Report on Scientific Work
Deep-sea benthos of the Reykjanes Ridge: biogeographic analysis of the fauna living below 1000 m.
Report
on preliminary phase of the project
Patterns and processes of the ecosystems of the northern
Mid-Atlantic (MAR-ECO)
Alexander Mironov, Andrey Gebruk
Shirshov Institute of Oceanology
Russian Academy of Sciences
Contents.
Introduction
I. Database on the composition and distribution of deep-sea fauna (>1000m) along the
Reykjanes Ridge.
II. Compilation of the annotated bibliography on the biogeography of the north Atlantic.
1. Biogeographic boundaries.
1.1. Definitions and methods.
1.2. Vertical zonation.
1.3. Boundaries of the Eastern Atlantic Boreal Region, including the southern shelf of Iceland.
1.4. Boundaries of the bathyal and abyssal provinces of the northern Atlantic.
2. Biogeographical structure of the Reykjanes deep-sea fauna.
3. Biogeographical history of some deep-sea taxa .
Introduction
The Reykjanes Ridge is a 800 km segment of the Mid-Atlantic Ridge extending from Iceland southward to the Charlie Gibbs Fracture Zone, at 5230' N. The ridge axis shows a depth gradient from 0 m at 63N, the southern Icelandic coast, to about 3000 m. The deeper southern part of the Reykjanes Ridge is swept by Upper North Atlantic Deep Water (potential temperature 2 to 4C) and the shallower northern end is bathed in warmer Sub-Polar Mode Water (potential temperature 4 to 7C) (McCartney, 1992; Copley et al., 1996). A high temperature hydrotermal activity has been detected at 6306' N - the Steinaholl vent-field (Olaffson et al., 1991; German et al., 1994).
The Reykjanes Ridge, including the southern shelf of Iceland, represents a great interest for biogeographers. Benthic fauna of this area was a subject of sharp climatic, geomorphological and hydrological changes in Pliocene and Pleistocene. As a result, the area was invaded by species from very remote regions: north Pacific and south Atlantic. Present environmental conditions on the Ridge are also characterised by sharp physical and chemical gradients. For example, temperature anomalies on the geomorphologically uniform Iceland shelf are so pronounced that result in faunal differences on the level of biogeographical regions: the northern shelf is a part of the Arctic Region, whereas the southern shelf is a part of the Eastern Atlantic Boreal Region (Briggs. 1974, 1995). Within the relatively limited area (the Reykjanes Ridge and southern shelf of Iceland) the Arctic fauna is replaced by boreal, European fauna – by American, and the autochthonous deep-sea fauna is replaced by allochthonous one.
Disagreements over main biogeographical concepts are significant. "There is too little sound knowledge of biogeographical processes, reflected by disparity about what can be assumed or should be investigated, a lack of integrated models of the biogeographical system and little recognition of process levels within such a system" (Rosen, 1988). One of the major reasons of these disagreements is the lack in biogeography of clear methodological definitions of basic terms and following them. In addition biogeographical terminology in Russian publications has some peculiarities. For these reasons the methodological definitions of some biogeographical concepts and approaches are given below(section II.1.1).
I. Database on the composition and distribution of the deep-sea fauna (>1000 m) along the Reykjanes Ridge.
Many of marine groups have been considered in the series "The Zoology of Iceland" which consists of more than 80 parts (Thorson, 1941; Einarsson, 1948; Madsen, 1949; Saemundsson, 1949 and others).
Another major initiative in the area is the long-term BIOICE Program (Benthic Invertebrates of Icelandic Waters), initiated in 1991 (Tendal, 1998). Consequently, it is now possible to present a reasonably accurate assessment of Icelandic shelf biogeography. The composition of the deep-sea benthic fauna is well documented for the Iceland, Faroes, British Isles and Rockall Trough areas (Mortensen, 1927; Lieberkind, 1929; Einarsson, 1949; Gage et al., 1983, 1985; Harvey et al., 1988 and others). Compared to our knowledge of the fauna of these adjacent areas, little is know about the fauna of the deep (> 1000m) southern segment of the Reykjanes Ridge.
