Patterns of production and sedimentation in the boreal and polar Northeast Atlantic PAUL WASSMANN. ROLF PEINERT and VICTOR SMETACEK Wassmann, P., Peinert, R. & Smetacek, V. 1991. Patterns of production and sedimentation in the boreal and polar Northeast Atlantic. Pp. 209-228 in Sakshaug. E., Hopkins, C. C. E. & 0ritsland. N . A. (eds.): Proceedings of the Pro Mare Symposium on Polar Marine Ecology, Trondheim, 12-16 May 1990. Polar Research 10(1). Pelagic systems are potentially capable of retaining and recycling all autochthonous organic material. although some losses due to sinking particles inevitably occur. Relating processes in the surface layers quantitatively to vertical particle flux is difficult because only a small percentage of the total production is lost annually via sinking in the open ocean. Further. only a few types of particles contribute to this flux and only a small proportion of these may actually reach greater depths. Measurements of vertical flux with sediment traps revealed seasonal and regional patterns also within the northwestern Atlantic and indicate imbalances betwccn particle formation and dcgradation. The classical pattern of spring bloom sedimentation followed by reduced loss ratcs has bccn found in \hclf and shallow water regions such as the Norwegian Coastal Current and fjords and is also cncountcrcd in the Barcnts Sea. I n the Norwegian Sea. however. the seasonal pattern appears different as thc seasonal maximum has been observed during late summcr/autumn. The physical environment determines nutrient aVdihbility and hence the particles potentially available for sedimentation. The relationship betwccn phyto- and zooplankton govcrns vertical flux seasonality. and zooplankters with different life cycles and feeding strategies further modify the principle patterns. Her- bivores with life-cycle strategies involving overwintering of large biomass and predictable seasonal appear- ance (copepods. cuphausiids) will have a different impact than opportunistic organisms with vcry Ion overwintering biomass, for example salps and ptcropods. The latter exhibit much greater interannual biomass variation and may thus contribute to interannual variability of the vertical flux. Shelf systems of similar latitude arc generally comparable w i t h rcspcct to their flux patterns and also share timilarities w i t h marginal ice zoncs. Open ocean systems as the Norwegian Sea. however, cxhibit different patterns which are similar to the subarctic Pacific. Puul Wassmann. N o r w g i a n College of Fishery Science. Uniiier.~itv of Trurnsc. P . 0. Box 3083 G u l e q , N-9001 Trornsg. N o r w a y ; Rolf Peitiert. Instirut f u r Meereskiiiide. Dusternbrookerweg 20. 0-2300 K i d I . Federal Republic of Germany; Victor Smetacek. Alfred- Wegener-lnstitut f u r Polar- und Meere.\ forschung. A m Handelshafen 12, D-2850 Bremerhaoen. Germany. Introduction Sediment trap deployments conducted in the seventies in localities as disparate as the deep Sargasso Sea (Deuser & Ross 1980), the Baltic Sea (Smetacek e t al. 1978) and the shallow waters of Kiel Bight (Smetacek 1980a) revealed a strik- ingly common feature. T h e annual sedimentation pattern was dominated by a strong pulse following the spring bloom with lower rates during the rest of the year. A n additional autumn maximum in Kiel Bight could be related t o the autumn blooms which occur there regularly but not in the other areas. This spring pulse was accepted as the para- digm case for waters experiencing a spring bloom. T h e combination of insolation, shallowing of the mixed layer d u e t o fresh-water run-off, melt- ing ice and/or solar warming and high con- centrations of winter-accumulated nutrients in the euphotic zone provide the basis for the spring phytoplankton bloom (Fig. 1). It starts normally in March in the boreal zone and a second bloom might take place in August/September when nutrients are mixed upwards from deeper layers. The spring phytoplankton bloom starts in April or later in the Arctic and since light reaches limiting values by September, only o n e bloom occurs. The low herbivore biomass at the end of the winter period and the resulting low grazing pressure results in a slow response of zooplankton to increasing food supply, particularly in the Arc- tic (Heinrich 1962; Smetacek et al. 1984; Peinert et al. 1989; Tande 1991 this volume). A s a result, extensive sedimentation of ungrazed phytoplank- ton and phytodetritus takes place during the sen- escent stage of the spring bloom. Despite the pulsed nature of vertical flux, 210 P . Wassmann, R . Peinert & V . Smetacek I . -- _-_-I .______ - - a BOREAL ltght 1 - - - - -\\\ -____-_-’ .______ - - a Slrathcat8on I- I and Numen1 concentral8on 8n the eupholc zone (- - - - - I - _ _ _ _ _ _ _ - regeneraled pinduction regenefaled production ~ _ _ _ - - - - - - - - - . - - - new prodvclbon new production Phytoplank ton Zooplankton annual integrations of sedimentation acquired from early t r a p deployments yielded a linear-log relationship with annual primary production in the overlying water column (Suess 1980). Recently, Wassmann (1990) has shown that such a simple model can be applied t o boreal coastal waters of the North Atlantic. In the early eighties, it was widely accepted that the sedimentation rate in a given region could be ascertained from knowledge of primary pro- duction: the higher the production, the greater the proportion settling o u t (Berger et al. 1989). The concept of primary production, however, underwent a differentiation since we now dis- tinguish “new” from “regenerated” production (Eppley & Peterson 1979) and the “new” pro- duction sets the upper limit t o particle flux out of agiven water column (Platt et al. 1988). It seemed reasonable t o assume that major flux pulses would , follow periods dominated by new production, i.e. the spring bloom (Fig. 1). Attention next focused on the processes that control the vertical flux of particles from the mixed layer (Cadee 1985; Alldredge & Silver 1988; Lampitt et al. 1990). During the eighties, sediment traps were deployed in a variety of localities. Results from these studies indicate that annual patterns in dif- ferent regions can deviate substantially from the Fig. 1 . Conceptual diagram of seasonal changes in physical factors, phytoplankton stocks, new and regenerated production and sedimentation for boreal and Arctic areas. they merely demonstrate its restricted appli- cability. Indeed, implications arising from an assessment of the recent data o n vertical flux are already overturning another, older paradigm: the implicitly accepted patterns of the annual cycles of phytoplankton and zooplankton as exemplified by the curves of Heinrich (1962) which represent different regions of the world ocean. W e now know that these might well be basic types dic- tated by the physical environment, but in fact the variability is much greater than previously envisioned. Sediment traps a r e powerful tools in furthering o u r understanding of the dynamics of pelagic systems (GOFS 1989). They not only collect the material settling o u t of the upper layer but also demonstrate which types of suspended particles d o not sink o u t , or d o so only rarely. Inter- regional and inter-seasonal comparisons ( H o n j o 1990) provide a very useful framework to inter- pret, in ecologically meaningful terms, the per- plexing data now being derived from sediment trap deployments. Complications of primary production rates in the world ocean (e.g. Koblentz-Mishke e t al. 1970) as well as satellite-derived phytoplankton pigment distribution in the North Atlantic (Brewer et al. 1986) indicate that the Northeast spring maximum paradigm that emerged during 1 Atlantic is highly productive. This area is charac- the seventies. Considerable interannual variation terised by vast shelf areas including t h e North Sea in rates and even patterns recorded from the same and the Barents Sea that a r e open t o exchange region have also come to light (Deuser 1988). T h e with ocean basins (Figs. 2 and 3). In contrast to new data d o not disprove the earlier paradigm. the circular flow of the North Pacific gyre, warm Production and sedimentation in the Northeast Atlantic 21 1 R O O Fig. 2. Topography of the boreal and polar Northeast Atlantic (200 and 1Mx) m contours) and sediment trap mooring sites referred to in the text. LI = Lindlspollene; F A = Fanafjorden; V P = Vdring Plateau; NCC = Norwegian Coastal Current; NA-I = Aegir Ridge-1; NB-1 = Jan Mayen; LB 1 Lofoten Basin; B A = Balsfjorden; BI-1 = Bjdrneya; BS = Barents Sea sampling area (thin broken line); FS-I = Fram Strait; GB-2 = Greenland Basin. Also shown is the sea ice cover: average minimum and maximum in May (----) and August (-) (Redrawn from Wadhams 1986). 30' 20" Atlantic Water is transported northward along the eastern side beyond 80"N, whereas both cold Arctic Surface Water flowing along the western side and deep bottom water are exported south- wards into lower latitudes (Fig. 3). While the North Pacific could be characterised as a more closed and circular system, the North Atlantic is an open, asymmetric system with horizontal advection being of great importance for struc- turing pelagic systems and, hence, also their sedi- mentation patterns. The depth of winter mixing varies regionally with maxima of several thou- sands of metres south off Iceland and in the central Norwegian Sea where North Atlantic Deep and Bottom waters are formed. Further, there is a large latitudinal gradient of incident light and the Nordic Seas are characterised by a seasonally variable sea-ice cover i n the west, north and northeast (Fig. 2 ) as well as by extensive fronts between Arctic Water and Atlantic Water (Fig. 3). The regional pattern of the timing a. 