, heavily dominated by Chordaria flagelliformis, Devaleraea ramentacea, Fucus distichus, and Dictyosiphon foeniculaceus, with more scattered Acrosiphonia arcta (Dillwyn) Gain, virtually disappears, becoming primarily a bed of Chondrus crispus (in Nova Scotia with Corallina officinalis L.) that grades under the deeper kelps in GOM and SNS. The extensive forest of Alaria esculenta in NLQ, grading from the subtidal fringe to 3–8 m in depth in intermediate and exposed areas, is largely replaced by a Saccharina and Laminaria forest in GOM and SNS, but with an abundant understory of bushy and filamentous reds (the latter lacking in NLQ). Exposed GOM bears some resemblance to NLQ, but the Alaria band is much narrower, being largely restricted to 0.5 m depth. Unlike NLQ, there is no forest or savanna of Agarum clathratum in GOM and SNS, but rather scattered plants, grading into a red algal zone. All in all, the rather strong zonation of NLQ rocky shores is much more gradational in GOM and SNS. For the 10 most abundant species in NLQ, an examination of the seaweed assemblage by ANOVA coupled with SIMPER results, showed that NLQ was significantly different from each of the other regions; the average dissimilarity for this group of species was 90.4 compared to GOM and 88.7 compared to SNS. We also examined the equivalent sets of the top 15 and top 20 most abundant species groups by each of these methods along with a chi square test, and again found that each of the three regions was statistically distinct. Dissimilarity results were over 89.0 for each region. Newfoundland/Labrador/Quebec (NLQ) Rocky shore. The cumulative mean biomass for each species with a biomass exceeding 1% of the total for any one depth zone is plotted by depth zone for exposed, intermediate, protected, and rocky pinnacle sites, from the northeastern Newfoundland, southern Labrador, and eastern Quebec stations in Figures 21–24. Species achieving less than 1% of total biomass in any given depth zone are listed in order of abundance in Table 2. For a total account of all species biomass by depth and exposure, see the Appendix D (1). The total biomass (taken from species means) for each of the wave exposure groups in NLQ is 20–25% lower than GOM and 35–73% lower than Nova Scotia. The rocky exposed sublittoral of this region can be divided into five overlapping zones: (1) a sublittoral fringe (less than 2.5 m); (2) an upper kelp forest (0.5–2.5 to 5–10 m); (3) a deeper kelp “savanna” (5–20 m); and (4) a deeper, principally 50 Northeastern Naturalist Vol. 18, Monograph No. 8 bushy, red algal zone (usually to the limit of hard bottom or 30–35 m). With lesser exposure, the zones are compressed to shallower water. The sublittoral fringe set of species in this case includes Fucus distichus, Chordaria flagelliformis, Dictyosiphon foeniculaceus, Devaleraea ramentaceum, and Acrosiphonia arcta. The upper kelp forest zone, depending on exposure, includes Alaria esculenta, Laminaria digitata, and Saccorhiza dermatodea. The deeper kelp savanna is made up primarily of Agarum clathratum and Desmarestia viridis with underlying and often open patches of small red algae. Finally, the deeper fringe of Ptilota serrata and Turnerella pennyi (Harvey) F. Schmitz with scattered plants of Agarum, which shows as greatly diminished biomass on Figures 21–24, is quite apparent and often expansive when diving. Much of the rock surface is more or less thinly coated (up to 10 cm thick) with coralline crusts (see Figs. 7, 14, 16). Figure 21. Cumulative mean biomass of dominant (greater than 1% of total biomass for any one depth zone) macroalgal species in standard depth zones, for exposed coasts in northeastern Newfoundland, southern Labrador, and eastern Quebec (NLQ). Species with less than 1% of total biomass are listed in Table 2; a complete listing of mean biomass for all species is given in Appendix D1. Note that area between the curves in these diagrams represents species biomass, and all species together are cumulative to give depth-zone regional mean biomass; e.g., at 10 m depth, Agarum clathratum had a mean, all-exposed stations biomass of 1806 g/m2, Desmarestia viridis had a mean biomass of 262 g/m2, Ptilota serrata had a mean biomass of 121 g/m2, and all miscellaneous spp. had a mean biomass of 34 g/m2. For species abbreviations used in this and other figures, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 51 In the progression from exposed, open coast sites to intermediate and protected, in-bay sites in NLQ, the compression upwards of both the available rock substrate at the bottom of the transect and each of the primary zones is distinctive. However, the most striking change with exposure lies in the upper sublittoral fringe. At exposed stations, the effects of moving winter sea ice on the sublittoral fringe are obvious: mean biomass is lower in the infralittoral and at 0.5 m than it is at 2.5 m; the primary species are annuals in the infralittoral, and Alaria esculenta, which recovers over a year or two from ice scour, dominates at 0.5 m. The rock surface can be very irregular; ridges and high spots are often scraped clear, although Chordaria flagelliformis recruits quickly in the spring, especially in the infralittoral and 0.5-m zones. At intermediate exposures, the perennial Fucus distichus is moderately abundant, and biomass peaks in the infralittoral at roughly 50% higher than at exposed stations. This greater abundance of Fucus distichus extends up into the lower intertidal zone. Finally, a combination of low salinities in the spring and the lack of constant wave action—allowing S. droebachiensis to extend right to the low water line in summer— greatly reduces biomass from low water down to 0.5 m at the protected stations. The primary upper kelp forest zone on average is 5–7 m deep on exposed coasts, 2 m deep at intermediate sites, and virtually disappears in protected bays, with Agarum clathratum and scattered individuals of Desmarestia viridis providing the last remnants of kelp forest, mostly in patches as a “kelp savanna”. However, it is striking that in our data Agarum clathratum and Desmarestia viridis both achieve maximum abundance on protected bay shores at relatively shallow depths (2.5 m). Table 5. Species abbreviations used in the figures. Abbreviation Species Abbreviation Species Aa Acrosiphonia arcta Fs Fucus serratus Ac Agarum clathratum He Himanthalia elongata Ap Ahnfeltia plicata Ld Laminaria digitata Ae Alaria esculenta Ms Mastocarpus stellatus A spp. Antithamnion spp. Ma Membranoptera alata An Ascophyllum nodosum M spp. Membranoptera spp. Ct Callithamnion tetragonum Mo spp. Monostroma spp. C spp. Ceramium spp. O spp. Other species Cof Chorda filum Pp Palmaria palmata Cf Chordaria flagelliformis Pf Petalonia fascia Cfr Codium fragile Pr Phycodrys rubens Coff Corallina officinalis Php Phyllophora pseudoceranoides Cp Cystoclonium purpureum Pc Polyides caprinus Da Desmarestia aculeata P spp. Polysiphonia spp. D spp. Desmarestia spp. Pop Porphyra purpurea Dv Desmarestia viridis Ps Ptilota serrata Dr Devaleraea ramentaceae Pl Pylaiella littoralis Df Dictyosiphon foeniculaceus Rc Rhodomela confervoides Di Dumontia incrassata Sal Saccharina latissima E spp. Ectocarpus spp. Sd Saccorhiza dermatodea Ef Elachista fucicola Sp Scagelia pylaisaei Eu Euthora (Callophyllis) cristata Sl Scytosiphon lomentaria Fid Fimbriofolium dichotomum Tp Turnerella pennyi Fd Fucus distichus 52 Northeastern Naturalist Vol. 18, Monograph No. 8 Thirty species with a biomass at any one depth zone greater than 0.1% of the total were tallied for the NLQ region in this survey, with an additional 12 species tallied in lesser or trace amounts. Although the species listing in Table 1 and the biomass display in Figures 21–24 provide some insight into the relationship between the larger macroalgae and the small species, these presentations do not fully identify the structure of the seaweed assemblages because the canopy biomass overwhelms smaller species. For example, in the list of 20 species making up 95% of the total biomass in NLQ, only one species characterizing the red bushy zone appears (Ptilota serrata), and, in general, the understory and epiphytic species are under-represented. For that reason, we have presented the same exposure/biomass groupings in Figures 25–27, but with species separated into three categories: canopy, understory, and epiphytic/small species, as a percent of total biomass at each depth zone. To some extent, this breakdown is arbitrary, and some species (e.g., Chordaria flagelliformis) fit into grey zones between canopy and understory. Where there was a question, we placed the species in the category that would allow more visible display. While the epiphytic category is straightforward, as epiphytic/small it becomes a catch-all for very small species. Figure 22. Cumulative mean biomass of dominant (greater than 1% of total biomass for any one depth zone) macroalgal species in standard depth zones, for intermediate exposure coasts in NLQ. Species with less than 1% of total biomass are listed in Table 2; biomass summation for all species is given in Appendix D(1). 2011 W.H. Adey and L.-A.C. Hayek 53 The basis for color-coding in these figures and other figures is described in the Discussion section. Several elements of these seaweed assemblage diagrams are quite revealing, among them: (1) great dominance of Agarum clathratum and Desmarestia viridis in the lower half to two-thirds of the subtidal macrophyte zone, especially in protected waters; (2) a sharp shift of species composition at 10, 5, and 2.5 m depth (exposed to intermediate to protected) between Alaria-dominated kelp forest (with abundant Chordaria flagelliformis) and Agarum-dominated kelp (with abundant Desmarestia viridis); (3) the lack of a sharp break between the sublittoral fringe, the infralittoral, and kelp forest; (4) an order-of-magnitude lower biomass in understory as compared to canopy (that does not occur in GOM and SNS); and (5) the importance of Ptilota serrata in deeper water (especially at exposed and intermediate stations). In shallow water, Agarum clathratum occurs beneath Alaria (and L. digitata when it is present); however, it does not reach its peak of absolute abundance until one to two depth zones below the other kelp species. It then drops quickly in abundance at deeper depths, from near-continuous cover to Agarum savanna to scattered individuals in the lowest zone. Figure 23. Cumulative mean biomass of dominant (greater than 1% of total biomass for any one depth zone) macroalgal species (by weight), in standard depth zones for fully protected coasts in NLQ. Species with less than 1% of total biomass are listed in Table 2; biomass summation for all species is given in Appendix D(1). 54 Northeastern Naturalist Vol. 18, Monograph No. 8 Especially striking, because it would not have been apparent in the earlier figures or Table 1, is the very marked dip in understory biomass at 2.5–5 m. The biomass scale on understory biomass (and epiphytic/small) is reduced an order of magnitude from that used for canopy biomass, and this dip is apparent at all exposures (although in protected waters, a significant rise on the deep side is prevented by lack of substrate). Apparently, on the NLQ rocky shore, the abundant sublittoral fringe understory species Devaleraea ramentacea and Acrosiphonia arcta are replaced in deeper water by Ptilota serrata, but a significant mid-depth understory species does not occur. Another important element of the kelp forest and upper Agarum savanna zone is the annual, large, bushy, brown alga Desmarestia viridis. Making up about a quarter of the mean biomass in protected waters and about 12% overall in the NLQ region, it is more scattered in occurrence than Agarum clathratum, but can occur as large upright bushes, often a meter high and wide but sometimes considerably higher. D. viridis is a chemically protected species, and along with the chemically defended A. clathratum and P. serrata, forms roughly 90% of the macroalgal biomass below 5 m in exposed and intermediate stations and 2.5 m in protected waters. Figure 24. Cumulative mean biomass of dominant (greater than 1% of total biomass for any one depth zone) macroalgal species (by weight), in standard depth zones, for isolated rocky pinnacles located within Newfoundland and Labrador bays. A total spp. compilation is given in Appendix D(2). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 55 In the epiphytic/small category, the basic patterns we have discussed for canopy and understory species are repeated. Dictyosiphon foeniculaceus dominates in shallow water and Euthora (Callophyllis) cristata and Scagelia pylaisaei Figure 25. Macroalgal species at exposed stations in NLQ separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of color coding: dark blue (Arctic), blue (Subarctic), purple (Boreal), white (cosmopolitan). O spp. = other, non-separated spp. For species abbreviations, see Table 5. 56 Northeastern Naturalist Vol. 18, Monograph No. 8 deeper. However, the remaining species depth/abundance patterns are more irregular from exposure zone to zone. That is to be expected for these smaller and epiphytic species that are likely patchy for a wide variety of reasons, including Figure 26. Macroalgal species at intermediate exposure stations in NLQ separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: dark blue (Arctic), blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 57 the precision of our collecting. A more extensive collecting network would likely be necessary to discern exact patterns within these small algae. Phycodrys riggii is a constant mid-depth understory species, although its biomass is relatively low. The only Arctic species collected regularly in these collections is Turnerella Figure 27. Macroalgal species at protected stations in NLQ separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: dark blue (Arctic), blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 58 Northeastern Naturalist Vol. 18, Monograph No. 8 pennyi, mostly in deep water. This species occurs primarily below 25 m, where summer temperatures are typically below 7–8 °C and often below 5 °C. It was observed in moderate numbers in bay stations where the thermocline can be shallower. At depths over 20 m in southern Labrador and the northern peninsula of Newfoundland, its “rosettes” can reach 0.5 m in diameter. (The Arctic Neodilsea integra was found in scattered localities as individual plants, Table 2). Devaleraea ramentacea was treated as an Arctic species by Lüning (1990), and Adey et al. (2008) followed this assignment in their analysis of Arctic endemics. Although its ancestry in the North Pacific is clear, D. ramentacea is likely Subarctic in its distribution; a quantitative analysis of Arctic shores is necessary for confirmation. Rocky pinnacles. The data for rocky pinnacle sites are presented separately and are plotted for comparison with rocky shore stations (Figure 24). All of these sites are protected from significant wave exposure, and while the sites occur within a few hundreds of meters of extensive intermediate or protected rocky shore bearing the community structure shown in Figure 23, they bear a radically different seaweed assemblage. While elements of the rocky shore community are present, the kelps Saccharina latissima in shallow water and Saccharina longicruris in deeper water become dominant and can produce very localized but dense canopies at these isolated situations. The mean biomass (g/m2) at rocky pinnacle sites considerably exceeds that on the local rocky shore and equals the means for Gulf of Maine and Nova Scotia sites, some 700 km further south; many of the plants are probably quite old, having survived many seasons in protected waters without grazing. With respect to the larger biomass of Saccharina spp., these stations are similar to rocky shore sites in GOM. However, their biodiversity is quite low in comparison, and associated species that are abundant in GOM are lacking (e.g., Chondrus crispus, Cystoclonium purpureum (Hudson) Batters, Ceramium spp., and Corallina officinalis). Furthermore, the dominant, rocky shore-associated species of NLQ (i.e., Agarum clathratum, Desmarestia viridis, Saccorhiza dermatodea) are generally present in modest quantities. The Gulf of Maine (GOM) On rocky shores in GOM (Figs. 28–30; minor species in order of abundance in Table 3), the standing crop of seaweeds typically is roughly 40% higher than on rocky shores in Newfoundland/Labrador/Quebec. While the same basic zonation is visible and a similar upward compression of the deeper zones with less exposure occurs, there are some striking differences in seaweed assemblage. The sublittoral fringe zone on the Maine Coast is highly dominated by the perennial, bed-forming Chondrus crispus. The ubiquitous annual Chordaria flagelliformis of NLQ shores (appearing there at all exposed stations and at half of more-protected stations) is only occasionally present in the GOM, appearing in a single depth zone in Maine at one of 17 stations in this study. This distinction carries downwards from the lowest intertidal as a major difference (Chordaria complex versus Chondrus complex) in the lower intertidal of the two regions. The main, upper kelp forest, while still having important quantities of Alaria esculenta at 2011 W.H. Adey and L.-A.C. Hayek 59 exposed stations, is generally quickly dominated by Saccharina spp. and Laminaria digitata with depth; the Saccharina kelp forest extensively encroaches, in deeper water, on what is the Agarum “savanna” in Newfoundland/Labrador. So extensive is this encroachment that this deeper zone nearly disappears, Agarum clathratum appearing as scattered individuals in a diminishing forest of Saccharina spp. The red “bushy” zone likewise becomes more limited, and Ptilota serrata, while often present, is usually no longer the bed former that it is in more northern waters; it does not appear in the diagrams, but does appear well down the list of lesser species (Table 3). Equally striking in comparing the Newfoundland/Labrador/Quebec diagrams (Figs. 21–24) with those of GOM (Figs. 28–30 and Table 3) is species diversity. On these diagrams, it would appear that NLQ has nearly twice as many species as GOM. However, the correlation was not significant, and is an artifact of the one percent cut-off to allow concentration on biomass dominants. Mostly, this apparent diversity in NLQ is a result of a multispecies flora in shallow water, whereas in GOM, Chondrus crispus beds strongly dominated those zones; however, Ptilota serrata also occurs at modest biomass at depth in NLQ (mean value typically = 100–500 g/m2) as compared to less than 100 g/m2 mean levels in GOM. Total species count, including those occurring at trace levels, is about Figure 28. Cumulative mean biomass of dominant (>1% of total biomass for any one depth zone) macroalgal species (by weight), in standard depth zones for exposed coasts in the Gulf of Maine. See Table 3 for minor species listing. Total spp. compilation is presented in Appendix D(3). For species abbreviations, see Table 5. 60 Northeastern Naturalist Vol. 18, Monograph No. 8 10% greater in GOM than in NLQ. In addition to Chondrus crispus in shallower waters, a host of bushy red macrophytes (Cystoclonium purpureum, Ceramium rubrum C/ Agardh, Neosiphonia harveyii (J.W. Bailey) M.-S. Kim, H.-G. Choi, Guiry, & G.W. Saunders, Phyllophora pseudoceranoides, and Corallina offi- cinalis), rare or occasional in Newfoundland, fill in as a consistent understory beneath the kelp in the forest or in the upper parts of the red zone in deeper water (see Table 3). There are also many colder water subtractions (at least in terms of significant biomass), including Saccorhiza dermatodea, Devaleraea ramentaceum, and Turnerella pennyi, but these are best discussed around the percentage composition diagrams which follow. In GOM percentage composition diagrams (Figs. 31–33), several striking features, compared to those of NLQ, appear: (1) dominance of the Saccharina kelps and Chondrus crispus, especially in intermediate and more protected waters; (2) general reduction of Agarum clathratum, except in the deeper zones at exposed stations (see Fig. 20); (3) general replacement of Ptilota serrata in the understory by a suite of low, bushy reds (especially Cystoclonium purpureum and Phyllophora pseudoceranoides, except at deeper, exposed zones); and (4) great reduction of the ubiquitous epiphytic species of NLQ, Dictyosiphon Figure 29. Cumulative mean biomass of dominant (>1% of total biomass for any one depth zone) macroalgal species (by weight), in standard depth zones for intermediate exposure coasts in the GOM. Minor species are listed in Table 3. Total spp. compilation is presented in Appendix D(3). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 61 foeniculaceus and Pylaiella littoralis in shallow subtidal zones and Scagelia pylaisaei in deeper water. While canopy biomass in GOM was nearly double that of NLQ shores, the understory biomass was increased by about 10 times. This pattern was occurring in spite of the considerably greater overlying biomass of kelps. Even more striking, the sharp dip in NLQ biomass in the 2.5–5-m depth range, as noted above, did not occur; in that range, GOM understory macrophytic biomass for all exposures was about 50 times that of NLQ. Finally, the articulate coralline Corallina officinalis occasionally occurred at stations in NLQ, but the summed, mean, all-exposure biomass logged in at 5.2 g/m2. This compares to 616 g/m2 for GOM and 6599 g/m2 for Nova Scotia in the equivalent analysis (Table 1). These numbers are characteristic of the issues of presence/absence vs quantitative data. C. officinalis occurred throughout NLQ as well as in GOM and SNS. However, in biomass terms, as shown in this study, levels were two and three orders of magnitude higher, in the southern areas. As Figures 31–33 show, Fucus distichus (edentatus/evanescens) occurred as a major component of infralittoral canopy, especially at intermediate and protected stations. However, where it did occur, at siltier sties, it was Figure 30. Cumulative mean biomass of dominant (>1% of total biomass for any one depth zone) macroalgal species (by weight), in standard depth zones for fully protected coasts within GOM. Minor species are listed in Table 3. Total spp. compilation is presented in Appendix D(3). For species abbreviations, see Table 5. 62 Northeastern Naturalist Vol. 18, Monograph No. 8 at considerably less biomass than the equivalent Chondrus crispus beds at more open sites. Fucus distichus did not make the one percent cut-off to appear in our tally of total biomass and is considerably less abundant than in NLQ (Table 1). Although the erratic nature of epiphytic species’ biomass is Figure 31. Macroalgal species at exposed stations in the GOM separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 63 apparent, several epiphytic species that were either not tallied in NLQ or appeared in very low amounts are conspicuous on these diagrams: these include Palmaria palmata, Neosiphonia harveyii (a Japanese invasive), Ceramium rubrum, Dumontia incrassata (S.G. Gmelin) Ruprecht, and Bonnemaisonia Figure 32. Macroalgal species at intermediate stations in the GOM separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 64 Northeastern Naturalist Vol. 18, Monograph No. 8 hamifera Hariot (an Asian invasive). All of these species have been recorded for much of the coast of Newfoundland, except for the northern peninsula, but clearly they are not important in biomass terms. Figure 33. Macroalgal species at protected stations in the GOM separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 65 The Gulf of Maine exposed-station curve in the Bray Curtis similarity ordination (Fig. 20) shows the sharp change of species assemblage from warm to cold types at 10 and 20 m. This is not readily apparent on Figure 28 for the GOM, in part because of reducing biomass with depth. However, it is easily seen in Figure 31, where species are separated by canopy, understory, and epiphytic/ small categories, and Agarum clathratum, Ptilota serrata, and Euthora cristata, respectively, become the dominating elements. Southern Nova Scotia (SNS) The southern Atlantic coastal waters of Nova Scotia (Fig. 19) are three to four degrees warmer than the Subarctic in winter, and typically a degree or two warmer than that of the Maine Coast (Fig. 2). It is also subject to greater exposure to large North Atlantic swells, from which the Maine Coast is somewhat protected, and generally has lower turbidity (and a resulting deeper photic zone). The composition of its benthic macrophytes (Figs. 34–36; Table 4) was quite similar to that in Maine (Figs. 28–30; Table 3) and differred markedly from that in NLQ (Figs. 21–23; Table 2) in generally the same ways as the Maine shore differed. Figure 34. Cumulative mean biomass of dominant (greater than 1% of total biomass for any one depth zone) macroalgal species in standard depth zones for exposed coasts in southern Nova Scotia (SNS). Species with less than 1% of total biomass are listed in Table 4. Total spp. compilation is presented in Appendix D(4). For species abbreviations, see Table 5. 66 Northeastern Naturalist Vol. 18, Monograph No. 8 Total macroalgal biomass in shallow water, roughly 5–10 kg/m2 at intermediate and exposed sites, was about the same as that in GOM and twice that in NLQ; it was also about the same as that reported for the Nova Scotian coast by Scheibling (1986) for sites fully recovered after sea urchin die-off. Agarum clathratum, the dominant kelp in NLQ, and 30% of total macroalgal biomass (Table 1), occurred at 1.4% of total biomass in SNS (as compared to 3.3% on the Maine Coast). Even in absolute biomass terms, Agarum occurred at one-quarter to one-half the levels of its occurence in NLQ, and abundance remained low below the kelp zone . Chondrus crispus, the dominant macrophyte in total biomass, at 23%, on the Maine Coast (but with only occasional, trace biomass in NLQ), occurred at 19% in SNS (even though the biomass [g/m2] was larger than that in Maine). This result was due to the very large increase of the kelp Laminaria digitata from 6th place in NLQ at 3.7% of total biomass and 4th place in Maine at 10.7% to first place in SNS at 34% of total biomass. Likewise, the prominent Chordaria flagelliformis of the sublittoral fringe in NLQ was a very minor species in SNS; along with its NLQ companions—Devaleraea ramentacea, Acrosiphonia arcta, and Dictyosiphon foeniculaceus—it is replaced Figure 35. Cumulative mean biomass of dominant (greater than 1% of total biomass for any one depth zone) macroalgal species in standard depth zones for intermediate coasts in SNS. Species with less than 1% of total biomass are listed in Table 4. Total spp. compilation is presented in Appendix D(4). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 67 by Chondrus crispus. The Saccharina spp., at about the same biomass as in Maine, rank considerably less in percentage in SNS because of the great increase in abundance of L. digitata and Chondrus crispus. However, Saccharina longicruris, so important in Maine at all exposures, occurred prominently only in more protected waters in SNS in this study. Alaria esculenta, in direct competition with Laminaria digitata, shows an even greater reduction from NLQ (15% of total biomass in NLQ to 7% in Maine and 0.6% in SNS). While the species Desmarestia viridis, Devaleraea ramentacea, Saccorhiza dermatodea, and Ptilota serrata were all moderately abundant in NLQ, they were minor species (or occurred at trace levels) in SNS (as was the case in GOM). As in Maine, the group of moderate-sized bushy red algae that fill in the blank understory mid-depth range of NLQ (i.e., Chondrus crispus, Phyllophora pseudoceranoides, Cystoclonium purpureum, Ceramium rubum, and Neosiphonia harveyii) were similarly important in SNS. However, Corallina officinalis, a very minor species in NLQ and at the one percent level of total biomass in Maine, was the 4th-ranked species at 6% in SNS. Most of the minor species occurring in Maine, but not northeastern Newfoundland, also occurred Figure 36. Cumulative mean biomass of dominant (greater than 1% of total biomass for any one depth zone) macroalgal species in standard depth zones for protected coasts in SNS. Species with less than 1% of total biomass are listed in Table 4. Total spp. compilation is presented in Appendix D(4). For species abbreviations, see Table 5. 68 Northeastern Naturalist Vol. 18, Monograph No. 8 in southern Nova Scotia. However, several additional warmer-water species also occurred at more than trace levels, including Ahnfeltia plicata, Polyides caprinus, Nemalion multifidum, Leathesia (difformis) marina, and Callithamnion tetragonum. Figure 37. Macroalgal species at exposed stations in SNS separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 69 The invasive species Fucus serratus, an infralittoral to mid-depth species that invaded Nova Scotia from Ireland and Scotland in the 19th century (Brawley et al. 2009), has successfully established throughout the southern GSL and on significant sections of the northeastern shore of NS and SNS. It was widespread and Figure 38. Macroalgal species at intermediate exposure stations in SNS separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 70 Northeastern Naturalist Vol. 18, Monograph No. 8 abundant enough to rank 5th in our species list at 5% of total biomass. Codium fragile, a late 20th-century invader to Atlantic Canada (Bird et al. 1993), was more scattered in its occurrence in SNS, but appeared in great enough abundance at a few of our stations to rank 6th at 3.7%. This number is somewhat questionable Figure 39. Macroalgal species at protected stations in SNS separated into canopy, understory, and epiphytic/small categories, and plotted as cumulative percent of biomass by standard depth zone. Total biomass is plotted for each category to the right. See text for discussion of the color coding: blue (Subarctic), purple (Boreal), white (cosmopolitan). For species abbreviations, see Table 5. 2011 W.H. Adey and L.-A.C. Hayek 71 because of its’ erratic occurrence and the great difficulty in producing a consistent wet-weight measurement for this siphonous species. On the seaweed assemblage diagrams for southern Nova Scotia (Figs. 37–39), the greater general biomass of Laminaria digitata, Chondrus crispus, and Corallina officinalis also stand out as significant differences from GOM. At exposed stations that have sufficient depth, the deepest end of the rocky bottom, shows the major elements remaining of northern waters: Agarum clathratum, Ptilota serrata, Scagelia pylaisaei, and Euthora cristata. Just as in GOM, this pattern is apparent when species are separated into canopy, understory, and epiphytic/small categories (Fig. 37), and is quite striking on the Bray-Curtis ordination curve for exposed shores (Fig. 20). Ptilota serrata, a major element of exposed and intermediate shores in NLQ, shows as the dominant understory species at 20 m in SNS; however, in total biomass at 10 and 20 m, it only appeared at 16% of its abundance in NLQ. Also, even though A. clathratum showed up moderately well at the deepest zones of exposed stations, the biomass was half of its level in NLQ and occurred as relatively small plants, scattered as individuals rather than being the savanna-former of NLQ. In protected waters, Agarum was further reduced, and Ptilota serrata was not found. The epiphyte Euthora (Callophyllis) cristata also appeared in deep water, although at very small biomass; Scagelia pylaisaei occurred in small amounts, but only at exposed stations. The invasive Codium fragile shows up as scattered “spikes” on the plots, indicating its erratic occurrence in the wild. On the other hand, Fucus serratus, in place in parts of Atlantic Nova Scotia for many decades, was more evenly distributed, occurring regularly in some regions of the coast, but still totally absent in others. Discussion This analysis of comparative macroalgal biomass is the second successful test of the thermogeographic model. The TM (Adey and Steneck 2001) demonstrates 20 regions of “sea water climate” (also called specific thermal regimes by those authors) that are persistent over moderate to extensive lengths of coastline through Pleistocene time (glacial and interglacial intervals). These abiotic TM regions predict many of the widely accepted shore biogeographic regions previously designated by biogeographers (e.g., Briggs 1974, 2003; Ekman 1953). Adey and Steneck (2001) also demonstrated quantitatively that the two principal TM-designated regions (Subarctic and Boreal) in the North Atlantic Ocean are occupied by two largely distinct groups of coralline red algae; these two groups intergrade with each other across the climatic transition zones between the regions. A major difference between the abiotic TM and classical shore biogeography based on organisms is that in the TM there are often significant transition zones between the model-specified regions (both in the North Atlantic and elsewhere). The data presented in this paper show that the fleshy macroalgal assemblage of the western North Atlantic Subarctic Region, like that of the corallines, is significantly different from seaweed assemblages to the south in the TMdesignated transitional zones, GOM and SNS (when compared on the basis of 72 Northeastern Naturalist Vol. 18, Monograph No. 8 biomass). As we have shown above, and will discuss below, all of the species intergrading with Subarctic seaweeds in the transitional zones have a shallow- water, European (Boreal) derivation except for a very few cosmopolitan species, known species introduced from distant shores, and a few smaller seaweeds with uncertain distributions, . Our results show that the North Atlantic Subarctic is a large and very distinctive region with its roots in the North Pacific. Even though species diversity is relatively low in this region, accumulating evidence suggests that its unique seaweed assemblage is part of a stable and rich ecosystem that has not been significantly affected by over-harvested, mostly offshore fisheries. The vast majority of the North American rocky shore ecological research to date has been carried out in the Gulf of Maine and Nova Scotia. This work had traditions based in Europe and its boreal flora and fauna. Even though a substantial body of research had been carried out in the Subarctic by scientists at Newfoundland and Quebec universities prior to our work, their research was largely ignored, and the very significant implications of the presence of the Atlantic Subarctic to evolutionary history and global change had remained unrecognized. The Subarctic is distinct from the Arctic During glacial episodes, as shown in the TM (Adey and Steneck 2001), Arctic rocky shore climate is concentrated in narrow transitional zones in the northernmost Pacific and Atlantic Oceans. Consequently, over Pleistocene time, significant areas of Arctic rocky shore are ephemeral, and as a result, the rocky, subtidal Arctic Region is very weak in biodiversity as compared to the Subarctic. While the Arctic Region has a distinctive flora (Adey et al. 2008), and there are certainly undescribed species, there are likely few genera. The strong difference between NLQ rocky bottom seaweed assemblages and those of SNS and GOM is not due to the southward inter-fingering of these Arctic species; only two significant species in these data, Turnerella pennyi and Devaleraea ramentacea, at about 2.7% of total biomass, have been considered to be Arctic in character (Adey et al. 2008) (and D. ramentacea is probably a Subarctic rather than an Arctic species). Another likely Arctic species, Neodilsea integra Tokida, was found in trace quantities at two stations. The prevailing biogeographic hypothesis (reviewed by Mathieson et al. 1991) has long accepted that the Strait of Belle Isle formed a northern boundary of a region that extended from Cape Cod to that Strait, where it met with the fringes of the Arctic. This investigation, as well as the study of Hooper and Whittick (1984) in central Labrador, shows that the entire southern half of the coast of Labrador is dominated by the same community that occupies the Atlantic Subarctic region that we describe in this paper. For the fully marine and estuarine stations in the Hooper and Whittick study at Makkovik Bay, Labrador (500 km north of the Strait of Belle Isle), the top ten Subarctic species of NLQ, as shown in Table 1, occurred at 67% of the stations, while the seven species regarded as Arctic by those authors occurred at only 27% of the stations. Also, of the seven species treated as Arctic by Hooper and Whittick (1984), only the deep water Turnerella pennyi occurred at more than one quarter of the stations. While this 2011 W.H. Adey and L.-A.C. Hayek 73 kind of presence/absence tally is not as definitive an indicator of biogeographic affinity as abundance in biomass terms, it securely demonstrates that the flora remains dominantly Subarctic well north of the Strait of Belle Isle. As Adey and Steneck (2001) showed for the crustose corallines, even in the northern quarter of the Labrador coast from Saglek Bay to Cape Chidley, only 38% of the coralline flora was Arctic in composition, the remainder being Subarctic (that data was collected in 1964, prior to significant warming; southern Labrador in that study was 95% Subarctic spp.). The sea-ice factor The unique seaweed assemblage that we describe here for the Atlantic Subarctic is not due to the presence of sea ice, although sea ice is certainly a factor in circumscribing the composition of the intertidal (Adey and Hayek 2005) and sublittoral fringe, the zone lying from the infralittoral to a meter or two below low water tide levels. Some of the species in this zone are annuals, quickly recovering space cleared by sea-ice abrasion in the late winter and spring (especially Chordaria flagelliformis). However, others, such as Alaria esculenta, are important perennials that may be heavily ice-scoured in the uppermost reaches of their range, but recover their dominance within a few years (Keats et al. 1985). As Keats showed, some Subarctic seaweed species are positively affected by ice scour (e.g., C. flagelliformis, Devaleraea ramentacea, Acrosiphonia arcta, and Dictyosiphon foeniculaceus), which removes the dominant Alaria , allowing these other species to recruit for a short time before Alaria reestablishes canopy. It is not necessarily the presence of ice, but rather the movement of that ice in an exposed seaway that plays the most significant role in forming seaweed assemblages. In mid-century, Stephenson and Stephenson (1954) compared seaweeds on the north and south shores of Prince Edward Island. The southern Gulf of St. Lawrence is a warm coast in summer, and as Adey and Steneck (2001) show, the subtidal portion of this area carries about the same number of Boreal species of corallines as the southernmost Gulf of Maine. However, extensive sea ice covers the entire area in most winters. As the Stephensons describe, Chondrus crispus is very abundant on the south coast, in the Northumberland Strait, where the ice is less mobile. On the north coast, open to northerly winds across a broad fetch in winter, moving ice and a weaker rock limits Chondrus crispus to patches, especially in hollows in the rock where ice scour is limited. Along parts of the open northern coast of the Gulf of St. Lawrence and in the many bays where shore-fast ice forms early and remains until well into the spring, without being moved by an ocean swell or strong northerly wind sea, perennials such as Fucus distichus can form long-lasting intertidal/infralittoral beds of macroalgae (Adey and Hayek 2005). Also, three species that are among the most dominant macroalgal species of the region (Agarum clathratum, Desmarestia viridis, and Ptilota serrata, with 44% of regional biomass) are mid-to-deeper-water species, unlikely to be affected by sea ice. Chondrus crispus is the dominant macroalga of the sublittoral fringe in the Gulf of Maine and Nova Scotia. It reaches into the margin of the Subarctic in decreasing abundance in southern and western Newfoundland. Because the 74 Northeastern Naturalist Vol. 18, Monograph No. 8 Subarctic Region is generally characterized by winter–spring sea ice, constant winter disturbance by ice abrasion—rather than the temperature parameters of the thermogeographic model—could be responsible for at least one of the key distinguishing features of the core Subarctic: a distinctive array of ephemeral sublittoral fringe species and the general absence of Chondrus crispus. The eastern Atlantic coast of Nova Scotia is characterized by rather warm late summer water temperatures (13–15 °C) and winter temperatures well above 0 °C. However, by late winter or early spring, local sea ice from Chedabucto Bay, as well as occasional sea-ice outwellings from the southern Gulf of St. Lawrence around Cape Breton Island have been known to reach as far southwest as Halifax (McCook and Chapman 1997). In the latter case, the described intertidal shore was extensively denuded by ice scour (including Chondrus crispus on the low shore). “Ephemeral algae” quickly covered this rocky shore; however, the original intertidal assemblage was mostly replaced within two years (Minchinton et al. 1997). In recent years, Gulf of St. Lawrence sea ice has only extended as an extensive and compact mass as far as central Cape Breton Island on the Atlantic east coast. However, several days of strong easterly/northeasterly winds during February and March, can bring loose pack ice inshore around Cape Canso and at least as far southwest as Tor Bay (Fig. 40). A series of infralittoral and 0.5-m depth stations were occupied in 2004 to compare the seaweed assemblage of the Nova Scotia coast east of Tor Bay with the western shore (southwest of Halifax). In the summer of 2004, the abrasive effects of sea ice were obvious in the eastern region at Louisbourg (which is frequently closed by ice in the late winter), but were also apparent on the shores of the outer islands off Canso and further west off Tor Bay. Evidence of scour appeared in the form of abundant bare patches on intertidal and shallow subtidal rock surfaces, especially ridges and pinnacles. Smooth rock domes, which in SNS would carry a dense carpet of Chondrus crispus, were generally bare or even polished rock, whereas adjacent crevices carried dense growth of this seaweed. These physical characteristics are reflected in the northeastern Nova Scotia column of Table 6, in the form of a 50% reduction in total collected biomass. Chondrus crispus correspondingly showed a nearly 50% reduction in biomass, as compared to SNS, presumably due to ice abrasion. However, at the eastern Nova Scotian stations, it retained about the same percentage of total biomass as in the ice-free western region. Chordaria flagelliformis, the annual filiform brown alga, was more abundant on this portion of the coast of Nova Scotia that experiences occasional sea ice than it was in southern Nova Scotia, but it was still less abundant than Chondrus and an order of magnitude lower in abundance than in the Subarctic. This entire subject needs additional documentation, but it strongly suggests that Chondrus crispus, in its characteristic infralittoral/sublittoral fringe zone of dominance, is limited in abundance by wave-driven sea ice, along with many other seaweeds. However, as demonstrated by its abundance in the Northumberland Strait, it is not blocked from the Subarctic by winter sea ice. Perennial Subarctic species, such as Fucus distichus in the lowest intertidal and infralittoral and Alaria esculenta in the shallow sublittoral fringe, survive sea-ice scour in crevices and, as 2011 W.H. Adey and L.