Deep-sea studies were performed north of 60N by the Danish Ingolf Expedition (Burton, 1928; Clark, 1923; Hansen, 1908, 1913; Heding, 1935, 1942; Jensen, 1912; Lieberkind, 1935; Meinert, 1899; Mortensen, 1903, 1907, 1933; Wesenberg-Lund, 1930, 1941 and others). Several deep-sea samples were also taken by the RV "Thor" (Mortensen, 1903, 1907). Under the PETROS program (PETRogenesis of Oblique Spreading) rock sampling took place on the Charles Darwin cruise CD80 (BRIDGE Cruise 10, September 1993) and included 169 dredge stations between 57N and 67N. Deep-sea biological samples (between 1000 and 2600 m) were recovered at 40 of these stations. 101 species were identified from 102 biological samples obtained between 225 and 2600 m. (Many of them were identified at the genus level only). The species belong mainly to Porifera, Cnidaria, Mollusca and Echinodermata (Copley et al., 1996). Several deep-sea samples were obtained on the Reykjanes Ridge north of 60N during BIOICE Program (Tendal, 1998). However, most of the PETROS and BIOICE samples of benthic invertebrates are still being processed, and publications with species identifications are very limited (Waren, 1996; Petersen, 2000, Ivanov and Scheltema, 2001).
Several Okean-grab samples of benthic fauna were taken during the 5th cruise of the vessels "Mikhail Lomonosov” along the 30W meridian (Table 1); two of them located on the Reykjanes Ridge, depths 1278 and 2836 m (Kuznetsov, 1960). An extensive collection of benthic fauna was obtained on the 4th cruise the R/V “Akademik Mstislav Keldysh"(1982) from the area between 5803'N and 5852'N, and between 2409'W and 3152'W. The depth of samples ranges from 1220 to 2950 m. 100 benthic samples were obtained at 92 stations using the Sygsby trawl (22 samples), the Okean-grab (0.25 m2) (66 samples), the geological dredge (5 samples) and the submersibles "Pisces 7" and "Pisces 11" (7 samples). The 28th cruise (1992) worked at 59N, depths from 667 to 1161 m (Cruise Report, 1991). Biological samples and video records were obtained on 6 dives of submersibles Mir-1 and Mir-2. On the 43th cruise (2000) a benthic sample was taken by submersible "Mir" at 4457'N, 2800'W, depth 2800m (Table 2).
The published results of “Akademik Mstislav Keldysh” and "Mikhail Lomonosov” expeditions focus on the following taxa: Madreporaria (Keller, 1985), Brachiopoda (Zezina, 1985, 2000), Polychaeta (Detinova,1985), Echinoidea (Mironov, 1985), Spongia (Tabachnik, in press), Asteroidea (Belyaev, Mironov, 1996) and Holothuroidea (Gerbruk, 1990). The total number of identified species is 134 (Table 3).
II. Compilation of the annotated bibliography on the biogeography of the north Atlantic.
1. Biogeographic boundaries
1.1. Definitions and methods
The words "biogeographic subdivision" denote a subject of very great breadth. Following Starobogatov (1982), biogeography is defined here as science on the distribution of any life manifestations. It studies a distribution of biological objects of any nature over the face of the earth. With such a wide interpretation, the delimiting of the branches of biogeographiy bases in the first place on delimiting of the types of biological objects: organism, population, community, taxon, biotic complex, adaptation, type of reproduction, feeding mechanism and etc. A problem of delimiting two main branches of biogeography (biotic and biocenotic geography or historical and ecological biogeography) has a long-time discussion (for more detail discussion see Beklemishev, 1969, 1982; Udvardy, 1969; Chernov, 1984; Myers and Giller, 1988; Rosen, 1988; Mironov, 1990, 1999; Razumovsky, 1999).