'biological spring" depicted in Fig. 4 averaged over a 10- year period has been compiled by Pavshtiks & Timokina (1972). The advent of biological spring is triggered by stabilisation of a shallow surface layer which arises earliest in the southern realm. particularly marked along the Norwegian shelf. However, the occurrence of an early spring off the Greenland shelf and across the Barents Sea is related to melting at the ice edge. Further north and in the open Greenland Sea, spring commences as late 2s July/August. The unstable water column northeast of Iceland is noteworthy. One would expect the timing of spring dev- elopment to have a major influence on sedi- mentation patterns. The actual situation, however, is much more complicated. We here discuss vertical flux data from the boreal and polar regions of the Northeast Atlantic with the aim to identify different factors con- trolling the vertical flux. 212 P . Wassmann. R . Peinert & V . Smetacek Fig. 3. Surface currents and fronts in the Northeast Atlantic: I = Norwegian Coastal Current: 2 = Norwegian Atlantic Current; 3 = North Capc Current: 4 = Murmansk Current: 5 = Kanin Current: 6 = East Spitsbergen Current: 7 = West Spitsbergen Current; 8 = East Greenland Current: 9 = East Icelandic Current: 10 = Irminger Current: 1 1 = Jan Mayen Current. Fronts in the Northeast Atlantic: 12 = Norwegian Coastal Front; 13 = Barents Sea Polar Front; 14 = Polar Ocean Front: 15 = East Greenland Front and Iceland Gap Front. Shaded areas = mixing of Arctic and Atlantic water. Case studies Northeast Atlantic shelf North Sea. - Vertical flux studies were carried out in the Fladen Ground area of the northern North Sea during the springs and early summers of 1976 (FLEX 76) and 1983 ( R E F L E X 83). Striking differences were recorded between these years. During 1976 a spring bloom comprising small diatoms developed and a strong sedimentation pulse consisting of diatom cells and phytodetritus followed (Davis & Payne 1984). A substantial portion of organic matter produced by this bloom settled o u t . I n 1983 an unusual. small. colonial prymnesiophyte (Corymbellus aureus) dominated the spring biomass while diatoms attained only secondary importance (Gieskes & Kraay 1986). No sedimentation pulse was recorded this year (Cadee 1986) and organic material produced by the bloom was apparently retained in the water column. h'orweginn fiords. - T h e comp..cated topography of the Norwegian coastal zone, which ranges from 58 to 71"N. gives rise to a large variety of fjord, trough and shelf systems. Land-locked fjords and open fjords are characterised by different threshold depths at the entrance and different average depths. Open fjords and shelf systems are closely connected and house the same plankton organisms. T h e ecology of land-locked fjords, however, differs significantly d u e t o restricted water exchange, the exclusion of various types of plankton and the stagnation of bottom water. A "typical" fjord thus does not exist. Seasonal sedimentation trap deployments i n fjords along the Norwegian coast revealed dif- ferent flux patterns. i n the unpolluted land-locked fjord Lindaspollene, for example, major losses were recorded in April and early May with much lower fluxes during the rest of the year (Fig. 5A) (Wassmann 1983; Skjoldal & Wassmann 1986). In open Fanafjorden in contrast. only little organic Fig. 4. Timing of “biological spring”. as defined as the beginning of the development of phytoplankton and concentration of C. finmarchicus. Euphausiacea. etc. near the surface (Pavshtiks & Timokhina 1972) in the Northeast Atlantic. Redrawn from Pavshtiks & Timokhina (1972) with additional information by U . Stefinsson (pers. comm.) and K . Tande (pers. comm.). Production and sedimentation i n the Northeast Atlantic 2 1 3 ROO matter from the diatom spring bloom which occurred in April settled out of the mixed layer within the fjord. T h e bulk of the bloom biomass was probably exported onto the adjacent shelf prior to sedimentation as a result of a wind induced up-welling event. A second phyto- plankton bloom, sustained by new nutrients in the newly up-welled water, however, settled out in the fjord itself in May/June. Another, minor increase in sedimentation was measured in autumn when water column stability decreased following strong winds (Fig. 5A). O p e n fjords could, thus, b e characterised as multi-pulse sys- tems (Lindahl 1987; Wassmann 1991), whereas land-locked fjords with threshold depths lower than the seasonal pycnocline would be single- pulse ones. Also in the open, subarctic Balsfjorden (about 70”N) a seasonality of sedimentation is evident from late winter t o early summer (Lutter et al. JUL-AUG 1989) (Fig. 5B). T h e spring maximum following the phytoplankton bloom, however, was much less pronounced compared to Fanafjorden. At greater depth seasonal variation was also less significant. This possibly reflects a spring bloom which starts in weakly-stratified waters (Eilertsen et al. 1981) a n d , a t the same time, is influenced by the grazing by over-wintering herbivorous zooplankters (Hopkins e t al. 1985). The Norwegian shelf. - T h e few measurements available from the Norwegian shelf (Fig. 5B) indi- cate a spring maximum of vertical flux and low values during summer, i.e. a pattern comparable to Balsfjorden and the Barents Sea. Spring trap collections from 1983 (Peinert 1986b) contained relatively little intact phytoplankton, indicating that no major phytoplankton sedimentation pulse had been encountered. Summer sedimentation rates were lower, less variable and comprised 214 P . Wassmann. R. Peinert & V . Smetacek Llndaspollene l 4 0 m ) A 600, D Fanallotden 0 Norwegian Costal Current t 0 A-A' a c; Vsring Plateau (500rn) 0 Months Fig 5 Sedimentation rates of particulate organic carbon (mg C m ' d - ' ) in land-locked fjords. open fjords. the Norwegian Coastal Current. the Barenrs Sea and the Vsring Plateau Data replotted from Wassmann (1983. 1984 1989). Peinert (1986b). Skjoldal & Wassmann (19%). Lutwr et al (1989). Bathmdnn et al (1990a) and Wassmann et al (19%)) much less freshly produced matter of phy- toplankton origin. T h e bulk of the t r a p collections during this season consisted of krill faeces, some- times loaded with tintinnid loricae. The horizontal spring phytoplankton distri- bution in t h i s region is very patchy and coupled to the complex hydrography of the Norwegian Coastal Current (Braarud et al. 1958; Same & Mork 1981). Frontal processes between the Norwegian Atlantic Current and Norwegian Coastal Current. topographically induced gyres on the shelf (Sundby 1984). as well as lateral injections of low salinity water from fjords pro- vide variable conditions for phytoplankton growth. Differences in water column stability, up/ down-welling, nutrient availability at the onset of the growth season and nutrient replenishment by entrainment influence the immediate growth environment. Differences in the initial phy- toplankton seeding populations and advection of grazers of different type may modify conditions and hence further contribute t o space/time-dif- ferences in the spring bloom. Vertical fluxes, accordingly, must b e expected to exhibit a n even greater variability in space/ time during spring as highly non-linear factors influence the direct sedimentation of phyto- plankton (nutrient exhaustion triggering mass sedimentation, differing settling behavior of phy- toplankters). During spring, maximum vertical flux values may well exceed those depicted in Fig. 5B (Peinert 1986a), but it may not b e easy t o document such events as they would b e short- term and spatially restricted. In general, however, a spring sedimentation pulse of freshly produced autotrophic biomass, based o n and limited by winter nutrients can b e expected regularly o n the Norwegian shelf. T h e shelf, because it is shallow, is not a favorable site for the overwintering of large populations of herbivores. T h u s , it has to be colonised by annually dominant grazers during each spring. T h e resultant time lag between spring phytoplankton growth and the onset of intense grazing pressure (mainly by copepods) primarily controls spring sedimentation rates. B a r e n f s Sea. - T h e amount and seasonality of sedimentation in t h e Barents Sea is comparable to those of coastal areas of the boreal zone (Fig. 5B). Highest sedimentation rates of u p t o 1 g C m-* d-' were recorded during a dense Phaeocystb pouchetii bloom in areas dominated by Atlantic Water during May/June (Wassmann e t al. 1990). The average sedimentation rate below the euphotic zone ( 5 0 m ) during spring was about 300 mg C m-? d-' (P. Wassmann unpubl. data) with significant variation in different areas of the Barents Sea. During s u m m e r , sedimentation rates were much lower and averaged 80 mg C m-* d - ' below the euphotic zone (Wassmann 1989). T h e patterns of phytoplankton production and sedimentation in the Barents Sea are strongly influenced by the seasonal and interannual dynamics of sea ice (Rey & Loeng 1985). While the southwestern part is always ice-free, the central. eastern and southeastern areas a r e cov- ered by sea ice during spring (Fig. 2). T h e mar- Production and sedimentation in the Northeast Atlantic 215 78-40" 09 Time (monlhs) Fig. 6. Results from a mathematical model by Wassmann & Slagstad (unpubl. results): Seasonal dynamics of primary pro- duction (-.-.-.-), sedimentation (-) (g C m-* d - ' ) and suspended phytoplankton biomass (-----) ( g C m-*) along a transect in the central Barents Sea. ginal ice zone moves to the north during summer with the greatest retreat in the eastern Barents Sea. Interannual differences, however, are significant (Wadhams 1986). The spring phy- toplankton bloom in the southern Barents Sea dominated by warm Atlantic Water takes place in May/June (Fig. 6C) and is dominated by diatoms, for example Chaetoceros socialis. and by P . pouchetii. Phytoplankton growth starts already in late March, but stabilisation of a shallow sur- face layer by solar insolation is slow in this area and hence phytoplankton biomass increases cor- respondingly slowly. A strong pulse of sedi- mentation during spring is usually encountered as zooplankton grazing pressure by the major herbivore Calanus finmarchicus is light. Grazing by this copepod, which also dominates in the Atlantic Water of the Norwegian Sea, was shown to counteract sedimentation (Peinert et al. 1987). In the southern part of the Barents Sea, zooplank- ton can expect a rather predictable development of the spring bloom during May/June (Fig. 6C). To what extent the bloom is influenced by grazing, however, depends both on the size of over- wintering populations, especially C . finmarchicus, as well as on the advection of its copepodites from the Norwegian Sea into the southern Barents Sea (see below). The southern Barents Sea is a one- pulse system since no additional autumn bloom takes place due to light limitation. In the marginal ice zone further north and east the situation is different (Fig. 6A, B). If the sea- ice conditions allow for light penetration, the phytoplankton spring bloom starts earlier and develops more rapidly than in the southern Rar- ents Sea. This is due to the rapid increase in stability caused by melting sea ice. Again, a strong pulse of sedimenting matter follows the spring bloom after nutrient limitation. If, however, win- ter nutrients are left at the end of the spring bloom due to a deterioration of the light conditions (caused by a new advance of the marginal ice zone), another, yet smaller bloom may take place later (Fig. 6B). The rapid