-A.C. Hayek 75 Figure 40. Sea-ice state in the Gulf of St. Lawrence and eastern Nova Scotia on 18 February 2004. Eighty to one hundred percent dense sea ice occupied the coast from Cape Canso to Tor Bay and west as far as Country Harbour. By 23 February, the dense ice had moved offshore, leaving 40–60% coverage, and by 1 March, only thin new ice remained. While the impact of dense sea ice likely lasted only a few weeks, this duration would have been quite sufficient to provide the extensive scour of intertidal and uppermost subtidal rock surface seen at both Cape Canso and Tor Bay in the summer of 2004. Map courtesy of Canadian Ice Service. 76 Northeastern Naturalist Vol. 18, Monograph No. 8 Table 6. Sublittoral fringe - regional comparison of exposed and intermediate stations (infra = mlw neap tides - mlw spring tides). For each species listed, biomass (g[wet]/m2) and percent of total biomass (in parentheses) is given. Tr = trace. Ice/no ice-5 yrs, Nova Scotia, Infra + 0.5 m Cornwall/ 0–2 m Subarctic Subarctic Gulf of Maine ne NS sw NS Devon, Conception Bay, NF 0.5 m Infra + 0.5 m Infra + 0.5 m sea ice no sea ice England, Infra Keats et al. (1985) (12 stations) (12 stations) (6 stations) (8 stations) (8 stations) Lewis (1964) Subarctic spp. Fucus distichus - 178 (5) 639 (17) 58 (1) 40 (0.5) 369 (3) 0 Devaleraea ramentacea 58 (1) 211 (6) 253 (7) 64 (1) Tr 9 (Tr) 0 Acrosiphonia arcta 51 (1) 68 (2) Tr 122 (2) 35 (0.5) Tr 0 Alaria esculenta 3418 (83) 1682 (48) 891 (24) 1059 (17) 750 (10) 64 (0.5) 0 Chordaria flagelliformis 240 (6) 777 (22) 1332 (36) 6 (0.1) 171 (2) 5 (Tr) 0 Dictyosiphon foeniculaceus 21 (0.5) 151 (4) 178 (5) 0 6 (Tr) 0 0 Saccorhiza dermatodea 257 (6) 327 (9) 176 (5) 18 (0.3) 589 (8) 0 0 Saccharina longicruris - - - 263 (4) 0 0 0 Pilayella littoralis 2 (Tr) 2 (Tr) 45 (1) 0 5 (Tr) Tr 0 Desmarestia viridis 4 (Tr) 55 (2) 81 (2) Tr 67 (0.9) Tr 0 Agarum clathratum - 54 (2) 27 (0.7) 15 (0.2) 0 0 0 % subarctic of total biomass (98) (99) (98) (26) (22) (3) (0) Boreal spp. Chondrus crispus (inc. Mast. stell.) - Tr Tr 2132 (34) 2038 (27) 3939 (28) (17) Laminaria digitata 65 (2) 32 (0.9) 18 (0.5) 1605 (25) 0 6091 (44) (20) Corallina officinalis - 0 Tr 105 (2) 557 (7) 1011 (7) (8) Fucus serratus 0 0 0 0 224 (3) 47 (0.3) (21) Palmaria palmata 18 (0.4) Tr Tr 26 (0.4) 0 95 (0.7) (3) Cystoclonium purpurem - 0 0 198 (3) 0 32 (0.2) 0 Himanthalia elongata 0 0 0 0 0 0 (2) Laurencia spp. 0 0 0 0 0 0 (23) Leathesia difformis 0 0 0 0 6 (Tr) 23 (0.1) (5) % boreal of total biomass (2) (1) (0.5) (62) (37) (80) (99) Cosmopolitan spp. Saccharina latissima - - 17 (0.5) 644 (10) 3022 (40) 2099 (15) 0 Rhodomela confervoides - - Tr 4 (Tr) 42 (0.5) 25 (0.2) 0 Total Biomass (all spp.) 4134 3538 3657 6319 7552 13,809 2011 W.H. Adey and L.-A.C. Hayek 77 noted by Adey and Hayek (2005), sometimes on broad rocky flats, while Chondrus crispus does not. Thus, it seems likely that it is unable to compete with the many Subarctic species of this zone because of low summer temperatures. As we have shown, Chondrus crispus, the dominant understory macroalga on GOM and SNS coasts, does occur in moderate abundance in Newfoundland on the south and west coasts (the former never having sea ice and the latter with occasional winter/spring sea-ice incursions) outside of the boundaries of the TMdefined Subarctic. Chondrus crispus also occurs on the Subarctic northeastern coast of Newfoundland as scattered individuals, especially at mid-bay locations. However, an extensive search of the northern Gulf of St. Lawrence Coast did not locate a single plant. Yet, as noted on this same sea-ice coast, the perennial Fucus distichus can be abundant across rocky flats and in crevices in the infralittoral and sublittoral fringe. Fucus distichus occupies the same basic shore zones as Chondrus crispus. Scouring sea ice can remove macroalgae from the intertidal and uppermost sublittoral of rocky shores. However, because of the very irregular rock surfaces in most of this region, considerable amounts of macroalgae can remain from season to season in crevices. Also, around many complex bay island shores—especially on northern shores—where shore-fast ice forms early and remains late, shore-fast ice can protect the shore from moving ice abrasion. Do industrial fishing and Strongylocentrotus create a faux region? As many researchers have shown, the sea urchin S. droebachiensis is a significant grazer of seaweeds, especially some species of kelps (e.g., Himmelman and Nedelec 1990, Miller 1985a). In abundance, it ranges from mid-Labrador to the western Gulf of Maine. As reviewed in our Introduction, it is widely accepted that a protozoan disease produces boom-and-bust cycles of S. droebachiensis south of Newfoundland, but not in Newfoundland, the northern GSL, and further to the north. Indeed, infected urchins can be seen to recover if transferred to water temperatures <12 °C (Scheibling and Hatcher 2001). Could it be that S. droebachiensis, free from its parasite in the Atlantic Subarctic Region and freed from finfish predation by overfishing, is solely responsible for a recent development of the “coralline urchin barren” or kelp savanna character of this region? It has been proposed (reviewed by Steneck et al. 2002) that the “barrens” status was created by industrial fishing. Because of the severe socioeconomic problems created by the collapse of the northern cod stocks in Newfoundland and Labrador in the late 1980s, the catch and stocks of cod over the past century were documented recently (Harris 1990, Myers and Cadigan 1995, Taggart et al. 1994). Since the middle of the 19th century, cod landings from northeastern Newfoundland and southern Labrador (including the off-lying shelf) remained remarkably stable, with about a 60–70% increase over a century and multi-year fluctuations of 10–20%. This was mostly a fixed-gear (cod traps) and largely inshore fishery until the 20th century. Beginning in 1960, with the addition of large-scale trawler and factoryship fishing on the Banks, cod landings markedly increased. However, depending upon the region, peak catches, at levels of three times the traditional level, were not achieved until the late 1960s or early 1970s (Harris 1990). One of us (W.H. Adey) surveyed by SCUBA the entire coast of Atlantic Nova Scotia, Newfoundland, and 78 Northeastern Naturalist Vol. 18, Monograph No. 8 the Gulf of St. Lawrence, working north to Cape Chidley, Labrador, in 1964 (Adey 1966). Based on station documentation and underwater photography taken at the time, there has been little obvious change in the “coralline barrens” character of these rocky bottoms during the last 45 years. Coring and dating of sea urchin spines in sediments (Hooper 1980) suggests that the system has been at least semi-stable for much longer. There was still an enormous biomass of codfish in this region when W.H. Adey carried out his regional bottom survey in 1964, and the inshore fishery, where the surveyed bottoms lay, doubled its catch between 1964 and 1968 (Pinhorn 1984). Surveys of fish populations in the Gulf of St. Lawrence also suggest that significant depletion of large, predatory fish by fisheries’ activity occurred considerably later (1980s and 1990s) than documented coralline/urchin “barrens” (Benoit and Swain 2008). Moreover, as we discussed earlier in our treatment of the S. droebachiensis, the considerable evidence from cod stomach analyses suggests that northern cod only rarely eat sea urchins. As further evidence relating to this issue, one of the abundant crustose corallines in the Subarctic, Clathromorphum compactum, is characterized by winter reproduction, well-defined yearly anatomical layering, and considerable thickness (Adey 1965). Electron microprobe/laser ablation/MS techniques applied to C. compactum’s layers and high magnesium wall structure has characterized growth-rate and sea-temperature patterns ranging over the last century (Halfar et al. 2011, Hetzinger et al. 2009). Specimens for these analyses derive from 1960s USNH Subarctic collections, as well as more recent field trips. This study is on-going, but clearly demonstrates that there has been little change in C. compactum growth over the last century that is not explained by temperature alone. Clathromorphum compactum is primarily a mid-depth (5–20 m) species in the Subarctic Region (Adey 1966), generally occurring below the shallow and dense Alaria esculenta kelp zone, and within the Agarum/ Desmarestia/ Strongylocentrotus “barren” or savanna. This continuity of growth suggests that dense mid-depth development of Laminaria or Saccharina kelps that would have provided shading and slower growth for coralline algae was not present earlier in the century. Strongylocentrotus droebachienis, the only grazing echinoid in the Subarctic Region, is more or less abundant at all times. In Nova Scotia and Maine, removal of sea urchins by disease, by humans in experiments, or in harvest for roe produces a great increase in abundance of Laminaria and Saccharina kelp species (Steneck et al. 2008). On those shores, sea urchins when moderately abundant are known to consume about 10–20% of yearly production of kelp (Campbell 2004). However, much of the standing crop built up during the previous year is delivered to beach drift as wrack by winter storms (Steneck et al. 2001, Vadas et al. 2004). Through amphipod and bacterial breakdown in the wrack and re-delivery to coastal waters as organic particulates during spring tides and storms, kelp and other macroalgae provide organic particulates to coastal ecosystems. Rapid spring and early summer growth by Laminaria and Saccharina kelps maintain a balance with these degrading elements, and kelp forests normally remain dense. 2011 W.H. Adey and L.-A.C. Hayek 79 This balance can be disrupted even on the outer coast by sea urchin population explosions. However, in protected waters, even moderate sea urchin populations can produce coralline/urchin “barrens”. When one of us (W.H. Adey) worked along the entire Gulf of Maine and southern Nova Scotia coast in the early 1960s, kelp forests were abundant on the outer coast and along the outer shores of smaller bays. However, in the larger bays (Penobscot, Blue Hill, and Frenchman bays), abundant coralline/sea urchin “barrens” were present. In contrast, Himmelman et al. (1983), working in the western ranges of the Subarctic Core Region, and Keats (1986), working in the eastern end (Fig 19), showed that it is dense beds of Alaria esculenta that occupy shallow exposed shores, and it is this same species, not Laminaria or Saccharina, that develops as kelp forest when sea urchins are removed from shallow rocky “barrens”. Also, as Himmelman (1991) showed for the Mingan Islands in the northern Gulf of St. Lawrence (Fig. 6), and as Figure 41 implies, even deeper rocky bottoms are typically more of a rich seaweed “savanna” than a barren. Key to this state is the three dominant species (Agarum clathratum, Desmarestia viridis, and Ptilota serrata) in the Subarctic region that make up 44% of the total macroalgal biomass (Table 1). These species are all chemically (and perhaps physically) protected from sea urchin grazing (Himmelman and Nedelec 1990). The same three species make up only 4.5% and 2% of the biomass in the GOM and SNS, respectively. Figure 41. Typical sea-urchin coralline patch or “barren” in a savanna of Agarum clathratum, with a patch of Ptilota serrata (lower right). The four visible larger asteroids are Leptasterias polaris. Photo taken at 10–20 m depth at Wall Island, an exposed site near Battle Harbour in southern Labrador. 80 Northeastern Naturalist Vol. 18, Monograph No. 8 Even the numerical abundance of all three species is reduced to 15% and 11% of its NLQ abundance in the GOM and SNS, respectively. If Strongylocentrotus droebachiensis is not present, the energy that these three protected species invest in chemical defenses puts them at a competitive growth disadvantage with other larger macroalgae. However, this Subarctic array of protected seaweeds is not produced by S. droebachiensis alone. As Figures 20, 28, 31, 34, and 37 demonstrate, in colder, deeper water in GOM and SNS, Agarum clathratum and Ptilota serrata (and to a lesser extent, Desmarestia viridis) replace the dominant Laminaria and Saccharina kelps and Cystoclonium purpureum and Phyllophora pseudoceranoides, the bushy red understory of midand shallow depths. This pattern occurs even when sea urchins are not abundant. In NLQ, where S. droebachiensis is always present on rocky shores, and where temperatures are colder from surface to bottom, these same species are mostly free from the competition of the Laminaria and Saccharina kelp species. Agarum clathratum can expand against intensive sea urchin grazing (Gagnon et al. 2005) and has been shown to block movement of sea urchin fronts (Gagnon et al. 2004). Ptilota serrata seems adapted to deep-water conditions on both exposed and intermediate shores, and occurs in the shallower, in-bay environments in only small quantities. Agarum and Desmarestia can become shallow-water dominants on protected bay shores in NLQ where sea urchins, free of wave action, are able to operate nearly to the low water line. The relationship between Strongylocentrotus spp. and the kelps it grazes, mediated by the predatory Enhydra lutris (L.) (Sea Otter), has been widely studied in the North Pacific, especially in the Aleutian Islands (Estes and Duggins 1995, Estes and Steinberg 1988, Steneck et al. 2002). When Sea Otters are abundant, sea urchins (several Strongylocentrotus species, including S. droebachiensis) are either small or absent in shallow water, and kelps are abundant. In shallow water, Alaria fistulosa occurs as a canopy, and is underlain by several Laminaria species, while in deeper water (>10 m), Agarum (cribrosum) clathratum and Strongylocentrotus spp. coexist (Dayton 1975). When the Sea Otter is absent, due to hunting or other reasons, large sea urchins reach into shallow water and the shallow kelp beds are decimated, leaving the deeper-water Agarum beds. If the predecessors of North Pacific kelp communities were to be filtered through the Canadian Archipelago during interglacials, leaving behind the Sea Otter, the seaweed assemblages and the state of the kelp communities that we now see in the Atlantic Subarctic would seem to be a likely result. It also seems likely that the Sea Otter could not make the Northwest Passage in one season, and then could not survive a northern Canadian winter. Most of the subtidal invertebrate and algal species of the ecosystem would not be so encumbered. The Aleutians are only fringe Subarctic (Fig. 2), and unlike the northwestern Atlantic, winter seawater temperatures are not near or below 0 °C with extensive sea ice. Adding this modification to a filtered Aleutian’s assemblage provides an ice-modified, high-energy, shallow kelp zone that is highly dominated by the fast-growing perennial kelp Alaria esculenta and a wide variety of annual seaweeds. In the Atlantic Subarctic, on intermediate and exposed shores, S. droebachiensis is unable to survive some 2011 W.H. Adey and L.-A.C. Hayek 81 combination of high-wave energy and low winter temperatures to prey heavily on the shallow kelp zone. The deeper zones are dominated by protected species (especially Agarum clathratum), just as in the North Pacific. It is only the Laminaria (and Saccharina) mid-depth spp. that are absent from rocky shores (and restricted to refugia); this scenario is very much what one might predict by simply excluding Sea Otters in the Canadian Archipelago filter. Rocky pinnacle macroalgal assemblages Typical NLQ bay rocky shores present a shallow sea urchin/coralline savanna, where Agarum clathratum and Desmarestia viridis rise to quite shallow water. However, in some protected bay localities, a few hundred meters away from such shores, rocky mounds free of urchins can have abundant Saccharina kelps with reduced A. clathratum and D. viridis. Although the metabolism of S. droebachiensis seems little affected by lowered winter temperatures, most urchins retreat to deeper waters in the winter or hide deep in crevices on the lower shore. On exposed shores, this is probably to escape the higher average wave action of the winter; in bays, it may be to escape the lowered salinities of spring, since echinoderms are stenohaline. In the summer, they move landward, often as aggregations or “fronts” of urchins grazing as they go. However, their ability to transit sedimentary bottoms, mud to gravel, appears limited, and provides the potential for isolated rocky pinnacles, especially in bays, to remain Green Urchin-free, and colonized by Saccharina kelps. Although it remains a subject of some contention (McNaught 1999, Scheibling and Hatcher 2001), sea urchin larvae appear to select coralline rather than kelp bottoms for settlement. This behavior may be adaptive in that a kelp bed could harbor many predators of recruiting sea urchin larvae. In addition, a coralline bottom, largely of Clathromorphum spp., would provide minimal but largely uncontested feeding grounds, in the form of a thick, photosynthetic epithallium, often coated with diatoms and cyanobacteria (Himmelman 1986). However, if S. droebachiensis cannot reach a sediment-isolated peak from below and will not settle on kelp beds, once kelp is established, such peaks will remain largely sea urchin-free. It would be useful to perform sea urchin-removal experiments on protected rocky shores that lie near rocky pinnacles, because such shores are mostly coralline “barrens” with high urchin populations. The TM shows that the exposed Subarctic shores discussed here have summer surface temperatures of 9–10 °C. Surface to bottom temperatures were routinely taken during station dives, and although these are snapshots, the mean of 11 exposed stations also provides a surface temperature of 10.0 °C (the thermocline is often sharp, dropping to 8.3 °C at 10–20 m and 3.9 °C at 30 m). However, the intermediate to protected (Bay) stations provided both surface and 10–20-m depth mean temperatures 1–2 °C higher than exposed stations, and this is approaching the 12–15 °C summer surface temperatures level of GOM and SNS. Higher summer temperatures and their effects on the growth rates of Saccharina species may well be a factor in determining rocky pinnacle seaweed assemblages. It would be interesting to search in NLQ for offshore rocky pinnacles (where summer 82 Northeastern Naturalist Vol. 18, Monograph No. 8 temperatures at 10–20 m will typically be in the 5–8 °C range) surrounded by gravel beds to determine whether or not the Saccharina spp. (instead of Agarum and Desmarestia) will appear as dominants immediately below the Alaria zone. Saccharina latissima appears