Biogeographical study is divided into two main stages: chorological and cause-and-effect analysis. In the former any biological object is regarded as a point, an area, or a three-dimensional space. This analysis estimates nonuniformity of object distribution and considers geometrical characteristics of areas and spaces occupied by biological objects: diversity of forms of areas and spaces, their overlap and coincidence. In the pure chorological (geometrical) analysis the environment is assumed to be homogenous. Violation of this assumption makes hardly possible correct cause-and-effect analysis of distribution patterns.
In the present study only the biotic geography will be considered. Taxa and biotic (faunistic + floristic) complexes are the objects of consideration in the biotic geography; populations and communities (= biocenoses) are the objects in the biocenotic geography. Mosaic patterns of species distribution are ignored in the biotic geography, but they are objects of detail study in the biocenotic geography. For purposes of a global-scale chorological analysis in the biotic geography, a species range is defined as a continuous area contoured by a single line, connecting the most outlying points of a species occurrence (Hesse, 1924; Beklemishev, 1969). From the chorological point of view, the specific nature of the biotic geography is based on the assumption of the species range continuity (for more details of this approach see Mironov, 1990, 1999).
Faunistic (biotic) boundary is defined as a zone where the boundaries of numerous species ranges come close together. Drude (1890) called these zones as “dividing”, Walret and Alehin (1936) as zones of “flora drop”, Kuznetsov (1936) as “synperates”, Schaffer (1956) as “floristic lines”, Turrill (1958) as “chorions”, Razumovsky (1999) as “boundaries of botanico-geographical regions”, Mironov (1985, 1990, 1999) as “zones of crowding”, Kafanov (1991) as “zone of increasing of species richness”. A biotic region (= biogeographical province) is a territory outlined by zones of crowding. A biotic (faunistic + floristic) complex is a group of organisms inhabiting a biogeographical province (table 1).
To reveal the crowding zones, the following four approaches are often used (Mironov, 1999).
The typology of species renges: species ranges are divided into groups based on the relative coincidence of species range boarders or their fragments (Alehin, 1944; Zezina, 1976; Hayden, Dolan, 1976; Semenov, 1982). Defining types of species ranges, Krylov and Semenov (1977) and Semenov (1982) use the approach of “classification differentiation level”, based on the probability of coincidence of species range maps estimated relative the size of the study area.
The method of biothic transects (= the coincidence-of- species ranges method): a study area is divided into equal latitudinal, longitudinal and (or) bathymetrical intervals; then for each interval a number of species range boarders is estimated and a graph of their distribution along the three axes is drawn (Vinogradova, 1958, 1962, 1977, ; Backus et al., 1965; Hayden, Dolan, 1976; Carney et al., 1983; Mironov, 1985, 1986; Gage, 1986; Gage and Tyler, 1991). Peaks of numbers mark biotic boarders. A species is regarded as present in the given interval even if it is found only in the adjacent intervals at both sides (a concept of species range continuity).
Applied to biothic boundaries between the ribbon-like "vertical" provinces, this method relies on the vertical range of species in order to measure the number of first and last occurrences. Backus et al. (1965) give further details of the method using a chi-squared statistic. One of the version of the method of biotic transects are the curves showing the cumulative appearance of new species plotted against the depth, latitude or longitude. The cumulative curves show the intervals with high rate of faunal change (Haedrich et al., 1980; Gage et al., 1985, 2000). Also they indicate the positions of the zones of crowding.