development of the spring bloom in the marginal ice zone of the Barents Sea typically decouples phytoplankton development from zooplankton grazing (Eilert- sen et al. 1989). This effect is most pronounced in the ice-covered North where a minor, light- limited spring bloom may take place as late as in August/September during some years, followed presumably by a correspondingly small pulse in sedimentation. In other years, no bloom takes place at all when sea ice covers the northernmost areas until insolation reaches limiting values i n September. Annual primary production de- creases, thus, with increasing latitude, and food supply is more unpredictable in the northern as compared to the southern Barents Sea (Rey et al. 1987). Plant biomass may not be available every year, depending on the ice situation. The major zooplankton species in the areas dominated by Arctic Water are C . glacialis and Calanus hyperboreus, and these species seem well adapted to an unpredictable food supply by a life-cycle of more than one year (Tande 1991). This makes them less vulnerable to seasonal and interannual variability in food supply. These copepods are much larger than C . finmarchicus and accordingly produce larger faecal pellets for which higher sinking velocities have to be assumed. It is thus not yet clear if their grazing on spring phytoplankton counterworks sedimentation to the same extent as that of C . finmarchicus. Capelin (Mallotus oillosus) is a commercially 216 P. Wosstmtiti, K. Peiriert Rr c'. Stnetocek important fish that is also a key species in the southern and central Barents Sea. After spawning close to the coast of northern Norway and the Kola Peninsula. the capelin stocks migrate i n the form of a 'front' northwards into the Barents S e a . Migrating capelin decimate zooplankton stocks in the southern and central Barents Sea. leaving behind only sparse numbers in the "wake". Con- sidering the efficiency with which capelin deplete zooplankton stocks over a bast area (Skjoldal & Rey 1989). their presence is likely to affect sedimentation patterns. The capelin stocks in the Barents Sea have been strongly reduced since 1987 as a result of fishing and increased predation by cod. Presumably zooplankton at present exerts greater control on the vertical flux than before. as grazing pressure and hence retention of essential elements during summer in the upper layers will have increased due to decreased sedimentation. The present situation in the Barents Sea might be shifted towards that encountered in the Norweg- ian Sea after the break-down of the herring stock in the sixties (see below). The shallow areas of the Barents Sea differ from the deeper regions in as much as pelagic herbivores have to share food resources with epi- benthic macrofauna. This holds true for the waters especially around Bjornoya and Spitsber- gen as well as in the eastern and southeastern part of the Barents Sea. Locally. some of the world's largest benthic biomass concentrations a r e found here (Zenkevitch 1963) exemplifying the tight pelago-benthic coupling characteristic of shallow, Arctic environments (Grebmeier et al. 1988: Hopner-Petersen 1989). Loss rates of phvto- plankton by sedimentation in these areas may be of minor importance when compared t o active removal by benthic suspension feeders. The biodeposition of pelagic suspended matter by the scallop Chlaniys islaridica. commonly found in the shallow parts of the Barents Sea. can locally exceed sedimentation by an order of magnitude (Vahl 1980). Further, the zooplankton of the banks south of Spitsbergen can be dominated by cirripede larvae during May/June ( F . Rey pers. comm.) which indicates the important role of this benthic suspension feeder on the plankton of the area. Ope11 h'orwegiari Sen The surface currents of the Norwegian Sea are strongly influenced by the warm saline water of the Norwegiar, Atlantic Current whose branches cover a large area of the central northeast North Atlantic. In the southwest an additional. Arctic Water source is supplied by the East Icelandic Current (Fig. 3). O n the Norwegian coastal shelf, \vater from the Norwegian Coastal Current con- tributes to the northward flow. Periods with tem- peratures and salinities well below the long-term mean occurred i n the East Icelandic Current in the late 1960s and were recorded in the 1970s thereafter in the Atlantic inflow to the Norwegian Sea (Dickson et al. 197.5; G a m m e l s r ~ d & Holm 1984). A down-stream delay of about three years between the Rockall Channel in the southwest and the entrance t o the Barents Sea in the north- east of the Norwegian Sea was recorded as well, suggesting that this event was advective (Dickson 6: Blindheim 1981). This climatic anomaly rep- resents one of the most persistent and extensive variations in ocean c!imate yet observed in this century (Dickson e t al. 1988) and influences the plnnkton dynamics in the Norwegian Sea (Blindheim 1989). The winter convection in the Norwegian Sea generally extends to as d e e p as 3 0 0 m and seasonally replenishes near-surface nutrient reserves. As the vernal stratification of the surface layers is almost entirely dependent on solar radi- ation. the spring phytoplankton bloom is delayed by one to two months as compared t o the con- tinuously stratified coastal waters (Halldal 1953; Rey 1981). T h e timing of stabilisation. however, varies between years, depending on meteoro- logical conditions. The Norwegian Sea is a transition zone with water sources of both Atlantic and Arctic origin. Accordingly the biota include both boreal and Arctic species but, as is typical for northern seas, are not very diverse. T h e spring phytoplankton bloom consists mainly of diatoms. but also years with dominance of naked flagellates were observed. Considerable interannual variations in the qualitative as well as the quantitative aspects of the spring development in the Norwegian Sea have been acknowledged (Paasche 1960). T h e summer vegetation is rather sparse and dom- inated by microflagellates, coccolithophorids and dinoflagellates. but is more diverse and represents a typical regenerating community. It is not known to what extent autumn blooms take place. The Norwegian Sea is generally rich in mac- rozooplankton with copepods (C. finrnarcliicus, C. hyperboreus), various euphausiids ( e . g . Production and sedimentation in the Northeast Atlantic 217 - 100- Meganyctiphanes noruegica) and pteropods ( L i m - acina retrouersa) being important grazers (Wiborg 1955). 0stvedt (1955) observed neritic plankton to spread out as far as Ocean Weather Ship M I K E which is well within the Norwegian Sea proper. There, Lie (1968) observed large variations in the timing of maximum zooplankton concentrations (up to three months), possibly also due to advec- tion from the Norwegian Coastal Current. Advec- tion from the Norwegian Coastal Current and also along the axis of the Norwegian Current results in blurring spatial distribution patterns of Arctic and boreal zooplankton. C. finmarchicus maintains itself in the Norwegian Sea because of ontogenetic vertical migration. The overwin- tering population drifts southward at depth whereas feeding stages are moved northwards with surface water (Noji 1989). Comparing vertical fluxes from oceanic environments of the Northeast Atlantic with those of the fjords and shelf is difficult because the former are derived from trap deployments below 500m which are greatly reduced due to remin- eralization of matter in the upper hundreds of metres. The seasonal sedimentation of POC at 500 m depth on the V ~ r i n g Plateau in the Norweg- ian Sea off the Norwegian shelf at 67"N is shown in Fig. 5C. Seasonal maxima were generally found during summer with low sedimentation rates in spring and autumn. Three years of trap deploy- ments, however, showed significant interannual variations. The single peaks recorded in April 1986 and November/December 1987 appear unexplained. Despite the scatter in the data, it is obvious that the "regular" seasonal maximum of vertical flux occurs in summer on the V ~ r i n g Plateau and is hence not related to the spring bloom, as is typical for shallow regions described above. Microscopic analyses of trap collections indi- cate that the various identifiable biogenic com- ponents tend to have elevated fluxes at the time of the general seasonal maximum (Fig. 7). It appears that sedimentation on the VGring Plateau is strongly influenced by zooplankton. Through the productive season, zooplankton feeding press- ure is high and loss rates, accordingly, are low. The abundance of copepods over-wintering at depth and of those in small numbers present in the surface in late winter and their ability to respond to increases in food supply must be con- sidered in order to assess grazer control of the phytoplankton spring bloom in the Norwegian P Copepod faecal pellets 300 , Months Diatoms 1490 300 0 300 Tintinnids loo 1 Forammilera lo\ 0 0 0 0 1 5 Pteropods Oval faecal pellets 50 Mlnipellets 2000 Months Fig. 7. Sedimentation rates at 500 m depth on the Vering Plateau during 1986 (open circles), 1987 (filled circles), and 1988 (open squares). Total flux (dry weight) (mg m-2 d-I) and contributions of different biogenic particles (numbers lo3 m-2 d-I). Data replotted from Bathmann et al. (1990a). 