Microhabitat diversity of Svalbard Bryozoa

Bryozoans are one of the major macrofaunal groups of the high polar regions. Here we present data on the nature of bryozoan assemblages in the Svalbard Archipelago sampled over 6 years between 1997 and 2002. Samples were collected with use of Van Veen grab, dredge and Scuba diving at depths ranging from 0 to 268 m. Among examined material (5026 items) bryozoans were present on one type of abiotic (stones) and 40 types of biotic substrata. The biotic substrata we investigated included algae (12 taxa) and invertebrates (28 taxa: Hydrozoa, Gastropoda, Bivalvia, Polychaeta, Crustacea, Ascidiacea, other Bryozoa). Both a priori ANOSIM (Global R = 0.512, P = 0.001) and cluster analyses reveal differences between the assemblages on different types of substrata. Cluster analyses distinguish two distinct groups of substrata. The first (stable) included stones, molluscs and Balanus balanus whilst the second (flexible) consisted of algae, hydrozoans and bryozoans. Bryozoan assemblages which colonized different substrata, varied in terms of species composition and their dominance. Bryozoan species were categorized as generalists, low‐specificity epibiotic species or locally abundant background species. No species were found to be substratum‐specific to any type. Among substrata with the richest bryozoan fauna were stones (156 taxa), Chlamys islandica (68) and Balanus balanus (62). There were clear patterns in the distribution of substrata colonized by bryozoans within the fjord system: those shallow near the mouth were rich whilst those in deeper water were depauperate. We consider environmental gradients such as siltation and processes related to depth to be responsible for such patterns.


Introduction
Bryozoans are one of the few phyla in which virtually all representatives are primary or obligate sessile organisms. This means that they are directly dependent upon a substratum resource for at least the most significant stages of their life history. As with many benthic taxa, selection of the habitat is not a random process but rather the result of behavioural pattern (Hayward 1980). Thus the distribution of adults can largely be the result of choice exhibited by the prospecting larvae (Zimmer and Woollacott 1977). There are a number of factors influencing the selection of substrata by marine invertebrates. The most important seem to be: response to gravity, light, current, presence of surface film, and rugosity of substratum (Ward and Thorpe 1989).
The phenomenon of substratum preference is not only observed among the bryozoans. It is also a well-recognized behaviour among barnacles, brachiopods, polyplacophoran and gastropod molluscs, and tubeworms (Spirorbis spp.) (e.g. Pechenik and Eyster 1989). Some bryozoans benefit from host substrata in more ways than just as an attachment surface. Membranipora membranacea (L.), for example, intakes organic carbon from the kelp on which it lives (Manriquez and Cancino 1996).
Bryozoan abundance and taxonomic diversity can be directly related to substratum nature (Hayward and Ryland 1998). One of the most studied biotic substrata used by bryozoans are algae (e.g. Różycki and Gruszczyń ski 1986;Manriguez and Cancino 1996;Lippert et al. 2001). Some species of algae are used more than others (Ryland 1959). The most likely factors explaining the pattern of settlement is texture of thallus, age of the fronds and amount of mucus (Ryland 1959). Some bryozoan species-substratum associations are very selective either as commensal or mutualistic relationships (Cadee and McKinney 1994;Ross and Newman 1996). The broad range of substrata occupied by bryozoans includes, for example, crabs , merostomata (Key et al. 1996) and isopod exoskeletons (Key and Barnes 1999), bryozoans (Barnes 1994), corals (Harmelin 1990), brachiopods (Barnes and Peck 1996), gastropods (Ryland 2001), and bivalves (Seed 1996).
In the Arctic little is known about microhabitat use by bryozoans other than that boulder fields can support rich and diverse assemblages (Dick and Ross 1988). Kluge's (1975) monograph is one of the most comprehensive on high arctic species but this merely indicates substrata on which given species were recorded. That Powell (1968) found 38 bryozoan species associated with scallop beds in arctic Canada shows that a substratum other than rock is also important there. Furthermore, in the Svalbard region, Różycki and Gruszczyń ski (1986) and Lippert et al. (2001) found that bryozoans were the most abundant epiphytes on nearshore algae.
Data about the other substrata occupied by bryozoans in the Arctic are lacking. Here we attempt a baseline study of the spectrum of substrata used by one of the most numerous components of nearshore Svalbard macrobenthos, the bryozoans. We attempt to describe the characteristics of bryozoan assemblages on the major different substrata used.