to be a generalist kelp of cosmopolitan distribution in the colder northern hemisphere. It will tend to “fill-in” where marginal niches are open. It is common in the European Boreal kelp forests, but generally less abundant than Laminaria digitata at shallower depths, and Laminaria hyperborea at deeper locations (Connor et al. 1997). In warmer Subarctic bays, where Strongylocentrotus droebachiensis can be limited by sediment barriers from rocky pinnacles, docks, and breakwaters, S. latissma can find suitable habitat without significant competition from Alaria esculenta. In these situations, the slow-growing Agarum clathratum is also a poor competitor. At 10 m on NLQ in-bay rocky pinnacles, A. clathratum provided a mean biomass of about 1600 g/m2, about 25% of the total biomass (which was mostly Saccharina spp.). As we noted earlier, these are probably multi-year plants. On in-bay rocky shores at 10 m, the macroalgal biomass was over 90% A. clathratum; however, the mean biomass was only 183 g/m2, indicating that when S. droebachiensis has few, if any, restrictions, even A. clathratum can be limited in standing crop. In the glaciated western Atlantic, many protected sites tend to be silty and turbid, and selective pressure to develop a long, rigid stipe to elevate the kelp blade in the water column appears to be a factor in the ongoing evolution of Saccharina longicruris. The deeper transition from S. latissima to S. longicruris on rocky pinnacles (Fig. 42), where considerably heavier silt loads are to be expected, appears indicative of such a process. Further field and genetic studies could provide a better understanding of these issues. Seaweed assemblages of the North Atlantic Subarctic The information now available from across the entire western Atlantic Subarctic and outlined in tables and figures in this paper, is presented in Figure 42 as a modification of Himmelman’s 1985 sketch (Fig. 5) of rocky shore bottoms on the south and west coasts of Newfoundland. The rocky pinnacle shown to the right represents a wide variety of types, including docks and breakwaters, from shallow to deep, widely scattered in bay environments. It comprises a very small fraction of the area of the rocky shore environments that the primary diagram represents. Also added to Himmelman’s diagram: Saccorhiza dermatodea occurs at low biomass but is a consistent element of the Alaria forest and is occasionally found as very large, multiyear plants (Keats and South 1985). In addition, Desmarestia viridis is an abundant element of the coralline/urchin savanna, along with Agarum clathratum and Ptilota serrata on exposed and intermediate shores. A Turnerella/Ptilota/coralline assemblage is common where rocky bottom continues below about 25 m at exposed sites and somewhat shallower in protected bay environments. Turnerella pennyi appears to be a protected species as well (Himmelman and Nedelec 1990), which is partly responsible for its consistent presence in a zone likely to be S. droebachiensis winter habitat. Although Agarum clathratum, Ptilota serrata, and Desmarestia viridis can all occur as both isolated and widely integrated patches, it is not possible without a 2011 W.H. Adey and L.-A.C. Hayek 83 Figure 42. Schematic of North Atlantic Subarctic rocky shore community revised after Himmelman (1985). The original diagram drawn for south and west Newfoundland has been modified for the core Subarctic to show the typical sedimentary restriction of rocky bottoms (generally decreasing in depth in protected waters), the general switch to Turnerella pennyi bottom below 25–30 m, the often great abundance of Desmarestia viridis, the sometimes extended Agarum and Ptilota beds, and the minor, but consistent appearance of Saccorhiza dermatodea in the Alaria kelp beds. The rocky pinnacle, or Saccharina refugium, has been added as a representative of a wide range of isolated hard bottoms in protected bays. These pinnacles have small total area, often have few sea urchins, and support a dense kelp community. very extensive survey to provide a quantitative assessment of total area covered. However, at two randomly selected areas (where weather or other delays were encountered during our field work), we were able to spend enough bottom time to map several exposed (Fig. 43) and intermediate and protected (Fig. 44) shores to provide a glimpse of the complex mosaic that is present on most shores. Along with the random sampling of the NLQ region that we describe in this paper, and the Himmelman (1991) and Keats (1986) studies, these figures provide a realistic view of the core Subarctic region for comparison with that from the Subarctic/ Boreal regions of the Gulf of Maine as reviewed by Mathieson et al. (1991) and presented earlier in Figure 4. At the exposed Cape Daumalen site, all rock ledge below about 2 m depth is densely covered with a kelp assemblage, shifting from Alaria esculenta, with mixed Laminaria digitata, to A. clathratum at 7–8 m and then Agarum/Ptilota/Turnerella below about 20 m. It is only the unstable, gravel/ pebble slope that is a coralline “barren” (with few urchins). At the more protected Red Island site, the side of the island with intermediate exposure is largely vegetated with kelps (Alaria esculenta and A. clathratum) and Desmarestia viridis. 84 Northeastern Naturalist Vol. 18, Monograph No. 8 However, the lower-sloped, protected side of the island is Agarum/Desmarestia “savanna” with roughly 60% coralline urchin “barren”. The North Pacific Subarctic vs the North Atlantic Subarctic As shown by Adey and Steneck (2001), the North Atlantic Subarctic is clearly a geographically extended arm of the very large North Pacific Subarctic (see also Cunningham 2008). With virtually the same seawater climates (see Fig. 2), the two subregions are separated by roughly 6000 km, the Canadian section being mostly rocky coastline of transitional Subarctic/Arctic water climate (generally ice-free for a few months in late summer). The north Alaskan/Canadian coast as far east as the Mackenzie River delta is mostly a coast of sandy silty sediments, although rocky patches are known (Dunton and Schell 1986, Dunton et al. 1982). It is this 1500-km Alaskan and westernmost Canadian section that provides the primary barrier to exchange of Subarctic, subtidal, rocky shore biota. All of the dominant seaweed species of the Atlantic Subarctic also occur in the North Pacific and are known to occur across the Arctic Canadian Archipelago (Lee 1980). Figure 43. Map of rocky bottom with an average slope of about 45° at Cape Daumalen off Canada Bay on the northeastern side of Newfoundland’s Northern Peninsula. The gravel-cobble slope, to the left, and area below the Agarum/Ptilota beds is primarily a mobile coralline encrusted “talus”. The upper part of the Alaria/L.digitata zone is almost entirely Alaria in composition; the lower end is dominated by L. digitata. In late June at this site, surface temperatures were about 8 °C (about the same as in the previous year in mid-August). However, below 10 m, temperatures drop to 5 °C, with 3.5 °C being recorded at 50 m. The NW/SE length of the section is about 30 m. 2011 W.H. Adey and L.-A.C. Hayek 85 Most of these species, where information is available, have an ancestry in the colder North Pacific, as we have noted at several points, and most of their congeners currently reside in the North Pacific (see also Coyer et al. 2006). The same is true of the principal grazer of the seaweeds, Strongylocentrotus droebachiensis; the separate populations of that species show evidence of sporadic gene interchange (Addison and Hart 2005). The Northwest Passage between the oceans through the Canadian Archipelago (including the shorter channel through Hudson Strait, Foxe Basin, and the Fury and Hecla and Bellot Straits), is the most direct shallow-water connection between the subregions. This passage would have been open following deglaciation; at about 7500 years BP—during post-glacial, pre-rebound time—much of the passage would have had 100–200 m deeper water (and correspondingly wider passages) than at present (Dawson 1992). The period 6000–7500 years BP would Figure 44. Map of rocky bottom seaweed assemblages around Red Island at the inner end of the St. Lunaire Road entrance channel (about 80 km north of the site mapped in Fig. 43). The right or ESE side of the rocky island (which is about 150 m in largest dimension) is open to North Atlantic swell (25° at 2 km), but largely protected from local wind sea; it is an intermediate site (see Appendix C), whereas the northwest side of the island is quite protected, even from local wind waves for the portion within St. Lunaire Road. The non-patterned surface represents coralline/urchin bottom. Below 10 m (not shown), the continuing, largely coralline-covered rocky bottom is sparsely vegetated with Turnerella pennyi, Ptilota serrata, and Phycodrys riggii. 86 Northeastern Naturalist Vol. 18, Monograph No. 8 also have corresponded with the climatic optimum or hypsithermal (Strong and Hills 2010), which would have produced warmer temperatures and likely more extensive summer ice-free waters in the Northwest Passage. Although the seawater climate in the Northwest Passage has been suitable for the transit (either way) of subtidal Subarctic species through at least some of most interglacials, many seaweeds have not made the passage (Adey et al. 2008). It would also appear that extreme seasonality, coupled with the “sediment barrier” of the north Alaskan coast and Mackenzie delta, has made it difficult for the Sea Otter, the principal higher predator of S. droebachiensis, to make the passage, at least during the latest interglacial. Thus, the Northwest Passage has acted as a filter and the determinant of the dominant character of the North Atlantic Subarctic benthic seaweed assemblages. It is also likely that below the sublittoral fringe and shallow kelp community (where wave action is a deterrent to S. droebachiensis), the interaction between protected Subarctic seaweeds and their grazer may limit the transit of additional, unprotected Subarctic species. The Pacific Subarctic is of very high biodiversity (Briggs 2003) and is particularly rich in bushy red algae (Adey et al. 2008, Perestenko 1996). The “missing understory” at the junction of the shallow and deep kelp zones, which we discussed earlier, may relate to the Northwest Passage filter. Quantitative research on seaweeds through the entire Northwest Passage, and into the western Bering and Okhotsk Seas, is essential to gaining an understanding of these issues. This research needs to be accomplished before global warming has had a significant impact. The history of the Subarctic vs Boreal regions and their transition zone There is a strong correspondence between the TM, classical biogeography, the distribution of coralline algae, and (with this paper) fleshy seaweeds. These relationships allows us to develop hypotheses for the Pleistocene history of the two principal northern North Atlantic biogeographic regions (Subarctic and Boreal), as well as the transition zone now occupying SNS and GOM. The Subarctic Region is based in the northwestern North Pacific, including the western Bering Sea. Most of the dominant Atlantic Subarctic genera and many species occur in both the North Atlantic and the North Pacific, as well as through the Canadian Archipelago. During glacial times, the western Atlantic Subarctic was cut off from the North Pacific; rocky shoreline length was greatly reduced and likely limited to cobble/boulder patches along the glacially created fore-bulge of the continental shelf from the current Grand Banks to the continental shelf off Cape Cod (Pielou 1991, Shaw et al. 2002). In the North Pacific, the shorelines of the Okhotsk Sea, Kamchatka, and the southwestern Bering Sea shifted to accommodate sea-level fall, but were not significantly impacted by continental glaciers (Anderson and Borns 1994). A very large Subarctic rocky shoreline, supporting Subarctic species, would have existed in the Pacific Ocean throughout Pleistocene time, with populations migrating up and down the shoreline with the rise and fall of sea level. Little latitudinal shift occurred (Adey and Steneck 2001). Across the Atlantic Ocean during glacial intervals, the British Isles were also subject to major glaciation and its peripheral effects. The waters off the present-day English Channel were Arctic in character. The North Atlantic 2011 W.H. Adey and L.-A.C. Hayek 87 Boreal, centered on the British Isles and Bay of Biscay during the mid- to later Holocene, had previously migrated south to the western Mediterranean during glaciations. Nevertheless, there is considerable contiguous, rocky shoreline to support a continuous biogeographic region slowly migrating throughout the cycling of the Pleistocene. In the western Atlantic, the Laurentide Continental Glacier covered the entire coastline that currently contains both the Subarctic biota and the transitional zones to the south. As noted above, some of this shore would have been sand and gravel, but due to local glacial advances, boulder patches were also likely on the outer shore. Because of the very steep ocean temperature gradient, these shorelines would have been more Subarctic than Arctic in character (Climap Project Members 1976). Although certainly smaller than the current Subarctic rocky shore, there would have likely been local refugia for Subarctic biota. During each interglacial time, these refugia would have provided the source populations for seaweeds to spread along the growing northwestern Atlantic rocky coast. During the mid-Holocene, as we have described, these residual populations would have been supplemented by new injections of Pacific Subarctic species through the Canadian Archipelago. In contrast, the Boreal elements of the western Atlantic rocky shore flora would have been moved south with each encroaching glaciation, on the shelf, off the shore of what is now the New Jersey to Cape Hatteras area, to find a supporting climate. Never having been glaciated, this shore—a sandy coastline with very scattered outcroppings of soft tertiary rock—would have generally been inhospitable for seaweeds. Rocky refugia, if present, would have been few and far between. Thus, it seems likely, as each interglacial cycle began (in the Holocene roughly 8000–10,000 yr. BP, with rebound complete and shores stabilized at about -60 m [Toscano and Macintyre 2005]), the northwestern Atlantic rocky shore would have been occupied almost entirely by a Subarctic flora from local refugia. From roughly 6000–7500 yr BP, this flora would have been supplemented by migrating North Pacific Subarctic species, as the Northwest Passage, initially very deep and extensive due to a glacial depression of hundreds of meters, was relieved of its continental ice sheet. The concurrent hypsithermal warming would have enhanced this process. Later in the Holocene, with stabilizing shorelines (on both sides of the North Atlantic), current-derived drift specimens of Boreal seaweeds would have begun to arrive slowly in the western Atlantic transitional areas (Maggs et al. 2008). Using genetic analyses, several common invertebrates of Atlantic rocky shores have been identified as having a European Boreal origin (e.g., Wares 2002, Wares and Cunningham 2001). Some are tied to early historical passage, while others such as Semibalanas balanoides (L.) (Northern Rock Barnacle) are thought to have arrived before the LGM. This species is primarily intertidal and attaches to any hard surface, including large mollusk shells; it is among the most likely of species to have survived in limited southern refugia mostly occupied by sandy/shelly shores. It is probable that only with the arrival of Europeans beginning in the 16th century, with their wooden boats and ballast, 88 Northeastern Naturalist Vol. 18, Monograph No. 8 could the rapid influx of Boreal flora begin. Although considerable additional genetic analysis of a wider range of Boreal-origin seaweeds is necessary, the New England and Nova Scotian rocky shore biota may represent a case of massive anthropogenic bridging of an oceanic biogeographic barrier. Comparative analysis of seaweeds of Subarctic and Boreal regions In earlier figures describing seaweed assemblages by depth and region, species were color-coded. These colors represent an estimated biogeographic status of each species, and are dark blue (Arctic), blue (Subarctic), purple (Boreal), and white (cosmopolitan). The cross-hatching indicates species that range somewhat beyond their designated region. Following the pattern exhibited by the corallines, a putative Subarctic macroalgal species abundant in the western Atlantic and northwestern Pacific and extending into the Arctic, if wide-ranging, would be expected to reduce its abundance southwards in the British Isles and would not extend south of the English Channel. An eastern Atlantic Boreal species following post-glacial transport across the North Atlantic could expand throughout the Subarctic fringe zone from Cape Cod to the southern Gulf of St. Lawrence and south and west Newfoundland. However, such a species would occur in very limited abundance in the core Subarctic, and would not extend into the Arctic. The following distribution information is based on Lüning (1990) and Algaebase.org for September 2010 (with due regard for the likely inaccuracy of occasional outlying citations). In Figures 45–47, comparative plots of macroalgal community structure, with species separated by canopy, understory and epiphytic categories, and based on percent cover relative to depth, are shown for the three regions. Such separation allows a fuller comparison of species composition with depth and exposure between the three regions under consideration. It is immediately apparent for all three exposures and levels in the community (canopy, understory, epiphyticsmall) that the NLQ region is dominantly occupied by colder-trending Subarctic (rather than Arctic) species with only a small fraction of warmer-water species in mid-depths. The GOM and SNS regions, on the other hand, are rich in warmerwater species at all exposures. It is particularly striking that the understory flora of GOM and SNS, except in deeper, colder waters, is almost exclusively made up of warm-water species. In general, exposed stations of both GOM and SNS have macroalgal-coated rock to greater depths than the protected or intermediate stations. The deeper zones, especially of GOM exposed stations, reach into colder waters and have high percentages of Subarctic species (Figs. 20, 45–47). The deepest zone of exposed SNS stations likewise has a flora that is dominantly Subarctic in origin. For the epiphytic flora, the more erratic occurrence of these species, coupled with taxonomic uncertainty and a lack of a full geographic understanding, make the comparisons more difficult. It is noteworthy that no single individual of two of the most prominent cosmopolitan species in this group (Ceramium rubrum and Neosiphonia harveyii, in GOM and SNS) occurred in these collections in NLQ (although both had been collected there in the past; South and Hooper 1980). Ceramium rubrum is likely a warmer water Boreal species, but that cannot be 2011 W.H. Adey and L.-A.C. Hayek 89 Figure 45. Comparison of seaweed assemblages (species composition at canopy, understory, and epiphytic/small levels) with depth at exposed stations between the core Subarctic (NLQ) and the fringe regions of the Gulf of Maine (GOM) and southern Atlantic Nova Scotia (SNS). See text for discussion of the color coding: dark blue (Arctic), blue (Subarctic), purple (Boreal), white (cosmopolitan). 90 Northeastern Naturalist Vol. 18, Monograph No. 8 Figure 46. Comparison of seaweed assemblages (species composition at canopy, understory and epiphytic/small levels) with depth at intermediate exposure stations between the core Subarctic (NLQ) and the fringe regions of the Gulf of Maine (GOM) and southern Atlantic Nova Scotia (SNS). See text for discussion of the color coding: dark blue (Arctic), blue (Subarctic), purple (Boreal), white (cosmopolitan). 2011 W.H. Adey and L.-A.C. Hayek 91 Figure 47. Comparison of seaweed assemblages (species composition at canopy, understory and epiphytic/small levels) with depth at protected stations between the core Subarctic (NLQ) and the fringe regions of the Gulf of Maine (GOM) and southern Atlantic Nova Scotia (SNS). See text for discussion of the color coding: dark blue (Arctic), blue (Subarctic), purple (Boreal), white (cosmopolitan). 