Method of the species richness: like in the previous method, a study area is divided into equal latitudinal, longitudinal and (or) bathymetric intervals; then the number of species per interval is estimated and the graph of their distribution along the three axes is drawn. The peaks or sharp changes in the number of species coincide usually with the peaks in number of species range boundaries (Vinogradova, 1958, 1962; 1977; Valentine, 1966; Mironov, 1986; Kafanov, 1991; Gage, Tyler, 1991; Roy et al., 1994; Roy and Martin, 2001). For example, a number of bivalves in the north-eastern Pacific decreases with increasing latitude. The strong species richness trend shows a stepwise pattern with the major declines in richness being concentrated at the provincial boundaries (Valentine, 1966; Roy et al., 1994). Various quantitative methods describing a use of species richness indexes in biogeography are considered in the monograph by Kafanov (1991).
Method of biotic similarity: determination of homogeneity or distinctiveness between sampling points or zonal intervals showed by the indices of faunistic similarity (Menzies et al., 1973; Haedrick et al., 1980; Ohta, 1983; Gage and Tyler, 1971; Parin et al., 1997). This method confuses biotic and biocenotic boundaries since it does no follow the major assumption of biotic biogeography - the continuity of a species range.
1.2. Vertical zonation.
Much of the work on the vertical distribution of benthic animals has been summarised by Menzies et al (1973), Vinogradova (1977), Carney et al. (1983), Gage and Tyler (1991). The disagreements over the supposed depth of biothic boundaries are significant. These disagreements are caused by: (i) real differences between the geographic regions in the depth of biotic boundaries; (ii) application of different methods; (iii) biocenotic and other boundaries are often confused with biotic ones. The boundaries of geomorphological structures, water masses, photosynthesis layer, sediment type, taxocenes, groups of similar communities have been used to show the vertical faunistic zones (Appelford, 1912; Bruun, 1956; Hedgpeth, 1957; Andryashev, 1977; Peres, 1982; Burukovsky, 1984; Golikov, 1985 and others).
The crowding zone at about 800 - 900 m (fig. 1) was revealed by the method of pooling data on the distribution of echinoids in the northern Atlantic, from 50 to 62 N and from 10 E to 70 W (Mironov, 1986). This result was supported later by Copley et al. (1996); the study of PETROS collection from the Reykjanes Ridge suggests two bathymetric faunal zones with transition between them at 800 to 1000 m, also found at 60 N. Hydrographic survey of the ridge axis suggests that this faunal zonation is influenced by the water mass structure (Copley, 1996).The transition between two water masses occurs at 800-1000 m: a warmer, more saline water covers the northern end of the ridge, with a cooler, less saline water at the southern end. The warmer water mass corresponds to the Sudpolar Mode Water and the cooler water mass - to the Upper North Atlantic Deep water (Schmitz, McCartney, 1993), with the salinity minimum showing the influence of the Labrador Sea Water (Read, Gould, 1992; Tsuchiya et al., 1992).
The similar faunal zonation was recognised by Le Danions (1948) in the Bay of Biscay, and by Menzies et al. (1973) on the continental slope of the eastern USA. According to Menzies et al. (1973), the Archibental Zone of Transition extends from 445 to 940 m in the north-western Atlantic, from Carolina to Georgia. The Archibental Zone is followed by the Upper Abyssal Zone, down to 2635 m. The neighbouring eastern areas differ somewhat from the Reykjanes Ridge in the position of the deep-sea boundaries. In the north-eastern Atlantic, from 20 to 50 N, the taxoniomic composition of echinoid fauna shows most change between 1000 and 1200 m (Mironov, 1986). At the same depth the benthic fauna of the Scotland slope changes conspicuously (Gage et al., 2000). Two peaks in the first and the last occurrences of species were reported for the Rockall Trough: at 800-1200 m and 1800 m (Gage, 1986). In the Norwegian Sea the conspicuous breaks of fauna are observed at the depths about 800 m and 1200 m (Krayushkina, 2000).