Svalbard is influenced mainly by two water masses. The first is the West Spitsbergen current, which is a branch of the warm (4uC) and highly saline (35 PSU) Norwegian Current. This current moves along the western and northern parts of Spitsbergen (Loeng 1991). Transportation of these warm water masses causes a milder climate compared to other areas at similar latitude (Gammelsrod and Rudels 1983). The second water mass influencing this area is the East Spitsbergen Current. This cold, dense and highly saline water originates in the Arctic Ocean and travels along the east coast of Spitsbergen down to the southern tip of the island at which it is termed the Sorkapp Current. The Sorkapp Current brings a watermass ranging in temperature from 21.5 to 1uC and salinity from 34 to 35 PSU to the western coast of Spitsbergen (Loeng 1991). As along most high polar shores, the Svalbard coast is subjected to ice scour from floating ice during the summer months. In winter the sea surface freezes on inner fjord areas forming fast ice. Fast ice occurs less often and less extensively in the central and outer parts of the fjords (Svendsen et al. 2002).
The main study site, Kongsfjorden, is 26 km long and on average 8 km wide. The whole area of the fjord is 208.8 km 2 . Its maximal depth is 428 m, average about 140 m. Of the total coastline length of the fjord (89.6 km) 15.9 km is glacier mouth. Typically important factors influencing hydrography of the fjord are bathymetry, vicinity of the ocean, glaciers and local climate. Kongsfjorden does not typically have a fjord sill at the entrance, this absence causes strong influence of ocean waters on the hydrological regime (Węsławski et al. 1991). The fjord does have a range of bottom types from the soft fine, muddy sediments through sand and gravel to hard rock (Hop et al. 2002). Svendsen et al. (2002) provide a detailed description of the physical and environmental characteristics of the fjord.

Field sampling and data analyses
Sampling was carried out during the 1997the , 1998the , 2000the cruises of R/V Oceania and 1999the , 2000 cruises of R/V Jan Mayen. Samples were collected from the intertidal to 268 m depth using three techniques: Van Veen grab, dredge and Scuba diving (hand picking and frame 0.25 m 2 ). Of the 213 samples, 93 were collected by Scuba diving, 112 by grab and eight by dredge. Collected material was dried in the case of rocks, stones, shells, and barnacles. Algae and ''bushy'' bryozoan substrata were fixed in 4% formalin or 70% ethanol. All substrata were identified to the highest taxonomic resolution possible.
Frequency of occurrence was calculated by use of the equation: where F is the frequency of given taxa (%), n i is the number of samples where given taxa were present and n is the number of all samples.
To compare faunal composition of bryozoan assemblages between different substrata, we used the PRIMER software package. We calculated Bray-Curtis similarity measures using frequency of occurrence square-root transformed data (Bray and Curtis 1957). Using this matrix, samples were then classified into groups by hierarchical agglomerative clustering using group-average linking. We illustrated this divisive classification of samples using a standard dendrogram. One-way analyses of similarities (ANOSIM; Clarke and Green 1988) and multiple pairwise comparisons were used to test a priori differences between assemblages on different substrata. ANOSIM uses the test statistic R, which is calculated using average rank similarities among pairs of replicates (in this case e.g. species of algae) within each of two groups (e.g. algae-stones) minus the average rank similarity of replicates between groups and is scaled to give a value between 21 and 1. Thus, R51 when all similarities within groups are less than any similarity between groups, R.0.75 when there is large difference with the groups either well separated, R.0.5 when overlapping but clearly different, R,0.25 when groups are barely separable at all and R50 when replicates within and between groups are equally similar. If R<21, then pairs consisting of one replicate from each group are more similar to each other than are pairs of replicates from the same group (Clarke and Gorley 2001).

Results
From 213 samples, 5026 potential substrata for bryozoans were examined. On 3547 of them bryozoans were present. Those used consisted of loose rock and 40 biotic types. Twelve taxa of algae acted as hosts (Rhodophyta, Phaeophyta) along with 28 animal taxa [Annelida (Polychaeta), Bryozoa (Gymnolaemata), Chordata (Ascidiacea), Cnidaria (Hydrozoa), Crustacea (Cirripedia and Malacostraca), Mollusca (Monoplacophora, Gastropoda and Bivalvia)]. The bathymetric range in which most of the different microhabitats were found is shown in Figure 2. Accumulation curves for assemblages on many substrata (Figure 3) reached asymptote, suggesting that our sampling regime was sufficiently representative for the area.
In total 174 taxa of bryozoans were determined: 139 to species, 28 to genus, five to family, one to order and one to phylum level. The most frequently occupied substrata were  loose rock with 156 taxa of bryozoans recorded on them. On biotic substrata the number of taxa present varied from one to 68. They were most abundant on the bivalve Chlamys islandica (Moller) (68 taxa) and the cirriped Balanus balanus (L.) (62 taxa). Of the algae, Phycodrys rubens (L.) Batters (45 taxa) and Alaria esculenta (L.) Greville (29 taxa) hosted most.
Species frequency of occurrence of top 10 species in the main groups are presented in the Table I.