92 Northeastern Naturalist Vol. 18, Monograph No. 8 confirmed at this time. Neosiphonia harveyii is a warmer-water species that was introduced from Japan. In the Newfoundland/Labrador/Quebec (NLQ) stations, independent of exposure, there is a conspicuous absence of understory species biomass in the 2.5–5-m transition zone between sublittoral fringe Devaleraea/Acrosiphonia dominance and deeper water Ptilota serrata dominance (Figs. 25–27). This absence occurs where Laminaria digitata and its epiphyte Palmaria palmata intrude into the lower end of the Alaria kelp forest. As noted earlier, it might be hypothesized that in this narrow, maximum canopy zone (where Alaria esculenta also reaches maximum size; see Fig. 8b) the thickness of the canopy reduces light and production. However, this effect is not seen in the GOM and SNS regions, where canopy biomass is markedly greater. It can be clearly seen in Figures 45–47 that this zone is occupied by Chondrus crispus, Cystoclonium purpureum, and Phyllophora pseudoceranoides in GOM and SNS. Since mesograzing effects seem unlikely, we conclude that in NLQ this zone is the mean position of the summer Strongylocentrotus droebachiensis grazing front. The protected epiphyte Palmaria palmata is an occasional epiphyte on the stipes of Laminaria digitata as is the calcified Corallina officinalis, but apparently no bushy reds that are adapted to this zone have made the passage from the North Pacific Subarctic. The strong and characteristic sublittoral fringe zone of NLQ is reduced in GOM and virtually disappears in SNS (Table 6). Also, the progressive shrinking of the deep-water red zones including Ptilota serrata, Euthora (Callophyllis) cristata, and Scagelia pylaisii is clearly shown, but less pronounced, as might be expected. Moving from the NLQ across the TM hyperspace to warmer waters in GOM and SNS, the Subarctic flora is “squeezed out” first in shallow and mid-depths, but it persists in deeper water. As we demonstrated in Table 6, the Subarctic species assemblage that comprises greater than 95% of the biomass in NLQ is reduced to 30% of the total biomass in GOM and just 3% in SNS, and these areas are all significantly different at the P < 0.01 level (see above). However, Sheffé tests showed that GOM and SNS assemblages were significantly different at the P < 0.01 level, while GOM and NLQ were distinct at the P < 0.00l level. Similar tables for mid-depths (2.5–5 m; Table 7) and deeper water (10–30 m; Table 8) show equivalent reduction of Subarctic species abundance. However, it is very striking that the lowest percentages of Subarctic species abundance in all three regions, and conversely, the greatest penetration of Boreal species “northwards” into the core Subarctic, occurs in mid-depths. While there may be effects of biotic interaction, as we suggested regarding the location of the sea urchin front, we are also likely seeing winter control (low temperature and sea ice) of Boreal species penetration northward in shallow water, and summer control (failure of temperature to rise above 5 °C for a lengthy period) in deeper water. In considering the distribution of coralline algae across the North Atlantic, Adey and Steneck (2001) had the advantage of a uniform, quantitative trans- Atlantic data set. Unfortunately, equivalent data on the abundance of macroalgae that would allow a similar plot for the eastern Atlantic are few. However, the semiquantitative data provided by the Marine Biotope Classification (MBC) for 2011 W.H. Adey and L.-A.C. Hayek 93 Britain and Ireland by Connor et al. (1997), combined with the more general review of Lewis (1964) and Hawkins et al. (1992), is adequate to provide an extension of the survey that we have carried for the western Atlantic, and similar to that accomplished for the corallines (Fig. 48). These data only cover canopy species on exposed shores, but they provide a picture that is remarkably similar to that seen in the coralline plots in Figures 2, 3, and 17. Figure 48 only represents a fraction of the diversity present; in constructing Tables 7 and 8, which include only species of significant abundance in the MBC system, there were an additional 16 species (understory and epiphytic) that do not cross the North Atlantic but are Boreal in distribution and another 12 species that are either Lusitanean or Mediterranean in character (Tables 9, 10). Another very striking feature of Tables 6–8 is the trans-Atlantic macroalgal composition as a function of water depth. In the sublittoral fringe zone (Table 6), the top five listed Boreal species (that occur in the western Atlantic) form 73% of the biomass in GOM, 80% in SNS, and 69% in the English Channel. The relative abundance of these species is reduced in the Channel region because there are additional Boreal/Lusitanean non trans-Atlantic species in the flora (several Laurencia spp.in the understory and Himanthalia elongata in the canopy). In contrast, only 9% of the Channel region flora is made up of trans-Atlantic Boreal species at 2.5–5 m (Tables 7 and 8), the remainder being made up of either cosmopolitan species (8%) or “European endemics” (another 15 eastern Atlantic species being registered as abundant). At 10–30 m, the contrast is even more striking, with only 3% by abundance of the Boreal species tallied in the English Channel region also occurring in the western Atlantic and 13% being cosmopolitan species (that occur in the western Atlantic). At 10–30 m, 17 additional eastern Atlantic species are tallied as abundant “endemics”, with about two-thirds being Boreal in character and one third having a more southerly distribution (Lusitanean/Mediterranean). Note that in Tables 7 and 8, the Helgoland to Morlaix abundance calculations are based on 100% for each category of canopy/understory/epiphytic (as are the right-hand percentages for NLQ, GOM, and SNS sums). While this basis allows a more balanced comparison between the floras, these numbers are only semi-quantitative. In summary, of the dominant seaweeds in the shallow subtidal zones of the central eastern Atlantic Boreal Region, a large percentage (70% infra/0.5 m; 10% 2.5–5 m) have been able to make the trans-Atlantic passage to the western Atlantic, whether by natural or human assistance. However, most of the dominant, deeper-water eastern Atlantic species have not expanded westward across the North Atlantic. In terms of the apparent mechanisms for trans-Atlantic passage, it seems reasonable that shallow-water species would be more likely to achieve the crossing (e.g., on ballast, stones; see Brawley et al. 2009). Natural crossing (e.g., rafting; Ingolfsson 1995) would be similarly limited to shallowwater species. However, there is another equally important factor. In European Boreal waters, strong thermoclines with depth, at least in the photic range, do not occur. In northern Scotland, summer surface-water temperatures of 11–12 °C typically fall only a few tenths of a degree by 30 m depth. In the Channel and in 94 Northeastern Naturalist Vol. 18, Monograph No. 8 Table 7. Mid-water (2.5-5 m) regional comparison of western Atlantic species at exposed and intermediate stations. For each species listed, biomass (g/m2) and percentage (in parentheses) of total biomass within its habitat zone (canopy/understory/epiphytic) is given. Notes: Makkovik (Central Labrador) number of occurrences at 15 marine/estuarine stations, Hooper and Whittick (1984); the regional sums are normalized to 100% based on the total number of species/ station occurences of 216. For the remainder of the regions, sum rows provide regional biomass total for this study followed immediately in brackets by % of total biomass, and then in parentheses by the summed percent within canopy, understory, epiphytic categories, each weighted equally. Br. Is. (British Isles), Channel (English Channel), and Portugal: occurrence as Luning (1990) or algaebase.org (2009); * = abundant; 0 = occurs in small numbers or at limit; X = absent. Helgoland to Morlaix: semiquantitative stations; see text (%). Tr = trace. NLQ GOM SNS Helgoland to Makkovik Biomass % Biomass % Biomass % Br. Is. Channel Portugal Morlaix Arctic Turnerella pennyi 11 0.3 1.5 0 0 0 0 0 0 Sum 5 Tr 0 0 0 Subarctic Canopy Chordaria flagelliformis 15 19 (0.6) 0 0 * 0 X X Alaria esculenta 11 616 (20) 11 (0.3) 95 (1.8) * 0 X X Saccorhiza dermatodea 5 63 (2) 22 (0.6) 28 (0.5) X X X X Agarum clathratum 15 1283 (41) 43 (1.1) 20 (0.4) X X X X Desmarestia viridis 4 744 (24) 47 (1.2) 84 (1.6) * 0 X X Saccharina longicruris 3 1304 (33) 0 X X X X Understory Devaleraea ramentacea 5 8 (41) 0 0 X X X X Acrosiphonia arcta 9 2 (11) 0 0 * 0 X X Petalonia fascia 15 0.2 (1) 0 0 X X X X Ptilota serrata 8 5 (27) 1.9 (0.2) Tr X X X X Phycodrys riggi 15 3 (13) 0 0 X X X X Epiphytic Dictyosiphon foeniculaceus 15 41 (52) 0 0 X X X X Pilayella littoralis 15 6 (7) 0 0 ? ? ? X Euthora cristata 11 1.3 (1.6) 0 0.2 (0.3) X X X X Scagelia pylaisii 15 0.2 (0.3) 0 Tr (Tr) X X X X Sum 75 2792 {86} (81) 1429 {27} (12) 227 {4} (1.5) 0 2011 W.H. Adey and L.-A.C. Hayek 95 Table 7, continued. NLQ GOM SNS Helgoland to Makkovik Biomass % Biomass % Biomass % Br. Is. Channel Portugal Morlaix Boreal (occurs in Western Atlantic) Canopy Laminaria digitata 5 372 (12) 1153 (29) 2222 (43) * * X X Fucus serratus 0 0 0 680 (13) * * X X Codium fragile 0 0 0 865 (17) * * * X Understory Cystoclonium urpureum 3 0 170 (17) 21 (4) * * * X Chondrus crispus 0 0 608 (61) 121 (20) * * * (8) Corallina officianlis 2 0.6 (3) 4 (0.4) 255 (60) * * * (1) Phycodrys rubens ? 0 62 (6) 1.5 (0.13) * * X X Phyllophora pseudoceranoides 0 0 152 (15) 82 (14) * * X (1) Bonnemaisonia hamifera 0 0 4 (1.4) 0 * * * X Polyides caprinus (rotundus) 0 0 1.7 (0.2) 0 * * * (5) Ahnfeltia plicata 0 0 1.7 (0.2) * * * (1) Epiphytic Palmaria palmata 10 28 (35) 34 (12) 7 (11) * * * X Neosiphonia harveyii 0 Tr (Tr) 48 (17) 18 (28) * * ? X Ceramium rubrum ? 0 189 (66) 32 (50) ? ? ? X Dumontia incrassata 0 0 8 (3) 0 * * * X Porphyra spp.(purpurea) 0 0 0.3 (Tr) Tr (Tr) * * * X Callithamnian tetragonum 0 0 0 3 (5) * * * X Antithamnion sp. ? 0 0 1.4 (2) ? ? ? X Membranoptera alata 5 Tr (Tr) Tr (Tr) Tr (Tr) * * X (1) Sum 11 400 {12} (17) 2436 {47} (74) 4409 {76} (89) 10 Cosmopolitan Saccharina latissima 4 67 (2) 1337 (34) 1187 (23) * * * (12) Desmarestia ligulata 0 0 0 0 * * * (4) Rhodomela confervoides 15 0.6 (3) 1 (Tr) 14 (2) * * X X Ceramium virgatum 0 0 0 0 * * * (18) Ulva lactuca 0 0 0 0 * * * (17) Polysiphonia urceolata ? 0 0 0 ? ? ? (24) Sum 9 68 {2} (2) 1338 {26} (11) 1201 {21} (8) 29 Total biomass 3260 5209 5837 96 Northeastern Naturalist Vol. 18, Monograph No. 8 Table 8. Deep-water (10–30 m) regional comparison of species occurring in the Western Atlantic. For each species listed, biomass (g/m2) and percentage (in parentheses) of total biomass within its habitat zone (canopy/understory/epiphytic) is given. Sum rows provide regional biomass total for this study followed immediately in brackets by % of total biomass, and then in parentheses by the summed percent within canopy, understory, epiphytic categories, each weighted equally. NLQ GOM SNS Helgoland to Biomass % Biomass % Biomass % Morlaix Arctic Turnerella pennyi 11.4 (4.2) 0 (0) Tr (Tr) X Neodilsea integra 0.04 (Tr) 0 (0) 0 (0) X Subarctic Canopy Alaria esculenta Tr (Tr) 3.7 (0.5) 7.3 (0.5) X Saccorhiza longicruris 0 (0) 10.4 (1.3) 0 (0) X Saccorhiza dermatodea 0 (0) 7.5 (0.9) 2.9 (0.2) X Agarum clathratum 947 (94) 356 44 475 (30) X Desmarestia viridis 60.9 (6) 11.1 1.4 9.5 (0.6) X Understory Phycodrys riggii 6.7 (2.5) 0 (-) 0 - X Ptilota serrata 248.4 (93) 28.8 (20) 31.9 (16) X Fimbriofolium dichotomum 0 (0.1) 7.1 (4.9) 0.3 (1) X Epiphytic Euthora cristata 0.41 (32.5) 9.4 (19.8) 6.2 (15.8) X Scagelia pylaisii 0.74 (59) 0 (0) 0.5 (1.2) X Sum 1276 {100} (97) 434 {49} (31) 534 {30} (22) X Boreal Canopy Laminaria digitata 0 (0) 168 20.6 216 (13.8) X Fucus serratus 0 0 58 (4) X Understory Chondrus crispus 0 7.3 (5) 3.0 (1.5) X Corallina officinalis Tr (Tr) 3.7 (2.5) 8.2 (4.1) X Cystoclonium purpureum 0 11 (7.6) 17.8 (8.8) X Phyllophora pseudoceranoides 0 30.7 (21.9) 118.6 (58.7) X Phycodrys rubens 0 56.2 (38.6) 6 (3.0) (10) Polyides caprinus 0 0 (0) 15.7 (7.8) X 2011 W.H. Adey and L.-A.C. Hayek 97 Table 8, continued. NLQ GOM SNS Helgoland to Biomass % Biomass % Biomass % Morlaix Boreal, cont. Epiphytic Ceramium rubrum 0 19.6 (18.9) 4.9 (11.5) X Neosiphonia harveyii 0 69 (65.5) 0.4 (0.9) X Palmaria palmata 0 0 (0) 4 (9.4) X Porphyra spp. (purpurea) 0 1.7 1.6 0 (0) 0 Sum Tr (Tr) 199 {22} (58) 453 {25} (41) (3) Cosmopolitan Canopy Saccharina latissima 0 247 (30) 743 (48) X Desmarestia aculeata 0 11 (1) 22 (1.5) X Chorda filum 0 0 27 (1.8) (8) Understory Rhodomela confervoides 0 0.7 (0.5) 0 X Dictyota dichotoma 0 0 0 (10) Epiphytic and small Lomentaria orcadensis 0 0 0 (5) Membranoptera alata (denticulata) 0.1 (9) 0.1 (Tr) Tr Tr X Chaetomorpha spp. 0 0 9 (21) X Ectocarpus sp. 0 0 17 (39) X Polysiphonia sp. 0 0 0 (5) Bryopsis plumosa 0 0 0 (5) Ulva lactuca 0 0 0 (5) Sum 0.1 (3) 258 {29} (10) 818 {45} (37) (12) Biomass Total 1277 891 1805 98 Northeastern Naturalist Vol. 18, Monograph No. 8 northwestern France, the top to bottom summer temperatures in the water column would typically be 15–17 °C. These are summer surface temperatures that are also characteristic of GOM and SNS (except near the mouth of the Bay of 2011 W.H. Adey and L.-A.C. Hayek 99 Table 9. Mid-water (2.5-5 m) regional comparison of Eastern Atlantic species not known in Western Atlantic at exposed and intermediate stations. For Helgoland to Morlaix sums are normalized to 100%. * = abundant; x = absent. British Isles English Channel Helgoland to Morlaix Portugal Boreal species Laminaria hyperborea * * (44) * Dilsea carnosa * * (5) * Cladostephus spongiosus * * (5) * Callophyllis laciniata * * (16) * Erythrodermis traillii * * (5) X Lomentaria orcadensis * * (12) * Lomentaria clavellosa * * (6) * Sum (35) Lusitanean/ Mediterranean Laminaria ochroleuca * * (20) * Saccorhiza polyschides * * (20) * Cryptopleura ramosa * * (16) * Erythroglossum laciniatum * * (8) * Brongniartella byssoides * * (5) * Halarachnion ligulatum * * (1) * Figure 48 (opposite page). Comparison of seaweed canopy assemblages for exposed canopy sites between the western North Atlantic (Subarctic core and fringe Gulf of Maine and Nova Scotia) with eastern North Atlantic sites in Scotland and the Irish Sea and in the English Channel. The data for the British Isles were extracted from semi-quantitative data (abundance on scale of 0–5) for 9 sites from North Cornwall to Shetland in the JNCC Report (Connor et al. 1997). The English Channel data were acquired from two JNCC Report sites on the Channel plus Helgoland and Baie de Morlaix (Luning 1990) and South Devon (Lewis 1964). See text for discussion of the color coding: blue (Subarctic), purple (Boreal), white (cosmopolitan), orange (Lusitanean). Fundy). However, in the western Atlantic, except in the vicinity of the mouth of the Bay of Fundy, sharp summer thermoclines are the rule, and in GOM and SNS, with surface temperatures at 12–14 °C, the 30-m-deep temperatures are typically 7–9 °C and 5–7 °C, respectively. In NLQ, the summer thermocline can be even sharper, with summer temperatures at 30 m often 2 °C or lower. While 12% of the mid-depth floral biomass in NLQ is of Boreal origin, the percentage is only 1% in shallow water, and just a trace in deep water. Unlike the biogeographically transitional Gulf of Maine and Atlantic Nova Scotian shores, the core Subarctic presents a climatic wall that is difficult for eastern Atlantic Boreal species to climb, even after crossing the North Atlantic. The great difference in thermal structure between the eastern and western Atlantic, even when summer surface-water temperatures are quite similar, along with the obvious basic transportation issues, acts to prevent the successful trans-Atlantic crossing of deeper-water species. In biomass terms, only three Boreal species have made small inroads into the Subarctic, primarily at mid-depths: two red seaweeds, Palmaria palmata and 100 Northeastern Naturalist Vol. 18, Monograph No. 8 Table 10. Deep-water (10–30 m) regional comparison of species not known in Western Atlantic among exposed and intermediate stations. (%) proportions within canopy, understory, epiphytic, normalized to 100%. English Helgoland to British Isles Channel Portugual Morlaix Boreal species Laminaria hyperborea X X X 100 Sphaerococcus coronopiliferus X X X (4) Callophyllus laciniata X X X (4) Erythroglossum laciniatum X X X (4) Delesseria sanguinea X X X (4) Kallymenia reniformis X X X (4) Drachiella spectabilis X X X (35) Phycodrys rubens* X X X (10) Haraldiophyllum bonnemaisonium X X X (10) Pterosiphonia parasitica X X X (5) Sum (57) Lusitanian/Mediteranean Cryptopleura ramosa X X X (25) Bonnemaisonia asparagoides X X X (18) Heterosiphonia plumosa X X X (11) Halopteris filicina (7) Dictyopteris polypodioides X X X (4) Brongniartella byssoides X X X (4) Meredithia microphylla (14) Spondylothamniun multifidum X X X (10) Sum (28) *Differentiation between P. riggii and P. rubens unclear; P. rubens may occur in the Western Atlantic especially in GOM and SNS. Corallina officinalis, and the kelp Laminaria digitata. As we have discussed, it is at mid-depths in the Subarctic that there are a minimum of understory species, probably related to Strongylocentrotus grazing fronts. This is a likely zone for protected Boreal understory species to gain a foothold, even under climatically marginal conditions. Laminaria digitata has evolved from the Arctic Laminaria solidungula, but appears to be absent from the Arctic (Lee 1980, Wilce 1994). In Europe and on the transitional western Atlantic shores, it occurs broadly in shallow water on exposed coasts; in the Subarctic, it finds a narrow refugial band between the Alaria esculenta zone and the S. droebachiensis grazing front. In the North Atlantic Subarctic, the branching Lithothamnion species (L. glaciale, L. lemoineae) and the cortex-bearing (epithallium) Clathromorphum species (C. circumscriptum and C. compactum) highly dominate the rocky-bottom coralline flora. They are also “protected” species (the branching species by physically making grazing difficult; the cortex species by having a sunken, intercalary meristem and giving up cortex to grazing). Large areas of mostly living and growing coralline surface densely covered with sea urchins and with individual plants able to survive minimally for many decades and probably centuries (Halfar et al. 2011), attests to the efficacy of this arrangement. Also, as we have discussed, there is growing evidence that urchins that remain on such 2011 W.H. Adey and L.