218 P. Wassmann, R. Peinert& V. Smetacek Sea (Peinert et al. 1989; Bathmann et al. 1990b). Spring phytoplankton biomass accumulation and sedimentation thus depends on the grazers and blooms may not take place each year. The trap data, never depicting spring flux maxima in vari- ous years indicate in fact that spring phy- toplankton outbursts may be the exception from the rule in this region. During summer, a pulse of faecal pellets was observed, presumably due to a declining recycling efficiency caused by ontogenetic migration of C. finmarchicus copepodites to greater depth (Bathmann et al. 1987). This event triggered a response in benthic metabolism (Graf 1989). Indirect evidence indicates that the presence of large copepod stocks counteract pellet sedi- mentation, although these are produced in large numbers (Smetacek 1980b). The tentative con- clusion is that copepods themselves break up their faeces (named coprorhexy by Lampitt et al. 1990) and contribute organic material to the microbial network hence promoting recycling of waste prod- ucts (Smetacek & Pollehne 1986). However, repackaging of sinking material can also take place at mid-water depths as reflected by oval faecal pellets and minipellets at certain depth horizons (Bathmann et al. 1990a) (Fig. 7). Detailed microscopy of V ~ r i n g Plateau trap samples reveal that not all particles sink out in concert nor are all flux patterns recurrent. The time series (Fig. 7) show, for example, that the pteropod flux maximum is in autumn in contrast to the summer maximum of the other biogenic components. Although not observed in 1987, the pteropod maximum may nevertheless be a typical feature for the Norwegian Current as it was found again in 1988 (Bathmann et al. 1990a) and was also described for the nearby Lofoten Basin (Honjo et al. 1987). Bathmann et al. (1991) show that intense feeding by dense swarms of repro- ducing pteropods enhance flux due to sinking out of feeding webs with adhering particles. After reproduction, mass mortality occurred which was reflected in large numbers of pteropod shells i n sediment catches. A shift of the seasonal maximum of the vertical flux towards summer/autumn seems to be typical for the oceanic areas dominated by the Norwegian Atlantic Current (Fig. 8A. B) which clearly sep- arates them from the coastal and shelf waters of the Northeast Atlantic. Compared to the Vsring Plateau, the flux maximum i n the Lofoten Basin (about 650 km northeast of the Vsring Plateau deployment) occurs even later during autumn (Figs. 8 A , B). Off Bjsrndya, west of the Barents Sea entrance, sedimentation rates are far higher as in the Northeast Atlantic proper; seasonality differs and winter values are greatly enhanced (Fig. 8A. B). This is explained by winterly sedi- ment resuspension and subsequent cross shelf transport of nepheloid layers from the Barents Sea into the Norwegian Sea by dense bottom water (Midttun 1984; Honjo et al. 1988; Quad- fasel et al. 1988). Western regions of the northeastern North Atlantic Greenland Sea and Fram Strait. - The water masses of this region consist primarily of two components: Very cold (-1.5 to 1.5"C) and low saline (34.0 to34.6%0) water from the East Green- land Current and slightly warmer and more saline water from the Irminger and West Spitsbergen currents. The area is heavily covered by sea ice during most of the year, the seasonal and interan- nual variability of sea ice, however, is less pro- nounced in the Fram Strait and Greenland Sea area as compared to the Barents Sea (Fig. 2). Sea- ice dynamics are dominated by the Transpolar Current and the East Greenland Current which transports Polar Water and ice from the Polar Sea along the Greenland coast into the Greenland Sea and Icelandic areas. The ice cover- age is still significant at 60"N in the western part of the Northeast Atlantic compared to an average latitude of sea ice in the Barents Sea of about 75"N on the eastern side. Heavy sea ice from the Polar Sea and the Fram Strait blocks the fjords of East Greenland, for example Scoresby Sound, for most of the year. The productivity of these fjords is, therefore, low. Our knowledge about the relationship between plankton production and sedimentation in the Fram Strait and Greenland Sea is limited. However, primary production, zooplankton abundance and hence probably also the vertical flux vary much between cold and warm years i n the waters off southern Greenland (Horsted 1989). Primary production is generally highest in the frontal regions between the East Greenland Current and the warmer water masses. The spring bloom starts in the outer parts of the fjords in southeast Greenland as early as March and max- ima are found in April and July/August (Smidt 1979). Microzooplankton has its maximum in July Production and sedimentation in the Northeast Atlantic 219 Norwegian Atlantic Current 0 61-1 (1700m) 8 NB-1 (2815rn) 0. East Greenland Current C o FS-I (2ooom) p o c GB-2 (1900rn) 0 GB-2 (3000m) 1 '0 J ' F I M ' A l M l J ' J7)A ' S '0 N ' D1 2 0 - B O L E - I D O F S - 1 - 0 NA-1 psi 0 GB-2 ._ c c 0 81.1 a N B - l Fig. 8. Sedimentation rates (mg m-* d-') of particulate organic carbon (POC) and particulate silicate (Psi) as obtained from d e e p areas dominated by the Norwegian Atlantic Current and East replotted from Honjo et al. 5 1 6 - E a - (% 12- moored sediment traps in 8 - Greenland Current. Data -0. -o-o.o-o-o-o-o J ' F M I A ' M ' J ' J A ' S '0 ' N ' D I (1987). Months and the macrozooplankton has a somewhat longer-lasting maximum from June-July onwards. Phytoplankton growth in glacial fjords shows a pronounced seasonal cycle and is enhanced in the outer, more mixed areas as com- pared to the inner, stratified waters (Horsted 1989) where light penetration is lower due to turbidity brought about by the sediment load from glacial meltwater. In the open waters of the Davis Strait the most important herbivores are the copepods C. fin- marchicus, C . glacialis, C . hyperboreus and o. similis. The composition of the zooplankton and the timing of its development are strongly affected by hydrography and ice conditions. C. fin- marchicus is an indicator of Atlantic Water, C. glacialis and C . hyperboreus are characteristic for Arctic Water, while Oithona similis indicates the region of the Polar Front and mixed waters (Pavshtiks 1968). Zooplankton development commences during June in coastal waters and in April to May in the Davis Strait. The timing between primary and secondary production off Greenland is coupled to variations of the East Greenland and Irminger currents that influence ice cover, temperature, and stratification as well as advection of nutrients and organisms. The dis- tributions and abundances of for example cod, salmon and shrimps show accordingly long-term and short-term variations (Horsted 1989), and so should the vertical flux of particles. Polynyas are regularly found in the marginal ice zone off East Greenland (Niebauer & Alexander 1989). The nutrient rich, Polar Water of the Trans- polar Current is exposed to insolation in these polynyas and the marginal ice zone and is trans- ported into open waters (Spies et al. 1988). Along the drift of water exposed to sunlight, sequences of planktonic developments are found in space which would be found in time at an ice-free, subarctic station. As in the Barents Sea, diatoms and P. pouchetii are the dominating phyto- plankton species (Gradinger 1987; Smith et al. 1987). Ice-algae from melting multi-year ice from the Polar Basin may add t o the phytoplankton biomass present in polynyas (Melnikov & Bon- darchuk 1987). The zooplankton is dominated by larger forms such as C. finmarchicus, C. hyp- erboreus and C . glacialis (Barthel 1990; Smith & Schnack-Schiell990); egg production and feeding take place prior t o the spring phytoplankton bloom (Smith 1990) and secondary production within the marginal ice-zone can be high (H. J. Hirche pers. comm.). The zooplankton com- munity shares common features with those of the Barents Sea as well: Atlantic species with annual 220 P. Wassmann, R . Peinert & V . Smetacek life cycles preferring the warmer water to the east of the marginal ice zone predominance (e.g. C. finmarchicus) are contrasted by Arctic species in the marginal ice zone and polynyas which prefer colder water and have a multi-annual life-cycle. Sedimentation data from these regions are sparse and originate from deep-moored traps. POC sedimentation rates as well as seasonal varia- bility i n the Greenland Basin and the Fram Strait are lower (Fig. 8C) as compared to waters dom- inated by the Norwegian Atlantic Current (Honjo 1990). Sedimentation, however, increases during summer and for biogenic silica one episode of increased vertical flux was encountered by the Greenland Sea trap (Fig. 8D). I t is not known to what extent the ice coverage of the deployment areas during the productive period might have limited primary production and hence sedi- mentation. Recent deployments including shal- lower traps, however, indicate a pronounced seasonal vertical flux pattern (B. v. Bodungen pers. comm.). Icelandic waters. - The hydrography around Ice- land is influenced by Atlantic Water from the Irminger Current and Polar Water from the East Greenland Current (Fig. 3). Whereas Atlantic Water and the Icelandic Coastal Current dom- inate the shelf and oceanic areas t o the south and west of Iceland, the waters in the north and east of Iceland are influenced by a complicated mix- ture of different water masses. The latter area is a highly variable frontal zone fed by the Irminger Current (Atlantic Water), the East Icelandic Cur- rent (Arctic Water), the East Greenland Current (Polar Water) as well as the clockwise turning Icelandic Coastal Current (Fig. 3). Most of the fjords of Iceland are deep and well connected to the adjacent shelf. During input of Polar Water, the marginal ice zone comes close to the northern Icelandic coast. Low primary production rates were observed dur- ing such years and attributed to increased strati- fication of surface water, preventing the supply of nutrients from below and hence constraining continued new production (Thoradottir 1977). The spring phytoplankton bloom starts in the southern and western waters of Iceland in near- coastal areas in late March, whereas further in offshore areas its advent is usually in the second half of May (Thoradottir 1986). In the north, the spring phytoplankton bloom ends earlier during cold periods than during warmer periods, usually before the middle of June. Autumn blooms are generally not reported from Icelandic waters. The zooplankton biomass follows that of phyto- plankton in the Atlantic Water and over win- tering is of significance in the fjords (Astthorsson & Jonsson 1988; Astthorsson 1990). The spring bloom of the areas to the south and west of Iceland is, thus, influenced by grazers. During periods when the influx of Atlantic Water is strong and the zooplankton stocks are rich, this may probably also be true for oceanic provinces to the north. Since new production is lower during times of dominating outbreaks of Polar Winter, less zooplankton is accordingly found as compared to warmer periods (Thoradottir & Astthorsson 1984). Extensive changes in the physical, chemical and biological properties in the waters have been recorded especially north of Iceland, reflecting climatical variations (Thoradottir 1977; Stefans- son & Jacobson 1989). While diatoms usually dominate the phytoplankton spring bloom in Ice- landic waters, colonial stages of P . pouchetii increase markedly during the cold periods. Since the 1960s this species has dominated the phy- toplankton community t o the north of Iceland. The P . pouchetii blooms reflect anomalies in the silicate-nitrate relationship ( S t e f h s o n 1990). To what extent P . poucherii colonies are grazed by zooplankton, sink into the aphotic zone and/or are broken down by microbial degradation is still a matter of dispute (Huntley et al. 1987; Hansen et al. 1990; Estep et al. 1990). However, P. p o u - chetii colonies have been observed to sink rapidly in Arctic