1.3. Boundaries of the Eastern Atlantic Boreal Region, including the southern shelf of Iceland.
According to Briggs (1974, 1995), the Eastern Atlantic Boreal Region extends from the English Channel northward to the Kola Fjord (to the base of the Murmansk Peninsula), and westward to the south Iceland. However, the opinions about subdivisions and limits of the region vary considerably. First distribution maps were published by Forbes (1856, 1859) and Woodward (1856). Their version indicated the Boreal Province extending from northern Norway to the Skagerrak, and the Geltic Province from the Skagerrak to the Saint Mathieu Cape (near the western entrance to the English Channel). Some authors (Coomans, 1962; Mars, 1963; Hall, 1964; Vasilenko, 1974; Kussakin, 1979; Golikov, 1980; Nesis, 1982) supported the concept of two provinces; others (Geptner, 1936; Anderson, 1942; Poll, 1947; Saemundsson, 1949; Bruun and Pfaff, 1950; Gurjanova, 1951;Hyman, 1955; LeGall and Cantacuzene, 1956; Ursin, 1960; Ryland, 1963; King, 1964; Briggs, 1974, 1995) did not recognized such subdivision.
Based on the level of endemism as a criteria of biothic boundary, the Eastern Atlantic Boreal Region can not be divided into separate parts. However, the Eastern Atlantic Boreal Region may be subdivided into several provinces or subprovinces, when the biothic boundary is considered as a crowding zone only. Thus, three subprovinces were distinguished by Brattegard and Holthe (1995, 1997) on the Norwegian shelf; five sublittoral-upper bathyal subprovinces were separated by Krayushkina (2000) in the Norwegian Sea.
On the map published by Forbes (1856) and Woodward (1856), the Boreal Province extends to both sides of the Atlantic. The American and European parts of the northern Atlantic do not separate in many biogeographical schemes (Geptner, 1936; Gurjanova, 1951; Vasilenko, 1974; Kussakin, 1979; Golikov, 1980). Most of the disagreement has emerged about the location of the northern and north-eastern limits of the region (see Briggs, 1974). Depending on authors attitude, location of the boarder changes from the southern Iceland (Vasilenko, 1974; Kussakin, 1979; Golikov, 1980; Briggs, 1995) to the northern Iceland (Geptner, 1936; Gurjanova, 1951; Lattin, 1967) and from the Lofoten Islands (Hall, 1964) to the Novaya Zemlja and Svalbard (Golikov, 1980). The shelves of the southern Iceland and the Azores are usually referred to different biogeographical subdivisions: Eastern Atlantic Boreal and Mediterranean-Atlantic Regions (Briggs, 1974,1995), Keltic and Mauritanian Provinces (Nesis, 1982), and other variants.
1.4. Boundaries of the bathyal and abyssal provinces of the north Atlantic
The biogeographical scheme for the bathyal zone (from 200 to 3000 m) of the Atlantic has not been proposed so far. Only the existence of separate provinces in single areas has been proposed. Mortensen (1907) distinguished three Atlantic deep-sea (below 100 fms) regions: the European, West African and East American. The European Atlantic deep-sea region comprises the northern Atlantic, to the east of a line from Denmark Strait to the Gibraltar Strait. It is separated from the Arctic abyssal region by ridges across the Denmark Strait, the Faroe Channel and between Iceland and the Faroe Islands. The West African deep-sea region comprises the tract from the Azores to St. Helena. The East American deep-sea region comprises the whole western half of the Atlantic, from the Davis Strait to at least off La Plata, and perhaps even farther southwards. Later Mortensen (1933) regarded the whole northern Atlantic as one region (the North Atlantic deep-sea region), extending from Davis Strait, southern Iceland and Faroes to Cape Hatteras, Azores, Madeira and Canaries. Mironov (1994; in press) suggested that the Azores together with the Atlantis-Great Meteor seamount area should be separated in a biogeographical (Azorian) province of the sublittoral-upper bathyal zone. Tunnicliffe et al. (1996) considered the hydrothermal fields of the Mid-Atlantic Ridge as a biogeographical bathyal province. The Lucky Strike hydrothermal field was separated by Van Dover et al. (1996) in another biogeographical bathyal province.