-A.C. Hayek 101 coralline surfaces can maintain themselves for many years by growing extremely slowly during that time. There is also some evidence that S. droebachiensis keys to coralline surfaces when recruiting. There is an apparently unprotected, thin, and smooth-crusted dominant coralline (Leptophytum leave) in the Subarctic, but it is a deep-water species, and tends to occupy pebble, shell, and maerl surfaces in unstable sediment environments (Adey 1970b); as we have noted, those bottoms are difficult for sea urchins to negotiate. In short, this species occupies a refugium from sea urchin grazing. Such protected coralline species appear to play little role in the Atlantic Boreal. In Boreal waters, on these rocky bottoms, these species are mostly replaced by crustose species largely lacking such highly developed grazing protection (Phymatolithon purpureum in shallow waters; Phymatolithon lamii and P. laevigatum in mid-depths, and the unbranched Lithothamnion sonderi in deep water) (Adey and Adey 1973). In brief, it would appear that the establishment of a long-term stable sea urchin/coralline savanna, with its “forest patches” of highly protected (or refugium-located) species was not simply a matter of species selection from an available pool, but rather the co-evolution of grazer and macroalgal assemblages. Invertebrate populations As with the coralline algae and seaweeds, the same dominant invertebrates that characterize shallow rocky bottoms can be found in NLQ as in GOM and SNS. However, the proportions are likewise markedly different. Himmelman (1991) described this difference for a single locality in the northern Gulf of St. Lawrence. Separately, based on a large number of random bottom photographs from all three regions, we are quantitatively documenting this difference for echinoderms, crabs, and large molluscs, easily identified species in photographs. The principal elements appear clearly in the emerging data. The sunstars (Solaster endeca (L.) and Crossaster papposus) and the asteriid Leptasterias polaris are seen on virtually every dive in NLQ, often in abundance, and are dominant elements in the photographs. In our station work in GOM and SNS, of these three sea stars, only a single C. papposus was seen (offshore of Cape Elizabeth, ME); on the other hand, Asterias vulgaris Verrill (= Asterias rubens L.) (Northern Sea Star), which occurs patchily in NLQ, is abundant. In NLQ, the crab Hyas araneus (L.) (Toad Crab) is abundant, and yet is rarely seen in GOM and SNS, where the cancriid crabs, especially Cancer irroratus Say (Atlantic Rock Crab) and Cancer borealis Stimpson (Jonah Crab), along with the invasive Carcinus maenas (L.) (Green Crab), are the dominant elements. Our emerging data suggest that the analysis of Himmelman (1991) for the invertebrate community structure of the northern Gulf of St. Lawrence is generally applicable to the entire NLQ and warrants additional detailed study. Conclusions The 3000-km radiating set of coasts centered on the Strait of Belle Isle is highly unique in the structure of its seaweed assemblages. Based on the abiotic thermogeographic model (TM), this region is called the North Atlantic Subarctic 102 Northeastern Naturalist Vol. 18, Monograph No. 8 (Adey and Steneck 2001; also NLQ in the present study). Using ANOVA and Sheffé contrasts, chi square tests, and the Bray Curtis coefficients in a similarity analysis with multidimensional scaling, we showed the difference between NLQ and the Boreal/Subarctic transitional coasts of GOM and SNS to be highly significant statistically. Although most of the dominant species of NLQ extend to, and even through, the southern Canadian Arctic Archipelago into the North Pacific Subarctic region, the principal species are not Arctic dominants; Arctic species play only a limited role, mostly in deep, much colder water in the Subarctic. Although all of the dominant species of the NLQ/Subarctic occur to the southwest in the GOM and SNS, they become significantly reduced in abundance; those coasts are dominated by seaweed species that are also important in shallow waters on European coasts. Only a few of the dominant seaweeds of the NLQ core reach the British Isles, and none extend as far as the English Channel as significant members of rocky subtidal communities. As we have shown, an extensive additional array of seaweeds, unknown in the western Atlantic, also occurs in Europe; the TM of Adey and Steneck (2001) has designated this flora centered on the English Channel as Boreal (Celtic), in keeping with a long tradition in biogeography. Many shallow-water Boreal species apparently crossed the Atlantic to the rocky shore of North America during the late Holocene, and perhaps mostly in historical times. Some of these seaweeds have become important members of rocky shore communities from Long Island Sound to Nova Scotia. While three of these European seaweeds have been tabulated in this study for the NLQ, they occur at low percentages in mid-depths; Boreal seaweeds are rarely present in shallow and deeper waters of the NLQ. This pattern is in accordance with the yearly vertical temperature structure of the water column. It may also relate to available space in the Subarctic seaweed assemblage, since two of the three Boreal species have some protection (from grazing); as we have noted, the third (Laminaria digitata) occurs only in narrow, high wave energy refugia. Even here, at the base of the Alaria zone, its occurrence is patchy as it is wedged between fast-growing Alaria and sea urchins moving up from below to feed. The distribution of subtidal rocky bottom seaweed assemblages across the North Atlantic is closely in agreement with the abiotic patterns of the TM (which is based on temperature, area, and time, including the cyclical glaciations of the Pleistocene). Equally striking, the proportional differences in the similarity graphic between mean values of NLQ, GOM, and SNS (Fig. 20) are very close to those of the TM North Atlantic temperature graphic (Fig. 2). On the other hand, the percentage of Boreal seaweed species in GOM and SNS, from collections taken mostly in the last 10 years, is significantly higher than those of coralline species from collections taken in the 1960s. This difference may result from sea-water temperature rise over the 30–40-year time span between collections; this hypothesis can be tested by re-collection and new analysis of coralline species cover. As we have discussed in this paper, it seems likely that there has been no extensive change in Atlantic Subarctic seaweed assemblages for at least a century, and based on emerging data on coralline longevity, probably much longer. Deglaciation processes during the early to mid-Holocene and the general lack 2011 W.H. Adey and L.-A.C. Hayek 103 of rocky refugia to the south suggest that the Subarctic nature of the northwestern Atlantic rocky shores was fully established by that time. However, subtidal rocky bottom communities of the biogeographically transitional zones of SNS and the GOM, where the invasion of many species from the European Boreal has occurred throughout historical times, was likely underway naturally in pre-Columbian times, and is continuing today. These invasions have made major changes in the flora and fauna, and probably in community structure and ecosystem function as well. Presciently, Vadas and Elner (1992) noted that the “instability of the north-west Atlantic ... was recognized ... over 60 years ago.” In that case, they were referring to Mt. Desert Island, in the middle of the Maine Coast. However, these changes have yet to make significant inroads into the “core” Subarctic Region, suggesting that the biogeographic regions designated by the TM are more biotically stable than the transitional areas or zones lying between the regions. Although considerable additional study is necessary, it seems likely that the North Atlantic rocky, subtidal Subarctic shore is a model of ecosystem stability. This stability relates to a balance between a very slowgrowing consumer/producer combination (sea urchin/corallines) and a suite of chemically protected seaweeds. The protected seaweeds, including corallines, provide structure, and the wave-energy controlled grazing of an unprotected kelp (Alaria) allows periodic urchin reproduction. Biostromal in nature, the so-called coralline “barren” provides 1–5 cm of bored and porous carbonate that is richly occupied by filter-feeding invertebrates. The coralline barren also provides a reservoir of small, growth-retarded sea urchins waiting to move up as adults after an autumn storm has caught the previous generation before they could migrate. This dynamic adds another dimension to the ecosystem. in their 1990 paper “Disturbance and organization of macroalgal assemblages in the Northwest Atlantic”, Chapman and Johnson stated that “since there have been no indications of large-scale oscillations in community state, system dynamics in Newfoundland and the St. Lawrence estuary appear fundamentally different from that in Nova Scotia and New England in that sea urchin-coralline barrens not only dominate the subtidal but are apparently a permanent feature of it.” Allowing for the use of the word “barrens”, with the assumption that anything but a kelp forest in cold, rocky waters was somehow degraded, this showed a recognition 20 years ago of the essentials presented in the present paper. Himmelman (1991) in his study of the rocky shore ecosystem in the Mingan Islands provided a similar view. Neither of these authors voiced awareness of the biostromal and reef-like nature of the coralline “barren”, but they certainly recognized the great difference in benthic community structure above the coralline surface. The Chapman and Johnson (1990) paper was based on a considerable number of publications by scientists working in Quebec and Newfoundland and particularly on a number of experimental and observational papers by Himmelman and by Keats (see Literature Cited). Not surprisingly, without a theoretical basis and model for understanding the marine biogeography of the North Atlantic, the larger number of ecologists working in what is a transition region (i.e., GOM, SNS) compared to the large Subarctic Region (e.g., Himmelman, Keats, and their colleagues) 104 Northeastern Naturalist Vol. 18, Monograph No. 8 have not always recognized the importance of the Subarctic literature to understanding the North Atlantic. Indeed, a single paper from this group (Keats et al. 1985) appeared in both Mathieson et al. (1991) and the more recent CORONA (Coordinating Research on the North Atlantic) volume (Cunningham 2008), both referring to the effects of sea ice on a shore community (also see Heaven and Scrosati 2008). With the successful tests of the TM, including the present report, we hope that earlier literature demonstrating how different rocky shores in NLQ are from GOM and SNS will find a larger audience and that the Subarctic Region will be given more attention. In conclusion, the Atlantic Subarctic is demonstrated by the TM to be a large and distinctive biogeographic region compared to the transition region found to its southwest in southwestern Nova Scotia and the Gulf of Maine. The Subarctic Region has largely resisted Boreal invasion to date, whereas the transition region has many European Boreal species in shallow water. However, global warming is likely to displace the Arctic biota, leaving the much larger region of the Canadian Arctic Archipelago to become the Canadian Subarctic Archipelago. Additional studies of both the Subarctic and the Arctic should be a priority for marine biologists. Acknowledgments This was a multiyear project that covered thousands of miles of coast and was carried out with a research vessel that at times used the services of many boatyards and ports in the Canadian Maritimes and the Gulf of Maine. It would be impractical to acknowledge individually everyone who helped us on our way, and here we credit only the key individuals involved directly in collection and research. First and foremost, Karen Loveland Adey and Erik Adey made our superb research vessel possible, were present at sea on virtually all legs, and finally supported both the extended analyses and the production of the manuscript. Although we had an array of crew members at different times, Alok Mallick, Alex Miller, and Nick Caloyianis were often aboard on passages and were also superb divers and provided a wide range of collecting capabilities and underwater photography. Nick especially produced incomparable video images to provide a record of stations in the Gulf of Maine. Sue Lutz, of the Smithsonian’s Department of Botany produced several of the key illustrations and, as always, was available for the seemingly endless array of changes and additions. Illustrator Alice Tangerini and digital designer Mary Alice Roth worked hard to make sense out of our hopes for the front cover. We are particularly grateful for the considerable thought and valuable recommendations provided by an anonymous reviewer and Dr. Ricardo Scrosati, as well as our editor Dr. Susan Brawley. Our Copy Editor, Keith Goldfarb, has been extremely supportive and inventive in all aspects of the final editorial process. Funding for this project was provided by Ecological Systems Technology, Inc. and the Botany Department of the National Museum of Natural History. Literature Cited Addison, J., and M. Hart. 2005. Colonization, dispersal, and hybridization influence phylogeography of North Atlantic sea urchins (Strongylocentrotus droebachiensis). Evolution 59:532–543. 2011 W.H. Adey and L.-A.C. Hayek 105 Adey, W. 1965. The genus Clathromorphum in the Gulf of Maine. Hydrobiologia 24:377–420. Adey, W. 1966. The distribution of saxicolous crustose corallines in the northwestern North Atlantic. Journal of Phycology 2:49–54. Adey, W. 1970a. The crustose corallines of the northwestern North Atlantic including Lithothamnium lemoineae n. sp. Journal of Phycology 6:225–229. Adey, W. 1970b. Some relationship between crustose corallines and their substrate. Scientia Islandica 2:22–25. Adey,W. 1971. The sublittoral distribution of crustose corallines on the Norwegian Coast. Sarsia 46:41–58. Adey, W., and P. Adey. 1973. Studies on the biosystematics and ecology of the epilithic crustose Corallinaceae of the British Isles. British Phycological Journal 8:343–407. Adey, W., and L-A. Hayek. 2005. The biogeographic structure of the western North Atlantic rocky intertidal. Crypogamie Algologie 26(1):35–66. Adey, W., and K. Loveland. 1998. Dynamic Aquaria: Building Living Ecosystems. Academic Press, San Diego, CA. 498 pp. Adey, W., and R. Steneck. 2001. Thermogeography over time creates biogeographic regions: A temperature/space/time-integrated model and an abundance-weighted test for benthic marine algae. Journal of Phycology 37:677–698. Adey, W., S. Lindstrom, M. Hommersand, and K. Muller. 2008. The biogeographic origin of Arctic endemic seaweeds: A thermogeographic view. Journal of Phycology 44:1384–1394. Anderson, B., and H. Borns. 1994. The Ice Age World. 208 pp. Scandinavian University Press, Oslo, Norway. Balch, T., and R. Schiebling. 2002. Temporal and spatial variability in settlement and recruitment of echinoderms in kelp beds and barrens in Nova Scotia. Marine Ecology Progress Series 205:139–154. Bartsch, I., C. Wiencke, K. Bischof, C. Buchholz, B. Buck, A. Eggert, P. Fuerpfeil, D. Hanelt, S. Jacobsen, R. Karez, U. Karsten, M. Molis, M. Roleda, H. Schubert, R. Schumann, K. Valentin, F. Weinberger, and J. Wiese. 2008. The genus Laminaria sensu lato: Recent insights and developments. European Journal of Phycology 43(1):1–86. Benoit, H., and D. Swain. 2008. Impacts of environmental change and direct and indirect harvesting effects on the dynamics of a marine fish community. Canadian Journal of Fisheries and Aquatic Sciences 65:2088–2104. Bird, C.J., M.J. Dadswell, and D.W. Grund. 1993. First record of the potential nuisance alga Codium fragile spp. tomentosoides (Chlorphyta, Caulerpales) in Atlantic Canada. Proceedings of the Nova Scotian Institute of Science 40:1–11. Brawley, S., J. Coyer, A. Blakeslee, G. Hoarau, L. Johnson, J. Byers, W. Stam, and J. Olsen. 2009. Historical innovations of the intertidal zone of Atlantic North America associated with distinctive patterns of trade and emigration. PNAS 106:8239–8244. Briggs, J. 1974. Marine Zoogeography. McGraw Hill, New York, NY. 475 pp. Briggs, J. 1995. Global Biogeography. Elsevier, Amsterdam, The Netherlands. 452 pp. Briggs, J. 2003. Marine centers of origin as evolutionary engines. Journal of Biogeography 30:1–18. Bustnes, J., and O. Lonne. 1995. Sea ducks as predators on sea urchins in a northern kelp forest. Pp. 599–608, In H. Skjoldal, C. Hopkins, K. Erikstad, and H. Leinaas (Eds.). Ecology of Fjords and Coastal Waters. Elsevier Science, Amsterdam, The Netherlands. 106 Northeastern Naturalist Vol. 18, Monograph No. 8 Campbell, D. 2004. Ecosystem modeling in Cobscook Bay, Maine: A Boreal macrotidal estuary. Northeastern Naturalist 11(Special Issue 2):355–424. Chapman, A.R.O., and C. Johnson. 1990. Disturbance and organization of macroalgal assemblages in the Northwest Atlantic. Hydrobiologia 192:77–121. Choi, H., G. Kraft, H. Kim, M. Guiry, and G. Saunders. 2008. Phylogenetic relationships among lineages of the Ceramiaceae (Ceramiales, Rhodophyta) based on nuclear small subunit rDNA sequence data. Journal of Phycology 44:1033–1048. CLIMAP Project Members. 1976. The surface of the ice-age earth. Science 191:113–117. Connor, D., M. Dalkin, T. Hill, R. Holt, and W. Sanderson. 1997. Marine biotope classification for Britain and Ireland. Volume 2. Sublittoral biotopes. Joint Nature Conservation Committee, Peterborough, UK. Report No. 230. Coyer, J., G. Hoarau, M.-P. Outot-Le Secg, W. Stam, and J. Olsen. 2006. A mtDNA-based phylogeny of the brown algal genus Fucus (Herterokontae: Phaeophyta). Molecular Phylogenetics and Evolution 39:209–222. Cunningham, C. 2008. Lessons learned from coordinating research on the North Atlantic (Corona). Ecology 89 (11 Supplement):S1–S2. Dawson, A. 1992. Ice Age Earth: Late Quaternary Geology and Climate. Routledge, London, UK. 293 pp. Dayton, P. 1975. Experimental studies of algal canopy interactions in a Sea Otter-dominated kelp community at Amchitka Island, Alaska. Fishery Bulletin 73:230–237. Dudgeon, S., I. Davison, and R. Vadas. 1990. Freezing tolerance in the intertidal red algae Chrondus crispus and Mastocarpus stellatus: Relative importance of acclimation and adaptation. Marine Biology 106:427–436. Dudgeon, S.R., R.S. Steneck, I.R. Davison, and R.L. Vadas. 1999. Coexistence of similar species in a space-limited intertidal zone. Ecological Monographs 69:331–352. Dunton, K., and D. Schell. 1986. Seasonal carbon budget and growth of Laminaria solidungula in the Alaskan high arctic. Marine Ecology Progress Series 31:57–66. Dunton, K., E. Reimnitz, and S. Schonberg. 1982. An Arctic kelp community in the Alaskan Beaufort Sea. Arctic 35:465–484. Ekman, S. 1953. Zoogeography of the Sea. Sidgewick and Jackson, London, UK. 417 pp. Estes, J., and D. Duggins. 1995. Sea Otters and kelp in a community ecological paradigm. Ecological Monographs 65:75–100. Estes, J., and P. Steinberg. 1988. Predation, herbivory, and kelp evolution. Paleobiology 14:19–36. Gagnon, P., J. Himmelman, and L. Johnson. 2003. Algal colonization in urchin barrens: Defense by association during recruitment of the brown alga Agarum cribrosum. Journal of Experimental Marine Biology and Ecology 290:179–196. Gagnon, P., J. Himmelman, and L. Johnson. 2004. Temporal variation in community interfaces: Kelp-bed boundary dynamics adjacent to persistent urchin barrens. Marine Biology 144:1191–1203. Gagnon, P., L. Johnson, and J. Himmelman. 2005. Kelp patch dynamics in the face of intense herbivory: Stability of Agarum clathratum stands and associated flora on urchin barrens. Journal of Phycology 41:498–505. Garbury, D., and G.R. South (Eds.). 1989. Evolutionary Biogeography of the Marine Algae of the North Atlantic Ocean. Springer-Verlag, Berlin, Germany. 427 pp. Gaudette, J., R. Wahle, and J. Himmelman. 2006. Spawning events in small and large populations of the Green Sea Urchin, Strongylocentrotus droebachiensis, as recorded using fertilization assays. Limnology and Oceanography 51:1485–1496. Graham, M., B. Kinlan, L. Druehl, L. Garske, and S. Banks. 2007. Deep-water kelp refugia as potential hot spots of tropical marine diversity and productivity. PNAS 104:16576–16580. 2011 W.H. Adey and L.-A.C. Hayek 107 Greene, C., A. Pershing, T. Cronin, and N. Ceci. 2008. Arctic climate change and its impacts on the ecology of the North Atlantic. Ecology 89(supplement 2008):524–538. Halfar, J., S. Hetzinger, T. Zack, W. Adey, G. Gamboa, B. Kunz, B.Williams, and D. Jacob. 2011. Coralline algae growth increment widths archive North Atlantic climate variability. Palaeogeography, Palaeoclimatology, and Palaeoecology. Available online at doi: 10.1016/j.palaeo.2010.04.009. Harris, L. 1990. Independent review of the state of the northern Cod stock. Final Report. Communications Directorate, Department of Fisheries and Oceans, Ottawa, ON, Canada. 154 pp. Hawkins, S., R. Hartnoll, J. Kain and T. Norton. 1992. Plant-animal interactions on hard substrate in the Northeast Atlantic. Pp. 1–32, In D.M. John, S.J. Hawkins, and J.H. Price (Eds.). Plant-Animal Interactions in the Marine Benthos. Clarenden Press, Oxford, UK. 570 pp. Heaven, C.S., and R.A. Scrosati. 2008. Benthic community composition across gradients of intertidal elevation, wave exposure, and ice scour in Atlantic Canada. Marine Ecology Progress Series 369:13–23. Hetzinger, S., J. Halfar, A. Kronz, R. Steneck, W. Adey, P. Lebednik, and B. Schone. 2009. High-resolution Mg/Ca ratios in a coralline red alga as a proxy for Bering Sea temperature variations from 1902–1967. Palaios 24:406–412. Himmelman, J. 1985. Urchin feeding and macroalgal distribution in Newfoundland, eastern Canada. Canadian Journal of Zoology 56:1828–1836. Himmelman, J. 1986. Population biology of Green Sea Urchins on rocky barrens. Marine Ecology Progress Series 33:295–306. Himmelman, J., 1991. Diving observations of subtidal communities in northern Gulf of St. Lawrence. In Therriault, J.-C. (Ed.). The Gulf of St. Lawrence: Small Ocean or Big Estuary? Canadian Special Publication of Fisheries and Aquatic Sciences 113:319–332. Himmelman, J., and Y. Lavergne. 1985. Organization of rocky subtidal communities in the St. Lawrence Estuary. Canadian Journal of Zoology 62:1015–1012. Himmelman, J., and H. Nedelec. (1990). Urchin foraging and algal survival strategies in intensely grazed communities in eastern Canada. Canadian Journal of Fisheries and Aquatic Sciences 47:1011–1026. Himmelman, J., A. Cardinal, and E. Bourget. 1983. Community development following removal of urchins, Strongylocentrotus drobachiensis, from the rocky subtidal zone of the St. Lawrence Estuary, eastern Canada. Oecologia (Berlin) 59:27–39. Himmelman, J., C. Dumont, C. Gaymer, C. Vallieres, and D. Drolet. 2008. Spawning synchrony and aggregative behavior of cold-water echinoderms during multi-species mass spawning. Marine Ecology Progress Series 361:161–168. Hommersand, M., S. Fredericq, and D. Freshwater. 1994. Phylogenetic systematic and biogeography of the Gigartinaceae (Gigartinales, Rhodophyta) based on sequence analysis of rbc L. Botanica Marina 37:193–203. Hommersand, M., O. De Clerck, and E. Coppejans. 