and subarctic waters of the Barents Sea (Wassmann et al. 1990), adding nutrient depletion of the euphotic zone and thus limiting further production in the waters north of Iceland. The highly productive front areas off Iceland and Jan Mayen support large zooplankton stocks which have been the basis for large pelagic fish- eries (for example herring, capelin and blue whit- ing). The herring has played a key role as a planktonic carnivore in Icelandic waters and of the Norwegian Sea (summarised by Devold & Jacobsen (1968) and Dragesund et al. (1980)). Herring feed mostly in the waters between Iceland, Jan Mayen and the Icelandic Gap Front during May to September. However, the northern Norwegian Sea, for example the waters west of B j ~ r n o y a . is also a feeding ground. The herring must have had a pronounced influence on the vertical flux of biogenic matter by grazing on Production and sedimentation in the Northeast Atlantic 22 1 zooplankton and by producing faeces, as fish faeces can contribute significantly to vertical flux (Staresinic et al. 1983; Heussner et al. 1987). As a consequence of increased stability due to increased supply of cold water from the East Greenland Current from 1966 on, productivity of the Icelandic Gap Front has since generally decreased (Stefansson & Jacobson 1989). It can be speculated that climatic induced reduction in zooplankton stocks also resulted in decreased stocks of carnivores as zooplankton resources might be limited in relation to food requirements of zooplankton feeders, for example herring (Pavshtiks & Timokhina 1972). The effect of the disappearance of herring on other stocks and the zooplankton community is difficult t o assess and so far no observations are available which indicate that the herring has been replaced by any other major planktivore. No vertical flux data are avail- able from Icelandic waters. We suggest that sedi- mentation today is different from that in the mid- 1960s and before. Evaluation of case studies The case studies clearly indicate that the relation- ship between annual patterns of primary pro- duction and those of sedimentation can vary considerably. The factors determining vertical flux interact with one another so that flux rates and patterns can exhibit large regional and interannual variability. In the following we exam- ine the role of the governing factors individually in order to assess their specific impact on processes governing particle production, modification and sedimentation. Nutrients, water column stability and spring phytoplankton growth rate The potential new production is dictated by depth of the mixed layer but also by initial nutrient concentrations. Nitrate is generally accepted to be the controlling nutrient species. Diatoms and also P. pouchetii blooms respond rapidly when ambient nitrate is exhausted and mass sedi- mentation is commonly observed. The deploy- ment of ultra-clean methods to monitor essential trace elements has provoked a debate whether iron in particular is the limiting nutrient in some areas of the Pacific Ocean and the Antarctic (Mar- tin & Fitzwater 1988; Banse, 1990; Martin 1990; Dugdale & Wilkerson 1990). This debate is beyond the scope of this paper and future results may elucidate a potential impact on the vertical flux of particles. At present, it could only be speculated that, for example, diatoms may respond in a different way to iron limitation com- pared to nitrate limitation. Aggregation and rapid sinking from surface layers following nitrate limi- tation (Smetacek 1985) may not be triggered if iron is the limiting element. A shallow mixed layer in spring is a prerequisite for development of the spring bloom. The depth of this layer in relation to light supply determines the rate of biomass accumulation during this period (Smetacek & Passow 1990). The shal- lowest mixed layers are found in sheltered waters but also along fronts between waters of different salinities, for instance on the borders of river plumes and at the edge of melting sea ice. Massive blooms covering restricted areas are characteristic of such shallow mixed layers, and blooms cul- minate as a result of ambient nutrient exhaustion. These are the blooms most likely to form aggre- gates on a large scale that sink out of the mixed layer rapidly (Smetacek 1985). The percentage of new production lost t o the system via vertical flux is highest under such conditions. In contrast, spring growth that commences in a deep mixed layer deplenishes nutrients and accumulates bio- mass slowly. The impact of grazing is obviously much greater on a phytoplankton population growing slowly as compared t o one growing rapidly. Accordingly, the percentage loss via sink- ing cells and phytodetritus is smaller and the proportion of organic material channelled t o the grazer pools is larger. However, depending on the degree to which available nutrients are utili- sed, total new production can be much higher in regions with a deeper mixed layer, hence also the potential for support. Although the average depth of the mixed layer is a characteristic feature of specific regions, it is still dependent on short-term weather conditions. In land-locked fjords, the annual variation will be restricted; in open ocean conditions, variation in the biomass accumulation rate can vary greatly. Thus, under cloudy, stormy conditions, mixing will be thorough and average cell division rates low. Under the influence of a stable high pressure zone, heating of the upper metres will lead to temporary stabilisation and rapid accumulation of biomass since cells and their progeny will con- tinue to divide at maximal rates. One can assume that in such a situation, the thin, warmer water 222 P. Wassmann. R . Peinert & V . Smetacek layers will not lie like a sheet over the colder winter water. Rather. the warming water will be susceptible to horizontal displacement and for- mation of warm water lenses on the kilometre scale. The horizontal distribution of algal biomass will be correspondingly patchy. Indeed, this mechanism is probably responsible for the patchy distribution of the spring bloom recorded in the Norwegian shelf by Peinert (1986a), along the Polar Front and in the marginal ice zone of the Barents Sea and Fram Strait (H. J . Hirche, pers. comm.) and also i n the open Baltic during PEX 86 (Passow 1990). Smetacek (1985) postulated that mass flux is preceded by aggregation of cells and chains. Aggregate formation in nutrient exhausted blooms has since been observed in nature (Kranck & Milligan 1988; Alldredge & Gotschalk 1989; Riebesell 1989). It follows that the higher the phytoplankton concentration the greater the speed of aggregate formation and hence also the proportion of the bloom lost to sedimentation. This is the case during blooms of sticky colonies of P. poucherii (Wassmann et al. 1990). Evidence for such a relationship is circumstantial at present but we see little reason t o suspect that it does not apply in nature. In this case horizontal patchiness of bloom biomass would result in even greater patchiness in vertical flux. Composition of phytoplankton Sediment trap data to date indicate that diatoms and coccolithophorids - both algae equipped with mineral covers - contribute a much greater pro- portion to vertical flux than flagellates without a mineral burden. This is illustrated by the situation described from the North Sea Fladen Ground in 1976 and in 1983: when diatoms dominated, heavy sedimentation was recorded whereas flux rates were very low when the colonial flagellate Corym- bellus aureus dominated the spring bloom. The factors responsible for this inter-annual shift in dominance are unknown but their effect on ver- tical flux is obvious. The degree to which the latter situation can be classified as unusual also awaits clarification. The fate of another, much more widespread. colonial flagellate (Phaeo- cystis) appears to be both retention within the mixed layer as well as sedimentation. Thus, in the North Sea, huge masses of foam emanating from declining Phaeocystis blooms accumulate on the coast (Batje & Michaelis 1986), a phenom- enon which never occurs in the case of the larger diatom blooms that precede Phaeocystis. We con- clude that this flagellate, following senescence of its blooms, does not sink out as readily as do diatoms but, under conditions not yet specified, can indeed settle out of the mixed layer as observed in the Barents Sea (Wassmann et al. 1990). Much more information is required on the sinking behavior of different phytoplankton groups and species to further our knowledge of vertical flux. In recent studies in the Baltic and the Barents Sea, Passow (1990) and Wassmann & Rey (unpubl. data), respectively, found significant species-specific differences in sinking behavior within a mixed spring bloom population. The role of grazers As one moves from shallower t o deeper water, the role of grazers increases in importance to an extent that grazing pressure can significantly retard bloom growth and hence prevent sinking out of cells and phytodetritus. Whether sedi- mentation occurs subsequent t o heavy grazing in the form of herbivore faeces or whether it is retarded altogether depends on the type of grazer dominating this trophic level. Protozoans, particularly the large forms that feed by extrusion of a feeding appendage (Jakobson & Anderson 1986) can well exert significant grazing pressure on phytoplankton including large, chain-forming diatoms. Digestion in these organisms seems to be fairly thorough and faeces attributed to this group generally consist of empty frustules. These so-called minipellets can be an important vehicle for transporting whole diatom frustules to the deep sea (Gowing & Silver 1985; Nothig & Bodungen 1989). However, we are far from quantifying the role of these organ- isms in pelagic systems. Their influence on sedi- mentation patterns is likely to be substantial and warrants investigation. Field data (Bathmann et al. 1987; Peinert et al. 1987; Noji 1989) and laboratory experiments (Lampitt et al. 1990) suggest that copepod grazing can counteract sedimentation by keeping matter in suspension. Being widespread, more evenly distributed horizontally (for example euphausi- ids) and unable to undertake direct horizontal migrations, copepods cannot easily leave their feeding grounds. In this context, coprorhexy (mechanical destruction of their own and other’s faeces; Lampitt et al. 1990) and coprophagy are Production and sedimentation in the Northeast Atlantic 223 important mechanisms. By conditioning feeding grounds they would ensure that resources are retained and recycled and not “wasted” by sedi- mentation to depths beyond the copepods’ reach. The