Despite some taxa occurring on various substrata, the assemblages were significantly different (ANOSIM global R50.512, with associated P50.001). Pairwise test for most of the groups of bryozoan assemblages are shown in Table II. Pairs which had significance levels (P) above 0.80 were not included (e.g. bryozoan-hydrozoan R520.36, P50.833) in the table. These data due to their high type I error would be meaningless for ecological interpretations. Cluster analyses generally confirm dissimilarities and similarities calculated by a priori pairwise ANOSIM R statistics.
Cluster analyses (Figure 4) visualized the differences between the substrata. The most similar assemblages were constituted on Balanus balanus and Chlamys islandica-95%. At the level of approximately 40% similarity we could distinguish two groups of substrata: one more stable, including stones, B. balanus and molluscs, the other more flexible, including algae, hydrozoans and bryozoans.
Following Barnes and Clarke (1995) we classified bryozoan colonizers of substrata into four types: generalist, host-specific, low specificity and background species (Table III). No host-specific species were found.
Rank abundance plots ( Figure 5) show that except on molluscs, shells and Balanus barnacles, assemblages were typically dominated by a few very abundant species. In most cases this differed between substrata and even between shallow and deep stones. The pioneer cheilostomes Harmeria scutulata (Busk), Hippothoa arctica Kluge, Tricellaria ternata (Ellis and Solander), and to a lesser extent Celleporella hyalina (L.), were the numerically dominant species.
Membranous, calcified and erect flexible morphological colony forms of bryozoans were present on all investigated types of substratum. Runner-like colonies were absent on hydrozoan and bryozoan substrata. Erect rigid forms were present only on more stable substrata types: stones, molluscs and Balanus (Table IV).
The dominant substrata used by bryozoans within the Kongsfjorden study site are shown in Figure 6. At the fjord mouth bryozoans mainly occur on stones but towards the middle fjord section this changes to algae. In the inner basin only bryozoans (Alcyonidium disciforme, Smitt) living on sediment were present. The least diverse microhabitats used by bryozoans were, therefore, at the two extremes of the fjord, the innermost and outermost sections. Diversity of microhabitat use decreased with depth, in many cases solely being stones. Bryozoans in the shallow area inhabited the most diverse type of substrata including rocks, algae, bryozoans, ascidians, hydrozoa, barnacles, molluscs, and crabs.
Bryozoan species playing important roles (dominants, most frequent ones) on certain microhabitats in Svalbard Archipelago are presented in Figure 7.

Discussion
Bryozoans from Svalbard waters are known to be species-rich (Gulliksen et al. 1999), our present study reveals that a rich diversity of substrata are also occupied by them. In a similar study in temperate waters Ward and Thorpe (1989) showed that the surface nature Table I. Frequency of bryozoan occurrence on seven major groups of substrates. Since assemblages on stones from the zone between 0 and 40 m depth differ very much from the stones from the 40+ m depth (similarity less than 40%; see the Figure 4) they were treated separately. (e.g. rough/smooth, convex/concave) rather than the type of substratum (e.g. rock or the shell of a particular mollusc species) was important for determining use by subtidal bryozoans. Our investigation supports this observation. The bryozoan community colonizing balanomorphs was similar to that on Chlamys islandica (see Figure 4). Although such substrata belong to different phyla (Crustacea, Mollusca), they are similar per se in shell surface structure. The same pattern was observed among algae and stones.  In contrast to Antarctic studies (Barnes and Clarke 1995), no host-specific bryozoan was found in this study (see Table III). This may result from a longer period of evolution and isolation of the Antarctic ecosystem in comparison to the Arctic (Dunton 1992).
Cluster analyses reveal two major groups of substrata: flexible (algae, hydrozoa, bryozoan) and stable (stones, shells, molluscs etc.). Glasby (2001) found large differences in abundance of given taxa on fixed versus moving artificial substrata. Previous studies of epiphyte abundance on seagrasses and macro-algae would support his results (Trautman and Borowitzka 1999). This is seen in the present study as well, as it shows differences in community structure between natural moving substrata (algae, bryozoans) and stable, nonmoving substrata (stones).
We suggest stony substratum per se is a much more stable substratum in comparison to all the flexible ones (algae, bryozoans, hydrozoans). There are many advantages to living on flexible substrata. Even if a given substratum is seasonal, usually its occurrence is predictable (e.g. algae). Stony polar substrata may be a focus of intense competition between encrusting organisms (Barnes and Kukliń ski 2003). Recruitment on flexible competitor-free substrata, therefore, is potentially beneficial. Manriquez and Cancino (1996) have also shown that bryozoans can absorb the exudates of algae. That might be especially advantageous in polar regions, where the food supply is highly seasonal. Settlement on algae in hydrodynamically favourable positions above a solid surface provides colonists with better flow and so higher nutrient supply and more efficient evacuation of waste (Wahl 1989).