2004. A morphological study and taxonomic revision of Euptilota (Ceramiaceae, Rhodophyta) from South Africa. European Journal of Phycology 39:369–394. Hooper, R. 1980. Observations on algae-grazer interactions in Newfoundland and Labrador. Canadian Technical Report of Fisheries and Aquatic Sciences 954:92–119. Hooper, R., and A. Whittick. 1984. The benthic algae of the Kaipokok Bay, Makkovik Bay, and Big River Bay Region of the central Labrador Coast. Naturaliste Canadien 111:131–138. Ingolffson, A. 1995. Floating clumps of seaweed around Iceland: Natural microcosms and a means of dispersal for shore fauna. Marine Biology 122:13–21. 108 Northeastern Naturalist Vol. 18, Monograph No. 8 Jenkins, S., P. Moore, M. Burrows, D. Garbary, S. Hawkins, A. Ingolfsson, K. Sebens, P.V.R. Snekrove, D. Wethy, and S. Woodin. 2008. Comparative ecology of North Atlantic shores: Do differences in players matter for process? Ecology 89(supplement 2008):S3S23. Johnson, L., S. Brawley, and W. Adey. In press. Secondary spread of invasive species: Historic patterns and underlying mechanisms of the continuing invasion of the European rockweed Fucus serratus in eastern North America. Biological Invasions. Keats, D. 1986. The effects of the experimental removal of Green Sea Urchins and of ice scour on sublittoral benthic macro-algal communities in eastern Newfoundland. Ph.D. Thesis. Memorial University of Newfoundland, St Johns, NL. 233 pp. Keats, D., and G.R. South. 1985. Aspects of the Reproductive Phenology of Saccorhiza dermatodea (Phaeophyta, Laminariales) in Newfoundland. British Phycological Journal 20:117–22. Keats, D., G. South, and D. Steele. 1982. The occurrence of Agarum cribrosum (Mert.) Bory (Phaeophyta) in relation to some of its competitors and predators in Newfoundland. Phycologia 21:189–191. Keats, D., G.R. South, and D. Steele. 1985. Algal biomass and diversity in the upper subtidal at a pack-ice disturbed site in eastern Newfoundland. Marine Ecology Progress Series 25:151–158. Keats, D., D. Steele, and G.R. South. 1986. Atlantic Wolffish (Anarchicas lupus) Predation on Green Sea Urchins (Strongylocentrotus drobachiensis) in Eastern Newfoundland. Canadian Journal of Zoology 64:1920–1925. Kraan, S., J. Rueness, and M. Guiry. 2000. Are North Atlantic Alaria esculenta and A. grandifolia (Alariaceae, Phaeophyceae) conspecific? European Journal of Phycology 36:35–42. Lane, C., C. Mayes, L. Druehl, and G. Saunders. 2006. A multi-gene molecular investigation of the kelp (Laminariales, Phaeophyceae) supports substantial taxonomic re-organization. Journal of Phycology 42:493–512. Langton, R., and R. Bowman. 1978. Food habits and food resource partitioning by Northwest Atlantic Gadiform fishes. National Marine fisheries Service, NFCL, Woods Hole, MA. Ref. No. 78-35:157 pp. Lee, R.K.S. 1980. Catalogue of the Marine Algae of the Canadian Arctic. Natural Museum of Canada, Publications in Botany, No. 9. Ottawa, ON, Canada. 82 pp. Lewis, J. 1964. The Ecology of Rocky Shores. English University Press, London, UK. 323 pp. Lindstrom, S. 2001. The Bering Strait connection: Dispersal and speciation in boreal macroalgae. Journal of Biogeography 28:243–251. Longhurst, A. 1998. Ecological Geography of the Sea. Academic Press, New York, NY. 398 pp. Lüning, K. 1990. Seaweeds, their Environment, Biogeography, and Ecophysiology. John Wiley and Co., New York, NY. 527 pp. Maggs, C., R. Castillo, D. Foltz. C. Henzler, M. Jolly, J. Kelly, J. Olsen, K. Perez, W. Stam, R. Vainola, F. Viard, and J. Wares. 2008. Evaluating signatures of glacial refugia for North Atlantic benthic marine taxa. Ecology 90:S108–S122. Mann, K. 1973. Seaweeds: Their productivity and strategy for growth. Science 182:975–981. Mathieson, A.C., C.A. Penniman, and L.G. Harris. 1991. Northwest Atlantic rocky shore ecology. Pp. 109–191, In A.C. Mathieson and P.H. Nienhuis (Eds.). Intertidal and Littoral Ecosystems. Elsevier, Amsterdam, The Netherlands. Mathieson, A., C. Dawes, E. Hehre, and L. Harris. 2010. Floristic Studies of Seaweeds from Cobscook Bay, Maine. Northeastern Naturalist 17(Monograph 5):1–48. 2011 W.H. Adey and L.-A.C. Hayek 109 McCook, L., and A.R.O. Chapman. 1997. Patterns and variations in natural succession following massive ice scour of a rocky intertidal seashore. Journal of Experimental Marine Biology and Ecology 214:121–47. McNaught, D. 1999. The indirect effects of macroalgae and micropredation on the postsettlement success of the Green Sea Urchin in Maine. Ph.D. Thesis. University of Maine, Orono, ME. Miller, R. 1985a. Succession in sea urchin and seaweed abundance in Nova Scotia, Canada. Marine Biology 84:275–286. Miller, R. 1985b. Seaweeds, sea urchins, and lobsters: A reappraisal. Canadian Journal of Fisheries and Aquatic Sciences 43:2061–2102. Minchinton, T., R. Scheibling, and H. Hunt. 1997. Recovery of an intertidal assemblage following a rare occurrence of scouring by sea ice in Nova Scotia, Canada. Botanica Marina 40:139–148. Myers, R., and N. Cadigan. 1995. Was an increase in natural mortality responsible for the collapse of northern Cod? Canadian Journal of Fisheries and Aquatic Sciences 52:1274–1285. Norall, T., A. Mathieson, and J. Kilar. 1981. Reproductive ecology of four subtidal red algae. Journal of Experimental Marine Biology and Ecology 54:119–136. Perestenko, L. 1996. Littoral and sublittoral vegetation of the southwest coast of the Okhotsk Sea and the Shantar Is. Botanical serial 1996:1–22. (In Russian) Pielou, E. 1991. After the Ice Age. University of Chicago Press, Chicago, IL. 366 pp. Pinhorn, A. 1984. Inshore exploitation of Atlantic Cod, Gadus morhua, in Labrador and Eastern Newfoundland waters. Journal of Northwest Atlantic Fisheries Science 5:79–84. Rosenzweig, M. 1995. Species Diversity in Space and Time. Cambridge University Press, Cambridge, UK. 436 pp. Scheibling, R. 1986. Increased macroalgal abundance following mass mortalities of sea urchins (Strongylocentrotus droebachiensis) along the Atlantic coast of Nova Scotia. Oecologia 68:186–198. Scheibling, R., and B. Hatcher. 2001. The ecology of Strongylocentrotus drobachiensis. Pp 271– 306, In J.M. Lawrence (Ed.). Edible Sea Urchins: Biology and Ecology. Elsevier Science B.V., Amsterdam, The Netherlands. Scheibling, R., and R. Stephenson. 1984. Mass mortality of Strongylocentrotus droebachiensis (Echinodermata: Echinoidea) off Nova Scotia, Canada. Marine Biology 78:153–164. Scrosati, R., D. Garbary, and J. McLachlan. 1994. Reproductive ecology of Chondrus crispus (Rhodophyta, Gigartinales) from Nova Scotia, Canada. Botanica Marina 37:293–300. Sears, J. 2002. NEAS Keys to the Benthic Marine Algae of the Northeastern Coast of North America from Long Island Sound to the Strait of Belle Isle. 2nd Edition. Northeast Algal Society, Contribution 2:1–161, v–xvii. University of Massachusetts, Dartmouth, MA. Shaw, J. 2006. Paleogeography of Atlantic Candian continental shelves from the Last Glacial Maximum to the Present with an emphasis on Flemish Cap. Journal of Northwest Atlantic Fisheries Science 37:119–126. Shaw, J., P. Gareau, and R. Courtney. 2002. Paleogeography of Atlantic Canada 13-O kyr. Quaternary Science Reviews 21:1861–1878. Sokal, R., and F.J. Rohlf. 1969. Biometry: The Principles and Practice of Statistics in Biological Research. Freeman and Company. San Francisco, CA. 776 pp. South, R., and R. Hooper. 1980. A catalogue and Atlas of the Benthic Marine Algae of the Island of Newfoundland. Memorial University of Newfoundland, Occasional Papers in Biology 3:1–136. 110 Northeastern Naturalist Vol. 18, Monograph No. 8 Stea, R. 2004. The Appalachian Glacier Complex in Maritime Canada. Pp. 213–232, In J. Ehlers and P.L. Gibbard (Eds.). Quarternary Glaciations: Extent and Chronology. Elsevier. B.V., Amsterdam, The Netherlands. 475 pp. Steneck, R., M. Graham, B. Bourgue, D.Corbett, J. Erlandson, J. Estes, and M. Tegner. 2002. Kelp forest ecosystems: Biodiversity, stability, resilience, and future. Environmental Conservation 29:436–459. Steneck, R., R. Bustamente, P. Dayton, G. Jones, and A. Hobday. 2008. Kelp forest ecosystems: Current status and future trends. Pp. 226–241, In N. Polunin (Ed.). Aquatic Ecosystems: Trends and Global Prospects. Cambridge University Press, Cambridge, UK. Stephenson, T., and A. Stephenson. 1954. Life between the tide marks in North America. IIIa. Nova Scotia and Prince Edward Island: Description of the region. Journal of Ecology 42(1):14–45. Strong, W., and L. Hills. 2010. Post-Hypsithermal plant disjunctions in western Alberta, Canada. Journal of Biogeography 30:419–430. Taggart, C., J. Anderson, C. Bishop, E. Colbourne, J. Hutchings, G. Lilly, J. Morgan, E. Murphy, R. Myers, G. Rose, and P. Shelton. 1994. Overview of Cod stocks: Biology and environment in the Northwest Atlantic region of Newfoundland with emphasis on northern Cod. ICES Marine Science Symposium 198:140–157. Taylor, W. 1957. Marine Algae of the Northeastern Coast of North America. University of Michigan Press, Ann Arbor, MI. 509 pp. Toscano, M., and I. Macintyre. 2005. Blanchon, P. comment on Toscano, M. and Macintyre, I. (2003): Corrected western Atlantic sea-level curve for the last 11,000 years based on calibrated 14C dates from Acropora palmata framework and intertidal mangrove peat. Coral Reefs 22:257–270. Vadas, R. 1968. The ecology of Agarum and the kelp bed community. Ph.D. Thesis. University of Washington, Seattle, WA. Vadas, R. 1977. Preferential feeding: An optimization strategy in sea urchins. Ecological Monographs 47: 337–371. Vadas, R., and R. Elner. 1992. Plant-animal interaction in the north-west Atlantic. Pp. 33–60, In D. John, S. Hawkins, and J. Price (Eds.). Plant-Animal Interactions in the Marine Benthos. Clarenden Press, Oxford, UK. Vadas, R., and R. Steneck. 1988. Zonation of Deep Water Benthic Algae in the Gulf of Maine. Journal of Phycology 24:338–346. Vadas, R., B. Beal, W. Wright, S. Nickl and Sheri Emerson. 2004. Growth and productivity of sublittoral fringe kelps (Laminaria longicruis) Bach. Phy. in Cobscook Bay, Maine. Northeastern Naturalist 11(Special Issue 2):143–162. Van den Hoek, C. 1975. Phytogeographic provinces along the coasts of the northern Atlantic Ocean. Phycologia. 14:317. Vermeij. C. 1978. Biogeography and Adaptation. Harvard University Press, Cambridge, MA. 332 pp. Wares, J.P. 2002. Community genetics in the Northwestern Atlantic intertidal. Molecular Ecology 11:1131–1144. Wares, J.P., and C. Cunningham. 2001. Phylogeography and historical ecology of the North Atlantic intertidal. Evolution 55:2455–2469. Wilce, R. 1994. The Arctic subtidal as habitat for macrophytes. Pp. 89–92, In C. Lobbanand and P. Harrison (Eds.). Seaweed Ecology and Physiology. Cambridge University Press, Cambridge, UK. 366 pp. 2011 W.H. Adey and L.-A.C. Hayek 111 Appendix A(1). Exposed stations used in this study. Atl. = N. Atlantic, LS = Labrador Sea, GSL = Gulf of St. Lawrence, and SBI = Strait of Belle Isle. Exposure Trans. to Open ocean or sand or Station Location Date Lat. (N) Long. (W) > 80 km >25 km/notes Land <10 km mud NLQ 03-2 Isle Cumberland, QC 7/24–25/03 51°13'2.4" 58°18'3.8" 168°NE–S–SW 192°W–N 25.5 m 80–100 km (GSL) 03-5 Wall Is., Cape St. Charles, Lab. 8/3–5/03 52°13'15.0" 55°37'5.0" 75°NE (LS) 50°SE–NE, >35 m 25–70 km (SBI) 03-8 White Dog Point, off St. Lunaire, NF 8/9/03 51°29'22.8" 55°27'21.0" 145°SE–NE (LS) 215°N–W–S 29 m 03-11 Ship Is., off Goshen, NF 8/13–14/03 49°39'4.2" 54°35'49.9" 98°N–W (LS) 262°E–S >30 m 03-14 Skerwink Head, off Trinity, NF 8/20–21/03 48°21'51.1" 53°20'26.8" 23°NE–E (LS) 72°S–E, 32–90 km 215°S–W–N 25 m Trinity Bay 03-17 Cape Freels/Lumsden, NF 8/22–24/03 49°19'19.8" 53°35'44.5" 204°E–N–E (LS) 156°E–S–W 7 m 03-19 St. Anthony Bight, NF 8/31–9/1/03 51°21'49.0" 55°33'31.5" 54°E–N–E (LS) 306°S–W–N 25 m 05-15 Barred Is., off Englee, NF 8/12–13/05 50°43'14.9" 56°6'22.0" 89°NE–SW (LS) 69°SE–SW 5 m 06-9 Partridge Pt., off Quirpon, NF 7/10–16/06 51°35'9.4" 55°25'18.6" 135°N–E (LS) 205°S–W >30 m 06-5 Noddy Pt., off Quirpon, NF 7/10–16/06 51°36'3.3" 55°28'6.5" 27°N–NW (LS) 76°NW–SW 257°S–E >30 m 48–70 km 06-2 Cape Daumalen, off Englee, NF 6/25–7/3/06 50°43'17.7" 56°6'6.1" 97°NE–SE (LS) Added because 263°N–W–S >30 m 05-15 to 5 m x̅ = 103° (03-2 x̅ = 86°) x̅ = 22 m GOM 04-31 Cape Elizabeth (CE), ME 9/14/04 43°33'52.0" 70°11'43.5" 122°NE–S (Atl.) 55°NE–SW 183°SW–N 93-10 Cape Elizabeth, ME 10/1/93 43°33'52.0" 70°11'43.5" 122°NE–S (Atl.) 55°NE–SW 183°SW–N ? 112 Northeastern Naturalist Vol. 18, Monograph No. 8 Exposure Trans. to Open ocean or sand or Station Location Date Lat. (N) Long. (W) > 80 km >25 km/notes Land <10 km mud GOM, cont. 07-22 West Cod Ledge, off CE, ME 7/10/07 43°33'6.4" 70°8'5.7" 156°NE–S (Atl.) 15°NE 189°SW–N >30 m (algae limit 25 m) 07-23 Seal Rock, off CE, ME 7/12/07 43°33'11.8" 70°12'52.6" 122°NE–S (Atl.) 55°NE–SW 183°SW–N 94-2 Jewell Is., off Casco Bay, ME 8/24–10/7/94 43°41'12.1" 70°5'3.7" 106°E–S (Atl.) 10°NE 264°NE–N–SW 30 m 92-3 Brothers, off Roque Is., ME 8/8–8/13/92 44°33'19.1" 67°26'11.6" 131°NE–SW (Atl.) 229°SW–N >15 m 07-13 Roque Complex, Grt Spruce Is., ME 6/25/07 44°33'19.3" 67°29'35.5" 93°E–S (Atl.) 267°S–W–NE 07-18 Roque Complex, Grt Spruce Is., ME 7/2/07 44°33'19.3" 67°29'35.5" 93°E–S (Atl.) 267°S–W–NE >20 m x̅ = 117° x̅ = 19 m SNS 04-19 Rose Pt., Lunenburg, NS 8/10/04 44°17'36.5" 64°13'44.7" 117°NE–SE (Atl.) 243°S–W–N 06-18 Ovens Reef, Lunenburg, NS 7/29/06 44°18'51.4" 64°14'20.7" 117°NE–SE (Atl.) 12 m 06-19 East Pt., off Lunenburg, NS 7/30/06 44°20'13.4" 64°12'23.0" 117°NE–SE (Atl.) >20 m 07-3 Cross Is., off Lunenburg, NS 6/2/07 44°18'2.8" 64°11'23.4" 117°NE–SE (Atl.) >30 m 07-5 Port Mouton Head, NS 6/7/07 43°52'19.4" 64°46'59.4" 86°NE–SE (Atl.) >20 m 04-20 West. Channel, Port Mouton, NS 8/11/04 43°53'19.9" 64°47'16.4" 86°NE–SE (Atl.) 274°S–W–N 07-6 Cape Roseway, off Shelburne, NS 6/8–11/07 43°37'44.2" 65°15'34.6" 124°NE–S (Atl.) 236°SW–N >30 m 04-25 Stokes Head, off Shelburne, NS 8/19/04 43°39'27.7" 65°15'24.8" 124°NE–S (Atl.) 04-1 Dover Island, NS 6/29–7/1/04 44°28'46.7" 63°51'29.2" 98°E–SW (Atl.)/ 238°SW–NE 24°25–30km SW x̅ = 106° x̅ = 22 m All stations x̅ = 109° 2011 W.H. Adey and L.-A.C. Hayek 113 Appendix A(2). Intermediate stations. Atl. = N. Atlantic, LS = Labrador Sea, GSL = Gulf of St. Lawrence, and SBI = Strait of Belle Isle. Dist. To Exposure Trans. to Station Location Date Lat. (N) Long. (W) open water Max fetch Local fetch sand or gravel NLQ 03-1 Cumberland Hbr., QC 7/23–7/26/03 51°13'33.2" 58°18'50.2" 1.3 km 6°(80km) SE (GSL) <1 km >20 m 03-4 Tilcey Is., Cape Charles, Lab. 8/2–8/4/03 52°13'26.5" 55°38'1.1" 1.1 km 5°E (LS) <2 km 20 m 03-9 St Lunaire Rd., NF 8/10/03 51°29'9.6" 55°28'31.1" 1.1 km 21°NE–E (LS) <3 km 10 m 05-9 St Lunaire Rd., NF 7/23/05 51°29'41.8" 55°28'2.6" 1.2 km 28°E–SE (LS) <4 km ? 06-12 Jaques Cartier Is., Quirpon, NF 7/10–16/06 51°35'50.3" 55°27'16.4" 0.3 km 0 (no line of sight; 0.3 km) <2 km 12 m 06--3 Outer Englee Hbr., NF 6/25–7/3/06 50°43'21.6" 56°6'21.9" 0.2 km 1°open ocean E (LS) <0.5 km 2 m 03-18 Burnt Is., off Twillingate, NF 8/28/03 49°40'45.9" 54°45'40.8" 1.5 km 23°(140km) NW (LS) <3 km 10 m 03-10 Goshen Arm, NF 8/13–14/03 49°38'41.7" 54°35'55.2" 0.9 km 17°open ocean NE (LS) <2 km >30 m 03-15 Trinity Hbr., NF 8/22/03 48°21'50.0" 53°20'49.8" 1.0 km 0 <2.5 km >20 m x̅ = 1.0 km x̅ = 11.2°open ocean x̅ = 15.5 m GOM 04-29 Bates Is., Casco Bay, ME 9/11/04 43°42'17.9" 70°4'44.3" 1.4 km 28°NE–S (Atl.) <3 km 93-1 Bates Is., Casco Bay, ME 8/14–15/93 43°42'17.9" 70°4'44.3" 1.4 km 28°NE–S (Atl.) <3 km 12 m 04-32 Portland Head, ME 9/15–18/04 43°37'43.0" 70°12'46.9" 0 km 32°NE–SE (Atl.) <9 km 93-3 Portland Head, ME 7/13/93 43°37'43.0" 70°12'46.9" 0 km 32°NE–SE (Atl.) <9 km 15 m 92-2 Englishman Bay, ME 8/8–13/92 44°35'15.8" 67°29'40.0" 6.1 km 20°NE–E (Atl.) <6 km 07-14 Englishman Bay, ME 6/26/07 44°35'15.8" 67°29'40.0" 6.1 km 20°NE–E (Atl.) <6 km 9 m 07-17 Roque Is., ME 7/1/07 44°34'34.7" 67°29'52.4" 3.9 km 21°E–S (Atl.) <3 km 12 m x̅ = 2.9 km x̅ = 25°open ocean x̅ = 12 m SNS 04-2 Mid Dover, NS 6/29–30/04 44°28'59.7" 63°51'38.3" 0.6 km 5°SE (Atl.) <0.5 km 04-18 Blue Rocks, Lunenburg, NS 8/5/04 44°21'15.5" 64°14'50.9" 7.4 km 24°SE (Atl.) <5.7 km 06-16 Feltzen, Lunenburg, NS 7/26/06 44°19'54.5" 64°17'6.9" 9.9 km 11°ENE (Atl.) <7.0 km 10 m 07-2 Feltzen, Lunenburg, NS 6/1/07 44°19'54.5" 64°17'6.9" 9.9 km 11°ENE (Atl.) <7.0 km 04-21 Massacre Is, Port Mouton, NS 8/12/04 43°54'23.7" 64°47'50.0" 4.1 km 25°SE (Atl.) <5 km 07-4 Massacre Is., Port Mouton, NS 6/6/04 43°54'23.7" 64°47'50.0" 4.1 km 25°SE (Atl.) <5 km 10 m 04-24 Sandy Pt., Shelburne Hbr., NS 8/18/04 43°41'14.6" 65°19'11.8" 7.7 km 12°SE (Atl.) <3 km 07-7 Sandy Pt., Shelburne Hbr., NS 6/11/07 43°41'14.6" 65°19'11.8" 7.7 km 12°SE (Atl.) <3 km 7.5 m x̅ = 5.9 km x̅ = 15.4° x̅ = 9 m All stations x̅ = 17.2° 114 Northeastern Naturalist Vol. 18, Monograph No. 8 Appendix A(3). Protected stations used in this study. Atl. = N. Atlantic, LS = Labrador Sea, GSL = Gulf of St. Lawrence, SBI = Strait of Belle Isle, and GOM = Gulf of Maine. Permanent Distance to Exposure sand/gravel Station Location Date Lat. (N) Long. (W) open water Max fetch Max local fetch line NLQ 03-3 Inner Baie de Jaques Cartier, QC 7/27–29/03 51°17'44.0" 58°17'17.5" 6 km 2°2.8 km 0.68 km 10 m 03-7 E. Port Markham, Lab. 8/6/03 52°22'20.8" 55°44'57.3" 12.7 km 5°4.3 km <1.0 km 14 m 03-12 SW Goshen Arm, NL 8/15/03 49°36'48.8" 54°38'32.3" 6.4 km 7°2.8 km <1.3 km 1 0 m 03-20 St. Anthony Hbr., NL 9/1/03 51°21'37.0" 55°34'26.7" 4 km 12°1.6 km <1.0 km 7 m 05-8 St. Lunaire Rd., NL 7/21/05 51°29'2.2" 55°28'36.6" 2.4 km 35°2.0 km <20 km 3 m 06-10 J Cartier Is, Quirpon Hbr., NL 7/10–16/06 51°35'26.3" 55°27'34.4" 1.5 km 18°1.6 km <1.2 km 6 m 06-13 Vincent Is, Quirpon Hbr., NL 7/10–16/06 51°35'23.0" 55°27'7.4" 1.5 km 21°1.6 km <1.0 km 6 m x̅ = 4.9 km x̅ = 0°open ocean x̅ = 3.7 km x̅ = 8 m GOM 93-2 Stave Is, Casco Bay, ME 9/4–9/7/93 43°42'47.4" 70°5'19.0" 9.2 km 15°4 km <3 km 6 m 07-19 Stave Is, Casco Bay, ME 7/6/07 43°42'47.4" 70°5'19.0" 5.7 km 5°(GOM) <4.2 km 94-3 Rogue Is., Casco Bay, ME 10/4–10/6/94 43°42'46.2" 70°5'21.0" 3.0 km 7°6 km <2 km 10 m 04-30 Rogue Is., Casco Bay, ME 9/11/04 43°42'46.2" 70°5'21.0" 3.0 km 7°6 km <2 km 10 m 07-15 Roque Is., Englishman Bay, ME 6/27/07 44°34'20.7" 67°31'0.0" 9.6 km 13°4.2 km <3 km 07-16 Roque Is., Englishman Bay, ME 7/1/07 44°34'34.1" 67°30'0.6" 6.9 km 6°(GOM) <4 km 12 m x̅ = 6.9 km x̅ = 2.2°open ocean x̅ = 3.2 km x̅ = 9 m SNS 04-3 Blind Bay, NS 7/2/04 44°31'3.3" 63°50'1.0" 4.4 km 0° <1.5 km 04-17 Battery Pt., Lunenburg Bay, NS 8/5/04 44°21'35.7" 64°17'44.1" 10.7 km 4°(Atl.) (shoals seaward) <9 km 06-17 Battery Pt., Lunenburg Bay, NS 7/27/06 44°21'35.7" 64°17'44.1" 10.7 km 4°(Atl.) (shoals seaward) <9 km 6 m 07-1 Battery Pt., Lunenburg Bay, NS 6/1/07 44°21'35.7" 64°17'44.1" 10.7 km 4°(Atl.) (shoals seaward) <9 km 04-22 Port Mouton, NS 8/13/04 43°55'4.0" 64°50'7.4" 7.0 km 10°(Atl.) (15°,6 km) <4 km ? (shoals seaward) 04-23 Birchtown Bay, Shelburne, NS 8/16/04 43°43'44.3" 65°20'27.2" 14.2 km 0° <5 km 07-9 Birchtown Bay, Shelburne, NS 6/14/07 43°43'44.3" 65°20'27.2" 14.2 km 0° <5 km 5 m x̅ = 9.3 km x̅ = 3.5°open ocean x̅ = 5.5 m All stations x̅ = 9.3 km x̅ = 1.9° x̅ = 7.5 m 2011 W.H. Adey and L.-A.C. Hayek 115 Appendix A(4). Rocky Pinnacle stations used in this study. Dist. to Mud/sand Station Location Date Lat. (N) Long. (W) open water Max local fetch depth NLQ 03-6 South Cove, Cape Charles Hbr., Lab. 8/4/03 52°13'10.2" 55°38'22.2" 3 km 37°,0.7–1 km 2.5 m 03-12 Southwest Cove, Goshen Arm, NF 8/15/03 49°36'44.1" 54°38'45.6" 6 km 5°, 3.7 km 10–12 m 03-13 Swains Is. Tickle, off Wesleyville, NF 8/17/03 49°8'28.7" 53°33'25.4" 6 km 5°, 2 km 3 m 03-16 Trinity Hbr, NF, high spots N of town 8/22/03 48°21'41.8" 53°20'19.0" 2 km 2 km 10–20 m 03-17 Trinity Hbr, NF, high spots S of town 8/22/04 48°21'56.5" 53°21'50.8" 4 km 1 km 10–20 m 03-20 St Anthony Hbr., Harbor Rock, NF 9/1/03 51°21'31.8" 55°34'5.6" 3 km 6°, 3 km 1 m 05-17 Englee Hbr., off Lion Hill, NF 8/14/05 50°43'30.3" 56°6'21.5" 0.5 km 6°, (LS) at 0.5 km mud-gravel 06-4 Englee Hbr., off Lion Hill, NF 7/2/06 50°43'30.3" 56°6'21.5" 0.5 km 6°, (LS) at 0.5 km mud-gravel 116 Northeastern Naturalist Vol. 18, Monograph No. 8 Appendix B. Station summary analysis. Protected/ Exposed Intermediate rocky pinnacles Σ NLQ # stations (sites) 11 (8) 9 (9) 15 (3) / 6 (6) 41/26 # quadrats 72 52 34 / 20 178 GOM # stations (sites) 8 (4) 7 (4) 6 (4) 21 (12) # quadrats 51 53 26 130 SNS # stations (sites) 9 (4) 8 (5) 7 (4) 24 (13) # quadrats 34 30 26 90 Total # stations (sites) 28 (16) 24 (18) 34 (17) 86 (51) # quadrats 157 135 106 398 Appendix C. Mean exposure status of sites used in this study. O = open ocean, D = distance, and SL = sediment line. Exposed Intermediate Protected O D SL O D SL O D SL NLQ 103° 0 22 m 11.2° 1.0 km 15.5 m 0.0° 4.9 km 8.0 m GOM 117° 0 19 m 25.0° 2.9 km 12.0 m 2.2° 6.9 km 9.0 m SNS 106° 0 22 m 15.4° 5.9 km 9.0 m 3.5° 9.3 km 5.5 m x̅ 109° 0 21 m 17.0° 3.3 km 12.2 m 1.9° 7.0 km 7.5 m 2011 W.H. Adey and L.