efficiency of retention would be intimately related to copepod population density and food availability and hence not solely dependent on the immediate growth environment. During spring, copepod over-wintering success, the timing of their ascent from hibernation, population density and grazing pressure in surface waters are factors that exert control on “successful” retention or loss by sedimentation of primary produced matter. This is also supported by a model of phy- toplankton growth and sedimentation under vary- ing influence of copepod grazing by Wassmann & Slagstad (unpubl. results). Assuming that no zooplankton grazing takes place, the model indi- cates that total primary production is solely deter- mined by physical factors and sedimentation is high (Fig. 9A). If the number of copepods in the model is increased (Fig. 9B, C), retention of suspended matter in the upper layers becomes more efficient, sedimentation decreases accord- ingly and the phytoplankton biomass maximum is lower as well. Biomass of grazers of the copepod type which is controlled by the copepod’s over- wintering success and advection during late winter and spring obviously has a dominating influence on the carbon flux in the southern Barents Sea. As a consequence, grazing pressure can be expected t o be highly variable from year to year, as the maximum macrozooplankton biomass shows a wide range between 1.5 and 15 g C m-2 (Skjoldal et al. 1987). Grazing by large, roving crustaceans such as euphausiids produces faeces that sink too rapidly for easy recycling at depths of their production. Thus these crustaceans inherently deplete their own and others’ feeding grounds. To sustain large Fig. 9. Results of a model run by Wassmann & Slagstad (unpubl. results): The effect of zooplankton grazing during early spring (variation in overwintering success and advection) in Atlantic Water of the Barents Sea at 72”30’N. The number of copepodite stage V are indicated. Primary production sedimentation (-) in g C phytoplankton (----) and zooplankton biomass (.....) in g C m - z . (-._._._) and ,,-Zd-l. , suspended 15 12 0.9 - 0.6 - u N E 0.3 0 s 0.0 c I B 2000 copepods I c 1.5 & 0 c - 20 Y 1.2 - -10 ................... Time (months) N E 0 m - - I C 0 1 a 0 N c m - 224 P. Wassmann, R . Peinert& V . Snietacek populations of this type of herbivore. however, horizontal migration to hitherto ungrazed blooms or continued new production by physical nutrient supply must compensate for losses via sedimented faeces. These organisms, however, also would benefit from copepods and other grazers that recycle matter in near surface layers. Other herbivores such as pteropods and some gelatinous plankters (salps. appendicularians) are less predictable i n their appearance. but may be abundant in very high numbers in some years and may significantly influence sedimentation. They all feed efficiently and non-selectively on small particles and, for example, faeces of salps can contribute significantly to the vertical flux (Bathmann 1988). Discharged pteropod feeding nets (Noji 1989) and abandoned appendicularian houses (Alldredge 1986) may play a role in accel- erating losses by scavenging small and slower sinking particles. Finally, the organisms them- selves (e.g. pteropods shells) can constitute a seasonally important fraction of settling inor- ganics (Bathmann et al. 1991). A further and indirect impact on sedimentation, which as yet can only be speculated upon, may result from changes in food web dynamics due to the presence of succession within herbivores. For the open Norwegian Sea, the maximum sedimentation, as encountered in late summer, could be related to a shift from a food web with copepods as major grazers towards a dominance of pteropods (Bathmann et al. 1990a). During years of very high salp abundance, as observed non-periodically and non-predictably in Norwegian waters (Brattstrom 1972). we specu- late that flux patterns would differ from normal years depending on the season of their appear- ance: if large salp biomass appeared early in spring, most of the phytoplankton biomass could leave the upper layers due to rapidly sinking faeces. This would leave copepods little chance to increase the size of their population. Such a scenario would result in a strong spring pulse in sedimentation and the observed summer or autumn pulse would be greatly reduced. Climatic change and the role of advection The fjord examples indicate that advection. dri- ven by the local wind field, can determine the annual flux in these coastal systems. I f , for example. the spring advective event referred to above had occurred weeks later, heavy sedi- mentation of the phytoplankton bloom could have been expected within the fjord. A second sedi- mentation pulse following an upwelling event could have taken place in addition to the first. We can postulate that in such open fjord systems, sedimentation patterns are governed by stochastic events, and hence are highly variable, whereas in closed fjords interannual variation will be lower due to limited advection. I n contrast to the subarctic North Pacific, the Northeast Atlantic on the whole can be viewed as a huge advective flow-through system. The major impact of this large scale latitudinal advec- tion along a gradient of physical environmental conditions on sedimentation will be different as compared to the direct influence in the fjords. It is the advection of grazers to latitudes eventually unfavourable for their reproductional success which we believe to be important. The Barents Sea provides examples of this: the number of copepods advected from the Norwegian Sea var- ies interannually due t o variations of the inflow of Atlantic Water (Blindheim & Loeng 1981). The grazing pressure on spring phytoplankton will vary accordingly. Long-term changes in the location and strength of major current systems influence the North Atlantic as a whole (Parsons & Lalli 1988). Cole- brook (1978) and Taylor (1978) observed that zooplankton is less abundant when the influence of the North Atlantic Current on the Northeast Atlantic is weak. Also, periods of hydrographic anomalies in the Norwegian and Barents seas do have implications for the pelagic food web. As a result of increased stratification due t o a strength- ening of the East Icelandic Current, primary and secondary production has decreased in the Ice- landic Gap Front and herring stocks have dis- appeared. Although fishing was involved in the decline of herring stocks, there is reason to believe that variability in the pelagic system due to changed hydrographical conditions contributed to the observed change as well. A similar dev- elopment was observed in the Barents Sea. Here, capelin disappeared as a result of climatically forced changes in zooplankton stock and fishing pressure. The pelago-benthic coupling and the food production for fish is strongly influenced by lateral transport of Atlantic Water into the southern Barents Sea (Fig. 9) and, thus, subject to climatic variations (Blindheim & Loeng 1981; Skjoldal & Rey 1989). The reasons for this varia- bility are as yet poorly understood, but we specu- Production and sedimentation in the Northeast Atlantic 225 late that the pelago-benthic coupling in the Nordic Seas may have changed significantly since the decline of the herring and capelin stocks. Also, vertical flux may have decreased during recent years since coprophagy and coprorhexy of rich and well-structured zooplankton communities result in retention of food sources in the upper layers. The vertical flux studies accomplished dur- ing recent years may, thus, not be representative for previous periods with large abundances of pelagic fishes such as herring and capelin. Since the Norwegian and Barents seas experience large climatic and biological variations, no ‘normal’ pattern of production and sedimentation can be derived from a few annual studies. The dominant copepods in the North Pacific differ in their life cycles from those in the Atlantic. One school of thought suggests that this difference is the underlying cause for the striking differences in annual cycles of phytoplankton (Frost et al. 1983; Frost 1987). In the subarctic North Pacific gyre, nutrient reserves (nitrate) are never exhaus- ted and spring blooms do not occur. This is dif- ferent in the North Atlantic. An establishment of the Pacific copepods in the North Atlantic, following from this reasoning, would entirely change its biology towards those patterns observed in the Pacific. We tend to believe, how- ever, that this would have been ‘tested out’ in the course of evolution and we rather suggest that the cause for the difference lies in the advective nature of the Atlantic as compared to the Pacific. This different nature of the North Atlantic may not allow the establishment of overwintering copepod populations that are tightly geared to the seasonal production regime in the overlying surface layers as in the Pacific. Colebrook (1985) has pointed out that inter-annual variation in copepod biomass is governed by the size of the overwintering population. Peinert et al. (1987, 1989) suggest that this might hold true for the Norwegian Sea and the model of Wassmann & Slagstad (unpubl. results) suggest that this is the case for the Barents Sea as well. If the seasonally first appearing copepods first serve to condition their environment by their activity in such a way as to open and maintain an adequate niche for their progeny, the size of the initial stock would indeed determine reproductive success in that year. Such biological gearing within populations and between trophic levels could be affected in unexpected ways by variability of the physico- chemical environment. It is the task of future studies to establish such mechanisms and unravel the various processes at work in shaping and driving marine pelagic systems. Information on vertical flux will continue to be a prerequisite for furthering our understanding of pelagic systems in this context. Acknowledgements. - P. Wassmann was supported by the Norwegian Research Council for Science and the Humanities (NAVF) and the Norwegian Ministry of the Environment. This publication is a contribution from the Norwegian Research Program for Marine Arctic Ecology (Pro Mare), publication No. 91 of the Sonderforschungsbereich 313, University of Kiel. and No. 332 of the Alfred-Wegener-Institute for Polar and Marine Research, Bremerhaven. References Alldredge. A . L. 1986: Aggregate dynamics: Biological processes. which form, alter and destroy aggregates in the ocean. Pp. 109-130 in Alldredge, A. L. & Hartwig, E. 0. (eds.): Aggregate dynamics in the sea: Workshop report. American Institute of Biological Sciences, Washington D.C. Alldredge, A. L . , Silver, M. W. 1988: Characteristics. dynamics and significance of marine snow. Prog. Oceanogr. 