Erect flexible and encrusting morphologies of bryozoans were present on all types of substrata, while erect rigid colonies only colonized stones, barnacles and molluscs. The solidity of the last three mentioned substrata is maybe why they are preferred by erect rigid morphological forms. Runner morphologies, in contrast, were absent on hydrozoans and bryozoans and dominated the morphological forms present on rocks below 40 m depth. Schäfer (1994) described these to be the primary macrobenthic colonizers of arctic rocks. Due to their primitive colony form (lowest integration within the colony), and fast, directional growth, they are very much adapted to inhabit unstable environments (e.g. small rocks, areas with high sedimentation rate) (McKinney and Jackson 1991; personal  Barnes and Clarke 1995) of some common Arctic bryozoans based on their occurrence as epifauna on biotic and abiotic substrata [in brackets: frequency of occurrence on the rocks (%)/ frequency of occurrence on the animals and algae (%)].

Bryozoan species type Exemplar species
Generalist Species which occur widely on rocks, stable hard substrata and animal, algal host Cribrilina annulata (15/11) Cylindroporella tubulosa (14/9) Harmeria scutulata (24/15) Host-specific species Species found only on specific animal host Not found Low-specificity epibiotic species Species which occur widely on organisms, algae but rarely on rocks Celleporella hyalina (3/38) Eucratea loricata (2/13) Tricellaria ternata (2/17) Background species Species which occur widely on rocks, but rarely on organisms, algae Electra arctica (21/2) Hippothoa arctica (22/1) Figure 5. Top 10 most abundant species on a given substratum type (in brackets, number of the species within the ''others'' group). observation). Their presence at these depths is probably indicative of disturbance in the area of occurrence, such as by seasonal input of fine sediment. Chemical and biological (e.g. avicularia presence) defence deployed by hydrozoans and bryozoans may explain the lack of runner morphologies colonizing them (Wahl 1989).
As we found across our Spitsbergen substrata, Lippert et al. (2001) found Celleporella hyalina dominated six algal species from the Kongsfjorden. This r-selected organism is abundant and widely distributed in the northern hemisphere (Hayward and Ryland 1999). The attributes of this species suit colonization of unstable substrata due to rapid patterns of growth and maturity (Seed and Hughes 1992).
The life history of Harmeria scutulata, which dominated shallow stones and bryozoan substrata, is largely unknown. Its lack of ovicells (peronal observation) and high abundance   Figure 5) would indicate an r-strategist. The habitat of this species is typically shallow waters repeatedly disturbed by ice-flows, fresh-water discharge and tides (Svendsen et al. 2002). Species with K-selected characters would seem unlikely to persist in such a disturbance regime. Harmeria scutulata certainly seems to be fast growing and able to colonize quickly. A strongly characteristic feature of the bryozoan communities we studied was the dominance of just a few species, despite the generally high levels of species richness ( Figure 5). This was especially true for stones (between 0 and 40 m depth, Harmeria scutulata 50% of all inhabiting species; stones below 40m depth, Hippothoa arctica 36%), algae (Celleporella hyalina 44%), hydrozoans (Tricellaria ternata 31%, Eucratea loricata 16%), and bryozoans (Harmeria scutulata 57%). Molluscs and barnacles were more evenly colonized by the bryozoan species, as has been found in both temperate and tropical equivalent habitats (Seed 1996).
Overall, all types of substrata in Svalbard waters are colonized by fast-growing bryozoans, which reach maturity very early (crucial for life-cycle closure) in the periodically foodlimited and often highly physically disturbed Arctic environments.
Two gradients of substratum distribution were distinguished within our fjordic study area ( Figure 6): the inner to the outer part of the fjord (geographic) versus that from the shallow to the deep part of the fjord (bathymetric). Such patterns are connected to clear environmental gradients (sedimentation, depth) along the fjord. The inner fjord has a much higher sedimentation rate, and a fine sediment bottom that is nevertheless more heterogeneous than the outer part (Svendsen et al. 2002). The deep outer region is mostly soft bottom with patchy stones or drop-stones occurrence where availability of substrata for bryozoans is very limited (Zaborska 2001). The shallow area was richer in substrata and sites potentially suitable for bryozoan colonization (Hop et al. 2002). The greatest difference between these two zones (shallow and deep), however, was the presence of algae in the shallow part, importantly adding three-dimensional structure to the system. We found algae to host different assemblages of Bryozoa than the other types of substrata (see Figure 4 and Table I).
Bryozoa are a very diverse group of macro-organisms (see Kluge 1975: 366 species) in comparison to the others in the Arctic (Gulliksen et al. 1999). One of their features which has made them so successful in that region is the ability to inhabit a wide range of substrata and to be highly successful as macrobenthic pioneer species, as identified here.