-A.C. Hayek 117 Appendix D1. Newfoundland/Labrador/Quebec (NLQ) species found in exposed, intermediate, and protected locations (mean biomas [wet] g/m2). Infra 0.5 m 2.5 m 5 m 10 m 20 m 30 m NLQ exposed Fucus distichus 58.9 0.0 0.0 0.0 0.0 0.0 0.0 Chordaria flagelliformis 1911.9 409.6 60.2 0.1 0.0 0.0 0.0 Alaria esculenta 162.9 1824.4 1530.4 751.5 0.3 0.0 0.0 Laminaria digitata 0.7 71.5 720.8 754.7 0.0 0.0 0.0 Saccorhiza dermatodea 0.0 55.0 218.3 31.8 0.0 0.0 0.0 Agarum clathratum 0.0 0.0 513.1 862.7 1806.2 377.5 20.7 Desmarestia viridis 19.3 0.1 275.7 568.9 261.7 0.0 0.0 Saccharina latissima 0.0 1.2 26.1 38.7 0.0 0.0 0.0 Desmarestia aculeata 0.0 1.7 1.6 0.9 0.0 0.0 0.0 Devaleraea ramentacea 352.6 376.5 31.5 0.0 0.0 0.0 0.0 Acrosiphonia arcta 110.7 112.5 7.9 0.7 0.0 0.0 0.0 Petalonia fascia 28.4 0.0 0.6 0.0 0.0 0.0 0.0 Scytosiphon lomentaria 2.1 0.0 0.1 0.0 0.0 0.0 0.0 Corallina officinalis 2.5 0.0 0.7 0.0 0.0 0.0 0.0 Rhodomela confervoides 0.0 0.0 1.5 0.1 0.0 0.0 0.0 Phycodrys riggii 0.0 0.0 0.6 3.0 31.4 1.0 0.0 Ptilota serrata 0.0 0.0 0.1 19.3 120.9 484.6 54.0 Turnerella pennyi 0.0 0.0 0.0 0.0 0.2 3.9 49.3 Ectocarpus spp. 25.0 4.3 0.3 0.2 0.0 0.0 0.0 Dictyosiphon foeniculaceus 314.3 7.8 152.5 0.0 0.0 0.0 0.0 Pilayella littoralis 5.9 2.3 20.0 0.1 0.0 0.0 0.0 Polysiphonia urceolata 0.0 0.7 0.0 0.2 0.0 0.0 0.0 Polysiphonia flexicaulis 0.0 0.0 0.3 0.4 0.0 0.0 0.0 Palmaria palmata 0.0 0.0 111.9 0.0 0.0 0.0 0.0 Euthora cristata 0.0 0.0 1.6 3.7 1.1 0.2 0.2 Membranoptera alata 0.0 0.0 0.0 0.1 0.1 0.2 0.0 Scagelia pylaisaei 0.0 0.0 0.1 0.0 0.7 2.9 0.0 Total biomass 2995.1 2867.5 3675.8 3037.0 2222.5 870.3 124.3 Total biomass sum 15,792.0 NLQ intermediate Chorda filum 0.0 26.4 0.0 0.0 0.0 0.0 0.0 Fucus distichus 2498.2 0.0 0.0 0.0 0.0 0.0 0.0 Chordaria flagelliformis 1863.6 1144.2 16.8 0.0 0.0 0.0 0.0 Alaria esculenta 35.0 1540.3 183.4 0.4 0.0 0.0 0.0 Laminaria digitata 0.0 0.0 10.0 0.0 0.0 0.0 0.0 Saccorhiza dermatodea 0.0 648.6 0.0 0.0 0.0 0.0 0.0 Agarum clathratum 0.0 107.3 1994.5 1761.3 1142.0 1390.0 900.0 Desmarestia viridis 195.7 110.3 1022.6 627.5 0.0 43.3 170.0 Saccharina latissima 0.0 66.4 203.0 0.0 0.0 0.0 0.0 Devaleraea ramentacea 239.1 35.0 0.0 0.0 0.0 0.0 0.0 Acrosiphonia arcta 132.9 24.3 0.1 0.0 0.0 0.0 0.0 Scytosiphon lomentaria 37.9 2.0 0.0 0.0 0.0 0.0 0.0 Corallina officinalis 0.0 0.0 1.7 0.1 0.1 0.0 0.0 Rhodomela confervoides 0.0 0.1 0.0 0.0 0.0 0.0 0.0 Phycodrys riggii 0.0 0.6 1.8 5.2 0.9 0.1 0.0 Petalonia fascia 26.5 0.4 0.1 0.1 0.0 0.0 0.0 Ptilota serrata 0.0 0.4 0.4 1.0 315.7 266.7 330.0 Chondrus crispus 0.1 0.0 0.0 0.0 0.0 0.0 0.0 118 Northeastern Naturalist Vol. 18, Monograph No. 8 Infra 0.5 m 2.5 m 5 m 10 m 20 m 30 m Turnerella pennyi 0.0 0.0 0.2 0.8 2.6 0.9 1.0 Neodilsea integra 0.0 0.0 0.0 0.0 0.0 0.2 0.0 Monostroma sp. 17.8 0.0 0.5 0.0 0.0 0.0 0.0 Elachista fucicola 3.9 0.0 0.0 0.0 0.0 0.0 0.0 Ectocarpus spp. 10.6 0.0 2.7 2.7 0.0 0.0 0.0 Dictyosiphon foeniculaceus 86.1 266.6 12.3 0.0 0.0 0.0 0.0 Pilayella littoralis 171.3 2.1 0.3 0.0 0.0 0.0 0.0 Polysiphonia urceolata 0.0 0.0 2.2 0.0 0.2 0.0 0.0 Polysiphonia flexicaulis 0.0 0.6 0.0 3.3 0.0 0.0 0.0 Ceramium sp. 6.7 0.0 0.2 0.0 0.0 0.0 0.0 Euthora cristata 0.0 0.0 0.0 0.1 0.1 0.5 0.1 Membranoptera alata 0.0 0.0 0.0 0.0 0.0 0.2 0.1 Scagelia pylaisaei 0.0 0.3 0.7 0.1 0.1 0.0 0.0 Poryphyra sp. 0.0 2.8 0.0 0.0 0.0 0.0 0.0 Chaetomorpha sp. 0.6 0.0 0.0 0.0 0.0 0.0 0.0 Total biomass 5325.9 3978.7 3453.3 2402.4 1461.7 1702.0 1401.2 Total biomass sum 17,347.0 NLQ protected Fucusdistichus 378.3 0.0 0.0 0.0 0.0 Chordaria flagelliformis 410.8 103.4 0.0 0.0 0.0 Alaria esculenta 0.0 192.8 0.0 0.0 0.0 Agarum clathratum 0.0 203.8 2292.5 336.4 183.1 Desmarestia viridis 0.0 206.9 1388.9 369.4 11.3 Acrosiphonia arcta 162.5 0.0 0.0 0.0 0.0 Scytosiphon lomentaria 0.0 2.4 0.0 0.0 0.0 Petalonia fascia 0.0 8.2 0.0 0.0 0.0 Phycodrys riggii 0.0 0.0 0.1 1.3 3.1 Ptilota serrata 0.0 0.0 0.0 1.2 3.8 Turnerella pennyi 0.0 0.0 0.0 0.3 6.2 Corallina officinalis 0.0 0.0 0.0 0.1 0.2 Neodilsea integra 0.0 0.0 0.0 0.0 0.1 Rhodomela confervoides 57.5 57.0 0.0 0.0 Ectocarpus spp. 0.0 0.0 0.0 0.1 0.0 Pilayella littoralis 6.7 0.1 0.0 0.2 0.0 Dictyosiphn foeniculaceus 29.2 297.0 0.0 0.0 0.0 Polysiphonia urceolata 0.0 0.0 0.1 0.0 0.1 Polysiphonia flexicaulis 0.0 12.5 0.4 0.0 0.0 Euthora cristata 0.0 0.0 0.0 0.0 0.7 Scagelia pylaisaei 0.0 0.0 0.6 0.0 0.1 Monostroma sp. 23.3 0.0 0.0 0.0 0.0 Total biomass 1068.4 1084.1 3682.7 709.2 208.7 Total biomass sum 6753.0 2011 W.H. Adey and L.-A.C. Hayek 119 Appendix D2. Species found in the rocky pinnacles canopy of Newfoundland/Labrador/Quebec (NLQ) (mean biomas [wet] g/m2). Species Infra 0.5 m 2.5 m 5 m 10 m 20 m NLQ rocky pinnacles Alaria esculenta 25.0 176.5 68.3 0.0 0.0 0.0 Chordaria flagelliformis 47.8 0.9 0.9 0.0 0.0 0.0 Agarum clathratum 0.0 0.0 100.8 110.0 1582.5 1510.0 Desmarestia viridis 0.0 331.8 18.3 0.0 14.8 1.0 Desmarestia aculeata 0.0 163.3 47.7 0.0 0.0 0.0 Saccharina longicruris 0.0 2456.7 4305.0 0.0 4790.0 0.0 Saccharina latissima 82.5 4536.7 616.4 30.0 1.9 0.0 Chorda filum 0.0 0.0 3.1 0.0 0.0 0.0 Saccorhiza dermatodea 0.0 211.7 53.3 0.0 0.0 0.0 Fucus distichus 915.0 6.9 205.0 0.0 4.1 0.0 Devaleraea ramentacea 960.0 1.6 0.0 0.0 0.0 0.0 Acrosiphonia arcta 1.9 0.2 0.0 0.0 0.0 0.0 Rhodomela confervoides 120.0 0.1 0.2 0.0 0.4 0.0 Polysiphonia spp. 0.0 0.0 0.0 0.0 1.4 0.0 Chondrus crispus 472.5 0.8 0.2 0.0 0.0 0.0 Petalonia fascia 0.0 0.2 0.0 0.0 0.0 0.0 Corallina officinalis 0.0 1.1 5.8 0.0 0.0 0.0 Ptilota serrata 0.0 0.0 0.0 0.0 0.3 1.0 Scytosiphon lomentaria 0.1 Elachista fucicola 2.5 0.0 0.0 0.0 0.0 0.0 Pilayella littoralis 23.8 0.0 0.0 0.0 0.0 0.0 Dictyosiphon foeniculaceus 5.0 0.7 0.0 0.0 0.0 0.0 Palmaria palmata 560.0 0.8 0.0 0.0 0.0 0.0 Porphyra sp. 1.6 0.0 0.0 0.0 0.0 0.0 Scagelia pylaiisii 0.1 0.0 0.0 0.0 0.8 0.1 Ulva lactuca 16.3 4.6 0.0 0.0 0.0 0.0 Chaetomorpha spp. 272.5 0.0 0.0 0.0 0.6 0.0 Ceramium spp. 0.0 0.0 0.0 0.0 1.0 0.0 Polysiphonia urceolata 0.0 0.0 0.0 0.0 0.1 0.5 Ectocarpus sp. 0.0 2.2 2.4 35.0 4.5 0.0 Stilophora rhizoides 0.0 0.0 3.1 0.0 0.0 0.0 Antithamnion sp. 0.0 1.1 0.0 0.0 0.0 0.0 Cladophora sp. 0.0 1.4 2.6 0.0 0.0 0.0 Monostroma sp. 2.5 0.3 0.0 Stictyosiphon tortilis 35.0 0.0 0.0 Delamerea sp. 0.0 0.1 0.0 0.0 Total biomass 3508.8 7899.5 5433.1 210.1 6402.3 1512.6 Total biomass sum 24,966.0 120 Northeastern Naturalist Vol. 18, Monograph No. 8 Appendix D3. Species found in the Gulf of Maine (GOM) (mean biomas [wet] g/m2). Species Infra 0.5 m 2.5 m 5 m 10 m 20 m 25 m GOM exposed combined Alaria esculenta 140.2 4090.5 10.0 15.6 11.1 0.0 Laminaria digitata 47.5 1094.1 1362.5 1794.4 489.1 0.0 Saccharina latissima 68.8 511.3 1195.0 1921.6 56.7 0.0 Saccharina longicruris 0.0 1051.3 3163.8 14.4 0.0 0.0 Saccorhiza dermatodea 0.0 72.0 0.0 11.1 0.0 0.0 Agarum clathratum 0.0 0.0 0.0 38.9 601.7 60.6 Fucus distichus 102.5 4.5 0.0 0.0 0.0 0.0 Desmarestia viridis 0.0 1.3 0.0 131.1 33.3 0.0 Desmarestia aculeata 0.0 0.0 0.0 153.3 0.0 0.0 Chondrus crispus 2951.5 131.3 233.8 37.8 4.4 0.0 0.0 Mastocarpus stellatus 80.5 4.7 0.0 0.0 0.0 0.0 0.0 Devaleraea ramentacea 173.3 132.0 0.0 0.0 0.0 0.0 0.0 Corallina officinalis 17.5 32.5 0.2 7.8 11.1 0.0 0.0 Cystoclonium purpureum 32.8 12.5 400.0 22.2 0.5 0.0 0.0 Rhodomela confervoides 28.6 267.6 0.0 3.0 0.2 0.0 0.0 Phyllophora pseudoceranoides 0.9 0.0 171.3 82.5 17.5 4.9 10.3 Ptilota serrata 0.0 0.0 0.0 0.2 51.8 12.6 0.1 Fimbriofolium dichotomum 0.0 0.0 0.0 0.0 3.0 1.0 0.0 Phycodrys rubens 0.0 0.2 25.6 51.1 88.6 6.5 0.7 Ceramium rubrum 127.8 7.8 39.6 0.0 0.0 0.0 0.0 Neosiphonia harveyii 155.0 109.4 0.0 36.2 0.0 0.1 0.5 Polysiphonia spp. 0.0 0.0 2.5 0.0 0.0 0.0 0.0 Dumontia incrassata 7.5 0.0 20.0 8.9 0.0 0.0 0.0 Palmaria palmata 24.6 72.9 118.8 11.2 0.0 0.0 0.0 Membanoptera alata 0.0 0.0 0.0 0.0 0.2 0.0 0.0 Antithamnion spp. 0.0 0.0 0.0 0.0 0.0 0.0 0.4 Bonnemaisonia hamifera 0.0 0.0 0.6 0.0 0.0 0.0 0.0 Porphyra (purpurea) 18.4 7.4 0.2 0.0 0.0 0.0 0.0 Polyides caprinus 0.0 0.0 2.5 0.0 0.0 0.0 0.0 Monostroma spp. 8.9 2.0 6.3 140.0 0.0 0.0 0.0 Euthora cristata 0.0 0.0 17.8 34.3 50.5 0.4 0.4 Total biomass 3986.2 7605.3 6770.4 4515.8 1420.0 86.0 12.2 Total biomass sum 24,396.0 GOM intermed combined Alaria esculenta 4.3 0.0 15.0 3.1 0.0 Laminaria digitata 0.0 65.7 1307.9 146.2 15.0 Saccharina latissima 0.0 1997.9 1388.3 843.8 685.1 Saccharina longicruris 0.0 0.0 1100.0 936.2 31.3 Saccorhiza dermatodea 0.0 0.0 75.0 0.0 22.5 Agarum clathratum 0.0 59.3 81.7 50.8 404.4 Fucus distichus 116.7 6.4 0.0 0.0 0.0 Desmarestia viridis 0.0 0.0 6.2 51.5 0.0 Desmarestis aculeata 0.0 0.0 0.0 0.0 33.4 Chordaria flagelliformis 0.0 25.0 0.0 0.0 0.0 Chondrus crispus 2929.3 2513.6 1973.6 186.2 17.5 Mastocarpus stellata 18.4 6.7 0.0 0.0 0.0 Devaleraea ramentacea 13.2 0.0 0.0 0.0 0.0 Corallina officinalis 2.0 368.2 9.2 0.0 0.0 Ahnfeltia plicata 0.0 0.7 6.7 0.0 0.0 Cystocloniun purpureum 0.0 747.6 55.2 203.9 32.5 2011 W.H. Adey and L.-A.C. Hayek 121 Species Infra 0.5 m 2.5 m 5 m 10 m 20 m 25 m Rhodomela confervoides 16.0 0.0 0.2 0.4 1.8 Phyllophora pseudoceranoides 0.0 1.4 152.9 201.4 69.6 Ptilota serrata 0.0 0.0 0.7 6.8 22.0 Fimbriofolium dichotomum 0.0 0.0 0.0 3.1 17.3 Phycodrys rubens 0.0 7.9 99.4 73.2 73.4 Acrosiphonia arcta 485.5 0.0 Ceramium rubrum 157.5 49.3 619.9 95.4 58.9 Neosiphonia harveyii 0.0 0.0 4.9 170.8 207.2 Polysiphonia spp. 0.0 3.6 0.0 0.0 2.5 Dumontia incrassata 26.0 3.4 2.5 2.3 0.0 Palmaria palmata 4.3 0.0 4.2 0.0 0.0 Bonnemaisonia hamifera 0.0 0.0 16.4 0.0 0.2 Porphyra (purpurea) 7.7 0.0 0.0 0.9 5.1 Polyides caprinus 0.0 0.0 2.5 1.5 0.0 Elachista fucicola 29.7 0.0 0.0 0.0 0.0 Monostroma spp. 41.9 5.6 0.9 0.0 1.3 Chaetomorpha spp. 98.3 14.3 55.5 0.0 0.0 Cladophora spp. 6.7 0.0 0.0 0.0 0.0 Euthora cristata 0.0 0.0 156.3 40.9 78.0 Total biomass 3957.4 5876.6 7134.9 3018.5 1779.1 Total biomass sum 21,767.0 GOM protected combined Alaria esculenta 0.0 0.0 82.5 0.0 Laminaria digitata 0.0 0.0 85.3 0.0 Saccharina latissima 0.0 1241.4 1802.5 1297.8 Saccharina longicruris 0.0 0.0 3530.0 611.1 Agarum clathratum 0.0 2.9 305.0 362.3 Ascophyllum nodosum 7.2 0.0 0.0 0.0 Fucus vesiculosus 50.0 0.0 0.0 0.0 Fucus distichus 171.3 171.3 0.0 0.0 Desmarestia viridis 0.0 0.0 412.1 0.0 Desmarestia aculaeata 0.0 0.0 0.0 30.4 Chondrus crispus 1334.0 1386.3 101.7 0.0 Mastocarpus stellata 11.4 11.4 0.0 0.0 Devaleraea ramentacea 108.7 0.0 0.0 0.0 Corallina officinalis 26.5 129.3 12.0 0.0 Cystoclonium purpureum 175.3 61.6 37.1 0.4 Rhodomela confervoides 0.0 6.0 30.3 0.9 Phyllophora pseudoceranoides 0.0 38.7 9.8 0.5 Ptilota serrata 0.0 0.0 5.4 0.4 Phycodrys rubens 0.0 2.9 12.8 0.0 Fimbriofolium dichotomum 0.0 0.0 0.0 0.8 Scytosiphon lomentaria 7.4 0.0 0.0 0.0 Acrosiphonia arcta 37.4 0.0 Ceramium circinnatum 4.0 0.0 0.0 0.0 Ceramium rubrum 0.0 19.3 0.0 28.9 Neosiphonia harveyii 102.3 62.1 89.7 4.0 Polysiphonia spp. 0.0 54.4 0.0 4.5 Dumontia incrassata 27.4 41.4 2.8 0.0 Palmaria palmata 0.0 1.1 2.9 20.6 Membranoptera alata 0.0 0.0 0.4 0.0 Bonnemaisonia hamifera 0.0 0.0 39.7 0.0 Antithamnion spp. 0.0 7.1 0.0 0.0 122 Northeastern Naturalist Vol. 18, Monograph No. 8 Species Infra 0.5 m 2.5 m 5 m 10 m 20 m 25 m Porphyra (purpurea) 113.0 187.6 0.0 0.0 Elachista fucicola 2.1 0.0 0.0 0.0 Monostroma spp. 1.7 0.0 0.1 0.0 Euthora cristata 19.3 20.8 Total biomass 2179.8 3424.9 6581.4 2383.4 Total biomass sum 14,570.0 2011 W.H. Adey and L.-A.C. Hayek 123 Appendix D4. Species found in southern Nova Scotia (mean biomas [wet] g/m2). Species Infra 0.5 m 2.5 m 5 m 10 m 20 m NS exposed combined Alaria esculenta 0.0 255.7 348.0 30.0 22.0 0.0 Laminaria digitata 0.0 24,162.9 6460.0 1578.0 543.5 20.0 Laminaria saccharina 0.0 3318.6 648.0 1360.0 1958.0 118.0 Agarum clathratum 0.0 0.0 0.9 70.0 682.0 544.0 Chordaria flagelliformis 0.0 0.7 0.0 0.0 0.0 0.0 Fucus distichus 31.7 2.9 0.0 0.0 0.0 0.0 Fucus serratus 0.0 179.3 146.0 870.0 104.0 0.0 Desmarestia viridis 0.0 0.0 0.0 11.4 1.5 0.0 Desmarestia aculeata 0.0 8.6 0.0 3.2 3.1 0.0 Codium fragile 0.0 0.0 13.7 3448.0 0.0 0.0 Chondrus crispus 8658.3 1353.6 189.7 24.9 7.0 1.9 Devaleraea ramentacea 0.0 35.7 0.0 0.0 0.0 0.0 Corallina officinalis 102.4 1065.7 2.8 24.6 16.0 1.6 Ahnfeltia plicata 5.6 27.9 0.0 0.0 0.0 0.0 Cystoclonium purpureum 0.6 35.0 7.9 34.4 52.4 0.0 Rhodomela confervoides 0.0 2.9 2.3 8.7 0.1 0.0 Phyllophora pseudoceranoides 0.0 0.0 2.9 6.5 3.7 0.0 Ptilota serrata 0.0 0.2 0.0 0.3 27.0 68.8 Fimbriofolium dichotomum 0.0 0.0 0.0 0.0 0.0 0.8 Phycodrys rubens 0.0 0.0 1.4 0.2 3.5 1.4 Turnerella pennyii 0.0 0.0 0.0 0.0 0.0 0.4 Ceramium rubrum 1.2 22.3 3.1 45.1 4.2 0.0 Neosiphonia harveyii 0.0 0.0 14.1 26.3 1.3 0.0 Elachista fucicola 0.0 7.1 0.0 0.0 0.0 0.0 Dumontia incrassata 0.0 5.0 0.0 0.0 0.0 0.0 Nemalion multifidum 16.1 5.0 0.0 0.0 0.0 0.0 Palmaria palmata 0.0 140.7 18.0 2.9 9.0 0.0 Membranoptera spp. 0.0 0.0 0.0 0.0 0.0 0.1 Scagelia pylaisaei 0.0 0.0 0.0 0.0 0.6 0.8 Antithamnion spp. 0.0 0.0 0.3 0.5 0.1 0.8 Callithamnion tetragonum 0.0 0.0 11.7 0.0 0.0 0.0 Acrothrix nova-angliae 0.6 0.0 0.0 0.0 0.0 0.0 Leathesia difformis 0.0 135.0 0.0 0.0 0.0 0.0 Pilayella littoralis 0.6 0.7 0.0 0.0 0.0 0.0 Monostroma spp. 3.7 0.0 0.0 0.0 0.0 0.0 Euthora cristata 0.0 0.0 0.5 0.1 1.7 0.8 Total biomass 8820.7 30,765.4 7871.4 7545.0 3441.0 759.3 Total biomass sum 59,202.0 NS intermediate combined Alaria esculenta 0.0 0.0 0.0 1.4 0.0 Laminaria digitata 2.2 200.0 272.0 577.9 84.9 Saccharina latissima 5.6 5071.9 1996.0 744.1 153.5 Saccorhiza dermatodea 0.0 0.0 30.0 51.7 8.6 Agarum clathratum 0.0 0.0 0.0 7.0 200.3 Fucus vesiculosus 45.0 0.0 0.0 0.0 0.0 Fucus distichus 0.0 1441.9 0.0 0.0 0.0 Fucus serratus 0.0 9.4 1704.0 394.0 69.3 Desmarestia viridis 0.0 0.0 62.6 263.7 27.1 Desmarestia aculeata 0.0 0.0 2.5 2.0 63.4 124 Northeastern Naturalist Vol. 18, Monograph No. 8 Species Infra 0.5 m 2.5 m 5 m 10 m 20 m Ascophyllum nodosum 0.0 0.0 748.0 0.0 0.0 Codium fragile 20.0 0.0 0.0 0.0 0.0 Chorda filum 0.0 0.0 8.9 16.0 80.0 Chordaria flagelliformis 1.7 18.1 0.0 0.0 0.0 UID filiform brown 0.0 0.0 0.0 5.9 11.7 Chondrus crispus 5064.4 679.4 238.1 34.5 0.0 Mastocarpus stellata. 1.1 0.0 0.0 0.0 0.0 Corallina officinalis 1013.9 1863.1 1361.8 29.0 7.0 Ahnfeltia plicata 0.0 1.3 0.0 6.3 0.0 Cystoclonium purpureum 2.8 125.1 116.7 238.3 1.3 Rhodomela confervoides 20.0 75.6 35.2 7.8 0.4 Phyllophora pseudoceranoides 0.0 0.0 0.0 318.8 352.3 Acrosiphonia arcta 2.2 8.1 13.0 0.0 0.0 Phycodrys rubens 0.0 0.0 0.0 4.4 13.1 Polyides caprinus 0.0 0.0 0.0 0.0 47.2 Pilayella littoralis 0.0 6.4 0.0 0.0 0.0 Ceramium rubrum 30.0 148.2 38.0 41.9 10.6 Neosiphonia harveyii 0.0 5.6 13.9 18.5 0.0 Polysiphonia spp. 1.1 7.5 0.8 5.3 1.8 Dumontia incrassata 1.1 0.6 0.0 0.0 0.0 Palmaria palmata 5.6 235.0 5.9 0.4 3.1 Pantoneura fabriciana 0.0 0.0 0.0 0.1 0.0 Antithamnion spp. 40.6 311.9 1.3 3.3 0.0 Ulva lactuca 0.0 1.9 0.0 0.0 0.0 Ectocarpus spp. 0.0 0.0 0.4 2.1 49.9 Cladophora spp. 0.0 2.2 0.0 2.2 0.0 Chaetomorpha spp. 8.6 3.8 0.0 0.0 26.9 Monostroma spp. 0.0 1.9 0.0 0.5 0.0 Euthora cristata 0.0 0.0 0.2 0.0 16.0 Total biomass 6266.0 10,218.8 6649.5 2777.1 1228.5 Total biomass sum 27,140.0 NS protected combined Laminaria digitata 0.0 0.0 945.0 447.5 Saccharina latissima 188.9 2977.2 250.0 281.2 Saccharina longicruris 0.0 0.0 1157.5 1837.5 Saccorhiza dermatodea 0.0 222.2 0.0 0.0 Ascophyllum nodosum 3.3 7.8 0.0 0.0 Fucus vesiculosus 551.7 240.6 0.0 0.0 Fucus distichus 442.8 412.2 2.5 0.0 Fucus serratus 0.0 9.4 560.0 1160.0 Desmarestia viridis 0.0 0.0 0.0 50.0 Codium fragile 0.0 319.4 80.0 26.8 Chordaria flagelliformis 8.3 67.8 0.0 0.0 Chondrus crispus 2267.2 1098.9 165.6 100.0 Corallina officinalis 325.6 684.5 44.2 55.0 Ahnfeltia plicata 0.0 42.8 4.9 31.4 Cystoclonium purpureum 7.2 185.6 13.0 34.3 Rhodomela confervoides 15.6 52.8 51.2 3.8 Phyllophora pseudoceranoides 0.0 1.7 162.5 237.5 Phycodrys rubens 0.0 0.0 0.0 0.2 Poyides caprinus 0.0 0.0 48.1 11.2 Dictyosiphon foeniculaceus 0.0 56.1 0.0 0.0 Ceramium rubrum 34.4 9.5 7.5 20.9 2011 W.H. Adey and L.-A.C. Hayek 125 Species Infra 0.5 m 2.5 m 5 m 10 m 20 m Neosiphonia harveyii 18.0 80.0 24.3 8.8 Polysiphonia spp. 0.0 0.0 13.3 49.3 Dumontia incrassata 0.0 0.0 0.0 0.1 Palmaria palmata 0.0 0.0 0.4 2.2 Antithamnion spp. 0.0 87.2 14.0 8.4 Ulva lactuca 0.6 8.9 0.0 0.0 Pilayella littoralis 0.8 11.1 0.0 0.0 Cladophora spp. 38.4 1.7 0.0 0.0 Chaetomrpha spp 22.4 2.8 49.6 35.2 Monostroma spp. 17.2 8.9 0.0 0.0 Euthora cristata 0.0 0.0 0.1 0.0 Total biomass 3942.4 6589.1 3593.5 4401.3 Total biomass sum 18,526.0 126 Northeastern Naturalist Vol. 18, Monograph No. 8 Appendix E (1). Algal species cited. Nomenclature follows Algaebase.org (18 February 2011). C = Chlorophyta (green algae), P = Phaeophyleae (brown algae), and R = Rhodophyta (red algae). Species Phylum division Acrosiphonia arcta (Dillwyn) Gain C Acrothrix (nova-angliae) gracilis Kylin P Agarum clathratum Dumortier P Ahnfeltia plicata (Hudson) E.M. Fries R Alaria esculenta (L.) Greville P Ascophyllum nodosum (L.) LeJolis P Bonnemaisonia hamifera Hariot R Brongniartella byssoides (Goodenough & Woodward) F. Schmitz R Bryopsis plumosa (Hudson) C. Agardh C Callithamnion tetragonum (Withering) S.F. Gray R Callophyllis laciniata (Hudson) Kutzing R Ceramium circinnatum (Kutzing) J. Agardh R Ceramium rubrum (Hudson) C. Agardh R Chondrus crispus Stackhouse R Chorda filum (L.) Stackhouse P Chordaria flagelliformis (O.F.Muller) C.Agardh P Cladostephus spongiosus (Hudson) C. Agardh P Clathromorphum circumscriptum (Kjellman) Foslie R Clathromorphum compactum (Kjellman) Foslie R Codium fragile (Suringar) Hariot C Corallina officinalis L. R Cryptopleura ramosa (Hudson) L. Newton R Cystoclonium purpureum (Hudson) Batters R Desmarestia aculeata (Linnaeus) J.V. Lamouroux P Desmarestia ligulata (Stackhouse) J.V. Lamouroux P Desmarestia viridis (O.F. Muller) J.V. Lamouroux P Devaleraea ramentacea (L.) Guiry R Dictyosiphon foeniculaceus (Hudson) Greville P Dilsea carnosa (Schmidel) Kuntze R Dumontia incrassata (O.F. Muller) J.V. Lamouroux R Elachista fucicola (Velley) Areschoug P Erythrodermis traillii (Holmes ex Batters) Guiry and Garbury R Erythroglossum laciniatum (Lightfoot) Maggs and Hommersand R Euthora cristata (C. Agardh) J. Agardh R Fimbriofolium (Rhodophyllis) dichotomum (Lepechin) G.I. Hansen R Fucus distichus L. P Fucus serratus L. P Fucus vesiculosus L. P Halarachnion ligulatum (Woodward) Kutzing R Himanthalia elongata (L.) S.F. Gray P Laminaria digitata (Hudson) J.V. Lamouroux P Laminaria hyperborea (Gunnerus) Foslie P Laminaria ochroleuca Bachelot de la Pylaie P Leathesia (marina) difformis Areschoug P Lithothamnion lemoineae Adey R 2011 W.H. Adey and L.-A.C. Hayek 127 Lithothamnion glaciale Kjellman R Lithothamnion sonderi Hauck R Lomentaria orcadensis (Harvey) F.S. Collins R Lomentaria clavellosa (Turner) Gaillon R Mastocarpus stellatus (Stackhouse) Guiry R Membranoptera alata (Hudson) Stackhouse R Membranoptera denticulata Kuntze R Neodilsea integra (Kjellman) A. D. Zinova R Neosiphonia harveyii (J.W. Bailey) M.-S. Kim, G Choi, Guiry & R G.W. Saunders Palmaria palmata (L.) Stackhouse R Pantoneura fabriciana (Lyngbye) M. J. Wynne R Petalonia fascia (O.F. Muller) Kuntze P Phycodrys riggii N.L. Gardner R Phycodrys rubens (L.) Batters R Phyllophora pseudoceranoides (S.G. Gmelin) Newroth & A.R.A. Taylor R Phymatolithon (polymorphum) purpureum (P.L. Crouan & H.M. Crouan) R Woelkerling & L.M. Irvine Phymatolithon (rugulosum) lamii (M. Lemoine) Y.M. Chamberlain R Phymatolithon laevigatum (Foslie) Foslie R Pilayella littoralis (L.) Kjellman P Polyides (rotundus) caprinus (Gunnerus) Papenfuss R Polysiphonia flexicaulis (Harvey) F. S. Collins R Polyisiphonia urceolata (Lightfoot ex Dillwyn) Greville R Porphyra purpurea (Roth) C. Agardh R Ptilota serrata Kutzing R Rhodomela confervoides (Hudson) P. C. Silva R Saccharina latissima (L.) C. E. Lane, C. Mayes, Druehl & G. W. Saunders P Saccharina longicruris (Bachelot de la Pylaie) Kuntze P Saccorhiza dermatodea (Bachelot de la Pylaie) Areschoug P Saccorhiza polyschides (Lightfoot) Batters P Scagelia pylaisaei (Montague) M.J. Wynne R Scytosiphon lomentaria (Lyngbe) Link P Stictyosiphon tortilis (Gobi) Reinke P Stilophora rhizoides (C. Agardh) J. Agardh P Turnerella pennyi (Harvey) Schmitz R Ulva lactuca L. C 128 Northeastern Naturalist Vol. 18, Monograph No. 8 Appendix E (2). Animal species cited. Invertebrates Mollusca Mytilis edulus (L.) (Blue Mussel) Modiolus modiolus (L.) (Horse Mussel) Arthropoda Cancer borealis Stimpson (Jonah Crab) Cancer irroratus Say (Atlantic Rock Crab) Carcinus maenus (L.) (Green Crab) Hyas araneus (L.) (Great Spider crab, Toad Crab) Semibalanus balanoides (L.) (Northern Rock Barnacle) Echinoderms Asterias vulgaris Verril (Northern Sea Star) Crossaster papposus (L.) (Spiny Sunstar) Leptasterias polaris (Muller and Troschel) (Polar Sea Star) Solaster endeca (L.) (Purple Sun Star) Strongylocentrotus droebachiensis (Muller) (Green Sea Urchin) Vertebrates Finfish Gadus morhua L. (Atlantic Cod) Anarhicas lupus L. (Wolffish) Mammal Enhydra lutris (L.) (Sea Otter)