20,41-82. Alldredge. A. L. & Gotschalk. C. 1989: Direct observations of the flocculation of diatom blooms: characteristics, settling velocities and formation of diatom aggregates. Deep-sea Res. 36. 159-171. Astth6rsson. 0. S. 1990: Ecology of the euphausiids Thys- sanoessa raschii, T. inermis and Meganyctiphanes noruegica in Isafjord-deep, northwest Iceland. Mar. Biol. 107,147-157. Astth6rsson. 0 . S. & Jonsson. G. S. 1988: Seasonal changes in abundance in Isafjord-deep. northwest Iceland. in relation to chlorophyll a and hydrography. Counc. Meet. Int. Counc. Explor. Sea 1988fL:3. 18 pp. Batje, M. & Michaelis. H . 1986: Phaeocystis pouchetii blooms in East Frisian coastal waters German Bight. North Sea. Mar. Biol. 93. 21-27. Banse. K. 1990: Does iron really limit phytoplankton pro- duction in the offshore sub-arctic Pacific? Limnol. Oceanogr. 35, 772-775. Barthel, K.-G. 1990: Relationships of food uptake and body components of Calanus finmarchicus, C . glacialis and C . hyperboreus to particulate matter and water characteristics in the Fram Strait. Polar Biol. 10, 34S350. Bathmann. U . 1988: Mass occurrence of Salpafusiformis in the spring of 1984 off Ireland: implication for sedimentation processes. Mar. Biol. 97, 127-135. Bathmann. U . , Noji, T. T., Voss. M. & Peinert. R . 1987: Copepod fecal pellets: abundance, sedimentation and content at a permanent station in the Norwegian Sea in May/June 1986. Mar. Ecol. Prog. Ser. 38. 45-51. Bathmann. U., Peinert. R. & Bodungen, B. v. 1990a: Pelagic origin and fate of sedimenting particles in the Norwegian Sea. Prog. Oceanogr. 2 4 . 117-125. Bathmann. U., Noji. T. & Bodungen, B. v. 1990b: Copepod grazing potential in late winter in the Norwegian Sea - a factor in the control of spring phytoplankton growth? Mar. Ecol. Prog. Ser. 60, 225-233. Bathmann, U . , Noji, T . & Bodungen, B. v. 1991: Sedimentation of pteropods in the Norwegian Sea in autumn. Deep-sea Res. 10, 1341-1351. 226 P . Wassmann, R . Peinert & V , Smetacek Berger. W.. Smetacek. V. & Wefer. G. 1989: Ocean pro- ductivity and paleoproductivity. Pp. 1-34 in Berger, w . , Smetacek. V. & Wefer. G . (eds.): Producfiuify o f f h e ocean: presenf and past. John Wiley & Sons, New York. Blindheim, J . & Loeng. H . 1981: On the variability of Atlantic influence in the Norwegian and Barents Sea. Fiskdir. Skr. Ser. Hauunders. 17, 161-189. Blindheim, J. 1989: Ecological features of the Norwegian Sea. Pp. 366-401 inRey,L.&Alexander.V. (eds.): Proc. 6fh Conf. Com. Arcfiquelnternatl, 13-15 May 1986. E . J . Brill. Leiden. Braarud,T.,Gaarder. R . K. & Nordli.O.1958: Seasonalchanges in the phytoplankton at various points of the Norwegian west coast. Fiskdir. Skr. Ser. Hauunders. 13. 77 pp. Brattstrom, H. 1972: On Salpafusiformis Cuvier (Thalaciacae) in Norwegian coastal and off-shore waters. Sarsia 48.71-90. Brewer. P. G., Bruland, K. W., Eppley, R. W. & McCarthy, J . J . 1986: The Global Ocean Flux Study (GOFS): Status of the U.S. GOFS Program. Eos 67,827-832. Cadee, G . C. 1985: Macroaggregatesof Emiliana huxleyiin sedi- ment traps. Mar. Ecol. Prog. Ser. 24, 193-196. CadCe, G. C. 1986: Organic carbon in the water column and its sedimentation, FIaden Ground (North Sea). Nefh. J. Sea Res. 20.347-358. Colebrook. J . M. 1978: Continuous plankton records: zooplank- ton and environment, North-East Atlantic and North Sea: 1948-1975. Oceanol. A c f a I. 9-23. Colebrook. J. M . 1985: Continuous plankton records: over- wintering and annual fluctuations in the abundance of zooplankton. Mar. Biol. 84.261-265. Davies, J . M. & Payne, R. 1984: Supply of organic matter to the sediment in the northern North Sea during a spring phy- toplankton bloom. Mar. Biol. 78,315-324. Deuser, W. G. 1988: Variabilityof hydrography and particle flux: transient and long-term relationships. Pp. 17S193 in Degens, E. T . , Izdar, E . I . , Honjo. S. (eds.): ParficleJlw in the ocean. Mitt. Geol.-Palaont. Inst.. Univ. Hamburg. SCOPE/UNEP Sonderband 62. Deuser, W. G . & Ross, E . H. 1980: Seasonal changes in the flux of organic carbon to the deep Sargasso Sea. Nature ( L o n d . ) 283.364-365. Devoid, F. & Jacobson, J . 1968: The formation and disap- pearance of a stock unit of Norwegian herring. Fiskdir. Skr. Ser. Hauunders. 15. 1-22. Dickson, R . R., Lamb, H. H. & Malmberg, S. A , , Colebrook 1975: Climatic reversal in the northern north Atlantic. Nature (Lond.)256,479-482. Dickson, R . R . & Blindheim. I . 1984: On the abnormal hydro- graphic conditions in the European Arctic in the 1970s. Rapp. P.-u. Rdun. Cons. I n t . Explor. Mer 18s. 201-213. Dickson. R . R . , Meinke. J . , Malmberg, S.-A. & Lee, A. J . 1988: The 'grcat salinity anomaly' in the northern North Atlantic 1968-1982. Prog. Oceanogr. 22,103-151. Dragesund. 0 . Hamre, 1 . & Ulltang. 0.1980: Biologyandpopu- lation dynamics of the Norwegian spring-spawning herring. Rapp. P.-u. Rdun. Cons. I n t . Explor. Mer 177.43-71. Dugdale, R . C. & Wilkerson. F. P. 1990: Iron addition experi- ments in the Antarctic: A reanalysis. Global Biochem. Cycles 4.13-19. Eilertsen. H. C . , Schei. B. & Taasen. J . 1981: Investigations on the plankton community of Balsfjorden, northern Norway. Sarsia66, 129-141. Eilertsen, H . C.. Tande, K. & Hcgseth. E. N. 1989: Potential of herbivorous copepods for regulating thc spring phytoplan kton bloomintheBarentsSea. Rapp. P.-u. Rdun. Cons. Int. Explor. Mer 188, 154-163. Eppley. R. & Peterson, B. J . 1979: Particulate organic flux and planktonic new production in the deep ocean. Nafure ( L o n d . ) Estep, K. W.. Nejstgaard, J . C . & Skjoldal, H. R. 1990: Grazing of copepods upon Phaeocystis colonies as a function of the physiological state of the prey. Mar. Ecol. Prog. Ser. 67,235- 249. Frost, B. W., Laury, M. R. & Hasset. R. P. 1983: Feeding behavior of large calanoid copepods Neocalanus crisfatus and N. plumchrus from the subarctic Pacific Ocean. Deep-sea Res. Frost, B. W. 1987: Grazingcontrol of phytoplanktonstock in the open subarctic Pacific Ocean: A model assessing the role of mesozooplankton, particularly the large copepods Neocalanus spp. Mar. Ecol. Prog. Ser. 39.49-68. Gammelsr0d. T., Holm, A . 1984: Variationsof temperature and salinity at Station M (66"N. 2"E) since 1948. Rapp. P.-u. Rdun. Cons. Int. Explor. Mer 185,188-200. Gieskes, W. W. C . & Kraay, G . W. 1986: Analysis of phy- toplankton pigments by HPLC before, during and after mass occurrence of the microflagellate Corymbellus aureus during the spring bloom in the open northern North Sea in 1983. Mar. Biol. 92.45-52. GOFS (1989): Sediment frap technology andsampling. Report of the U.S. GOFS working group of sediment trap technology and sampling, Woods Hole. 93 pp. Cowing, M. M., Silver, M. W. 1985: Minipellets: a new and abundant sizeclassofmarine fecal pellets. J. Mar. Res. 43,395- 418. Graf, G . 1989: Benthic-pelagic coupling in a deep-sea benthic community. Nature ( L o n d . ) 341.437439. Gradinger, R. 1986: Die Phyfoplankfonuerteilung in der Fram- straJ3e zwischen Gronland und Spitzbergen. Diplom Thesis, Univ. Kiel. 94 pp. Grebmeier, J . M., McRoy, C . P. & Feder, H. M. 1988: Pelagic- benthic coupling on the shelf of the northern Bering and Chukchi Sea I. Food supply and benthic biomass. Mar. Ecol. Prog. Ser. 4 8 , 5 7 4 7 . Halldal, P. 1953: Phytoplankton investigations from weather ship M in the Norwegian Sea, 1948-49. Hualrdd Skr. 38. 91 pp. Hansen, B..Tande. K. & Berggreen, U. C . 1990: On the trophic fate of Phaeocystis pouchefii (Hariot). 111. Functional responses in grazing demonstrated on juvenile stages of Cal- anusfinmarchicus (Copepoda) feddiatoms and Phaeocystir. J. Plankton Res. 12,1173-1 187. Heinrich, A. K. 1962: The life histories of plankton animals and seasonalcyclesof planktoncommunitiesin theoceans. J. Cons. I n f . Explor. Mer 27,15-24. Heussner, S . , Monaco, A. & Fowler, S. W. 1987: Charac- terisation and vertical transport of settling biogenic particles in the northwestern Mediterranean. Pp. 127-147 in Degens. E. T.. Izdar, E. I. & Honjo, S . (eds.), Parriclepux in the ocean. Mitt. Geol.-Palaont. Inst.. Univ. Hamburg, SCOPE/UNEP Sonderband 62. Honjo, S. 1990: Particle fluxes and modern sedimentation in polaroceans. Pp. 687-739inSmith. W. 0. (ed.): Polar Ocean- ography. Academic Press, New York. Honjo, S., Manganini, S. J.. Karowe, A . & Woodward, B. L. 1987: Particle fluxes. North-Eastern Nordic Seas: 1983-1986, Woods Hole Oceanogr. Inst. Tech. Rept. WHOI-87-17.85 pp. Honjo. S . , Manganini, S. J . & Wefer, G. 1988: Annual particle 282,677-683. 30, 1-13. Production and sedimentation in the Northeast Atlantic 221 flux and a winter outburst of sedimentation in the northern Norwegian Sea. Deep-sea Res. 35. 1223-1234. Hopkins, C. C. E . , Tande, K. S., Grbnvik, S . . Sargent, J. R. & Schweder. T. 1985: Ecological investigations of the zooplank- ton community of Balsfjorden. Northern Norway: Growth. and quantification of condition, in relation to over-wintering and food supply in Metridia longa. Calanusfinmarchicw, Thy- sanoessa inermis and Thysanoessa raschi. Pp. 83-102 in Gray, J . S. & Christensen, M. E. (eds.): Marine biology of polar regions and effecr of stress on marine organisms. John Wiley and Sons, New York. Horsted, S. A . 1989: Some features of oceanographic and bio- logical conditions in Greenland waters. Pp. 4 5 6 4 7 6 in Rey, L. & Alexander, V. (eds.): Proc. 6rh Conf. C o m . Arctique Inrernat., 13-15 May 1985. E . J. Brill, Leiden. Huntley, M., Tande. K. & Eilertsen, H. C. 1987: On the trophic role of Phaeocystis pouchetii (Hariot). 11. Grazing of Calanus hyperboreus (Krbyer) on diatoms and different size categories of P. pouchetii. J . Exp. Mar. Biol. Ecol. 110,197-212. Hbpner-Petersen, G . 1989: Benthos, an important compartment innorthernaquaticecosystems. Pp. 162-178inRey. L. &Alex- ander, V. (eds.): Proc. 6rh Conf. Com. Arctiquelnrernat. E . J. Brill, Leiden. Jacobson, D. M. & Anderson, J. M. 1986 Thecate heterotrophic dinoflagellates: feeding behavior and mechanisms. J . Phycol. 22,249-258. Koblentz-Mishke, 0 . J., Volkovinsky. V. V. & Kabanova, J . K. 1970: Plankton primary production of the world ocean. Pp. 183-193 in Wooster, W. (ed.): Scientific Exploration of rhe South Pacific, Proceedings of the symposium, Nat. Acad. Sci., Washington, D.C. Kranck, K. & Milligan, T. G. 1988: Macroflocs from diatoms: in situ photography of particles in Bedford Basin, Nova Scotia. Mar. Ecol. f r o g . Ser. 44, 183-189. Lampitt, R. S., Noji, T. & Bodungen, v. B. 1990: What happens to zooplankton fecal pellets? Implications for material flux. Mar. Biol. 104, 15-23. Lie, U. 1968: Variationinthequantityofzooplanktonandpropa- gation of C. finmarchicus at station M in the Norwegian Sea 1959-66. Fkkdir. Skr. Ser. Haounders. 14.121-128. Lindahl, 0. 1987: Plankton community dynamics in relation to water exchange in the Gulmarford, Sweden. Ph.D. thesis, Univ. Stockholm. 138 pp. Lut1er.S.. Taasen,J. P.. Hopkins, C. C. E. & Smetacek, V. 1989: Phytoplankton dynamics and sedimentation processes during spring and summer in Balsfjord, Northern Norway. Polar Biol. 10,113-124. Martin, J. H. 1990: Iron in Antarctic waters. Nature (Lond.)345, 156158. Martin, J. H . & Fitzwater, S. E. 1988: Iron deficiency limitsphy- toplankton growth in the north-east Pacific subarctic. Nature (Lond.) 331,341-343. Melnikov, A. & Bondarchuk, L. L. 1987: Ecology of mass accumulation of colonial diatom algae under drifting Arctic ice. Oceanology 27.233-236. Midttun, L. 1984: Formation of heavy bottom water in the Bar- ents Sea. Counc. Meer. Int. Counc. Explor. Sea 1984/C. 17. 9 PP. Niebauer, H. J., Alexander, V. 1989: Current perspectives on the role of ice margins and polynyas in high latitude ecosystems. Pp. 121-144in Rey.L. &Alexander. V. (eds.): Proc. 6thConf. Com. Arcrique Internat., 13-15 May 1985. E. J . Brill. Leiden. Nothig, E.-M. &Bodungen. B. v. 1989: Occurrence andvertical flux of faecal pellets of probably protozoan origin in the south- eastern Weddell Sea (Antarctica). Mar. Ecol. frog. Ser. 56. Noji, T. 1989: The influence of zooplankton on sedimentation in the Norwegian Sea. Ber. Sonderforschungsbereich 313, U n i u . Kiel17. 183 pp. Bstvedt, 0. J. 1955: Zooplankton investigations from the weather ship M in the Norwegian Sea. Hualrdd. Skr. 40.93 pp. Paasche, E. 1960: Phytoplankton distribution in the Norwegian Sea in June. Fiskdir. Skr. Ser. Hauunders. 12. 77 pp. Pars0ns.T. R. & Lalli, C. M. 1988: Comparative oceanicecology of the plankton communities of the subarctic Atlantic and Pacificoceans. Oceanogr. Mar. Biol. Ann. Reu. 1988.317-359. Passow, U. 1990. Vertikalverteilungund Sedimentationvon Phy- toplanktonarten in der mittleren Ostsee wahrend des Friihjah- res 1986. Ber. Inst. Meereskunde, Kiel. 203 pp. Pavshtiks, E. A . 1968: The influence of currents upon seasonal fluctuation in the plankton of the Davis Strait. Sarsia 34.38% 392. Pavshtiks, E. A. & Timokhina, A. F. 1972: History of inves- tigationson plankton in the Norwegian Sea and main results of Soviet investigations. Proc. R . SOC. Edinburgh 73,267-278. Peinert, R. 1986a: Saisonale und regionale Aspekte der Pro- duktion and Sedimentation von Partikeln im Meer. Ber. Son- derforschungsbereich 313, Unio. Kiel2. 108 pp. Peinert, R . 1986b: Production, grazing and sedimentation in the Norwegian Coastal Current. Pp. 361-374 in Skreslet, S. (ed.): The role of freshwater ourfow in coastal marine ecosystems. Springer Verlag, Berlin. Peinert, R., Bathmann, U., Bodungen, B . v. & Noji, T. 1987: The impact of grazing on spring phytoplankton growth and sedimentation in the Norwegian Current. Pp. 149-164 in Degens, E. T.. fzdar. E. 1. & Honjo, S. (eds.): Particleflux in fhe ocean. Mitt. Geol.-Palaont. Inst., Univ. Hamburg. SCOPE/UNEP Sonderband 62. Peinert, R.. Bodungen, B. v. & Smetacek, V. 1989: Food web structure and loss rates. pp. 34-48 in Berger, W., Srnetacek, V. & Wefer, G . (eds.): Producrioity of the ocean: present and past. John Wiley and Sons, New York. Platt, T . , Harrison, W. G., Lewis, M. L . , Li, W. K. W.. Sathy- endranath, S., Smith, R. & Vezina, A . F. 1988: Biological productionandtheoceans: thecaseforaconsensus. Mar. Ecol. f r o g . Ser. S2.77-88. Quadfasel, D., Rudels, B. & Kurz, K. 1988: Outflow of dense water from a Svalbard fjord into the Fram Strait. Deep-sea Res. 35,1143-1150. Rey, F. 1981: The development of the spring phytoplankton out- burst at selected sites off the Norwegian coast. Pp. 649-680 in Saetre, R. & Mork. M. (eds.): TheNorwegian Coasral Currenr, Univ. Bergen. Rey, F. & Loeng, H . 1985: The influence of ice and hydrographic conditionson thedevelopment of phytoplanktonin the Barents Sea. Pp. 49-63 in Gray, J. S. & Christensen, M. E. (eds.): Marine biology of polar regions and effect of srress on marine organisms. John Wiley and Sons, New York. Rey,F.,Skjoldal, H. R. ~4Slagstad.D. 1987:Primaryproduction in relation t o climatic changes in the Barents Sea. Pp. 29-46 in Loeng, H. (ed.): The effect of oceanographic conditions on distribution and population dynamics of commercial fish stocks in rhe Barenu Sea. Proc. 3rd Soviet-Norwegian Symp., Murmansk, 2 6 2 8 May 1986. Institute of Marine Research, Bergen. Riebesell, U. 1989: Comparison of sinking and sedimentation rate measurements in a diatom winter/spring bloom. Mar. Ecol. f r o g . Ser. 54, 109-119. 281-289. 228 P . Wassmann, R . Peinert & V . Smetacek Sztre. R. & Mork. M. I Y 8 1 : The Norwegian Coarfal Curreri( Univ. Bergen. 795 pp. Skjoldal. H. R. & Wassmann, P. 1986: Sedimentation of par- ticulate organic matter and silicium during spring and summer in Lindispollene. western Norway. M a r . Ecol. Prog. Ser. 30. 4 9 4 3 . SkjoMal, H . R.. Hassel, A , . Rey, F. & Loeng. H . 1987: Spring phytoplankton development and zooplankton reproduction in the central Barents Sea in the period 1979-1984. Pp. 59-89 in Loeng. H . (ed.): The effect of oceanographic condiriom on distribution a n d p o p u l a t i o n d y n a m i c s o f commercial f i h srocks in rhe Barenrc Sea. Proc. 3rd Soviet-Norwegian Symp.. Murmansk. 2 6 2 8 May 1986. Institute of Marine Research. Bergen. Skjoldal. H . R . & Rey, F. 1989: Pelagicproductionandvariability of the Barents Sea ecosystem. Pp. 2 4 s 2 8 3 in Sherman. K. & Alexander, L. M. (eds.): Biomars a n d geography o f large marine ecosysrems. Westview Press. Boulder. Colorado. Smetacek, V. 1980a: Annual cycle of sedimentation in relation to plankton ecology in western Kiel Bight. O p h e l i a . Suppl. 1 . 65-76. Smetacek, V. 1980b: Zooplankton standing stock. copepod fae- cal pellets and particulate detritus in Kiel Bight. Esruar. Coarr. Mar. Sci. 11.477-490. Smetacek. V. 1985: Role of sinking in diatom Iife-history cycles: ecological. evolutionary and geological significance. M a r . Biol. 84.239-251 Smetacek. V , Brockel. K. v.. Zeitzschel, B. & Zenk. W. 1978: Sedimentation of particulate matter during a phytoplankton spring bloom in relation to the hydrographical regime M a r . B i d . 47.21 1-226 Smetacek, V . . Bodungen. B. v.. Knoppers. B . , Peinert, R . . Pol- lehne. F.. Stegmann. P. & Zeitzschel. B. 1984: Seasonalstage, characterizing the annual cycle of an inshoce pelagic system. Repp. P . - u . RCun. Cons. Inr. Explor. Mer 183. 126135. Smetacek. V. & Pollehne. F. 1986: Nutrient cycling in pelagic systems: a reappraisal of the conceptual framework. Ophelia 26.401-428. Smetacek. V. & Passow, U . 1990: Spring bloom initiation and Sverdrup's critical depth model. L l m n o l . O c e a n o g r . 35. 2 2 1 234. Smidt. E . L. B. 1979: Annual cycles of primary production and of zooplankton at southwest Greenland. M e d d . G r m l . , Bios- cience 1 . 1-53. Smith. S. L. 1990: Egg production and feeding by copepods prior to the spring bloom of phytoplankton in Fram Strait. Green- land Sea. M a r . Biol. 106.59-69. Smith. S. L. & Schnack-Schiel. S. B 1990: Polar Zooplankton. Pp. 527-598inSmith. W. 0. ( e d . ) : P o l a r O c e a n o g r a p h y . Aca- demic Press. New York. Smith. W. 0.. Baumann. M. E. M.. Wilson. D. L. & Aletsee. L. 1987: Phytoplankton biomass and productivity in the marginal icezoneofthe FramStraitduringsummerof 1984 1. G e o p h y s . Res. 92,87174786. Spies. A . , Brockmann. U. H. & Kattner. G . 1988: Nutrient regimes in the marginal ice zone of the Greenland Sea in summer. Mar Erol. P r o g . Ser. 47. 195-204. Staresinic. N . Farrington. J . , Gagosian. R . B . . Clifford. C. H . & Hulburt, E. M. 1983: Downward transport of particulate matter in the Peru upwelling: role of the anchoveta Engraulis u p w l l i n g . Its Sedimenrary R e c o r d . Plenum Press, New York Stefansson. U . 1990: Anomalous silicate-nitrate relationships associated with Phaeocysris pouchetii blooms. Eos 71(2): 77. Stefansson. U. &Jacobson, J . 1989: Oceanographicvariations in the Iceland Sea and their impact on biological conditions. Pp. 427-455 in Rey. L. & Alexander. V. (eds.): f r o c . 6 t h Conf. Corn. Arcrique Inremar.. I s 1 5 May 1985. E . J . Brill. Leiden. Suess. E. 1980: Particulate organic carbon flux in the oceans: surface productivity and oxygen utilization. Narure ( L o n d . ) . 288.26&263 Sundby. S. 1984: Influence of bottom topography on the cir- culation at the continental shelf of northern Norway. Fiskdir. Skr. Ser. Hauunders. 17. 501-519. Tande. K S. 1991: C a l a n u in North Norwegian fjords and in the Barents Sea. Pp. 389-407 in Sakshaug. E.. Hopkins. C . C . E. & Britsland. N. A . (eds.): Proceedings of the Pro Mare Symposium on Polar Marine Ecology, Trondheim, 12-16 May 1990. Polar Research IO(2). Tay1or.A. H. 1978: Long-termchangesin North AtlanticCurrent system and their biological implication. Proc. R . Soc. Edin- burgh 76.22s243. Thbradottir. T . 1977: Primary production in northern Icelandic waters in relation to recent climatic changes. Pp. 655-665 in Dunbar. M. J . (ed.): Polar Oceans. Arctic Institute of America, Canada. Thoradbttir. T . 1986: Timing and duration of spring blooming south and southwest of Iceland. Pp. 345-360 in Skreslet. S. ( e d . ) : The role of freshwater ourflow i n coarral marine ecos- w e m s . Springer Verlag. Berlin. Thorad15ttir. T . & Astth6rsson. 0. S. 1984: Chlorophyll a and zooplankton distribution in Icelandic waters in spring 1982. 1983 and 1984. Annales Biologiques 41.63-66. Vahl. 0 . 1980: Volume of water pumped and particulate matter deposited by the Iceland scallop C h l a m y s islandica in Balsfjord, northern Norway. Pp. 639-644 in Freeland, H. J . , Farmer. D . M. & Levings. C. D. (eds.): Fjord Oceanography. NATO Conf. Ser. (Ser. IV; Mar. Sci.). New York. Wadhams. P. 1986: The ice cover. Pp. 21-84 in Hurdle. B. G . (ed.): The Nordic Seas. Springer Verlag. Berlin. Wassmann. P. 1983: Sedimentation oforganicand inorganicpar- ticulate material in Lind!ispollene. a stratified, land-locked fjord in western Norway. M a r . Ecol. P r o g . Ser. 13.237-248. Wassmann. P. 1984: Sedimentation and benthic mineralization of organic detritus in a Norwegian fjord. M a r . Biol. 83,83-94. Wassmann. P. 1989: Sedimentation of organicmatter andsilicate from the euphotic zone of the Barents Sea. R a p p . P . - u . RCun. Cons Inr. Explor. M e r 1 8 8 . 108-114. Wassmann. P. 1990: Relationship between primary and export production in the boreal. coastal zone of the North Atlantic. Limnol. Oceanogr. 35.464-471. Wassmann. P . . Vernet. M.. Mitchell. G . & Rey. F. 1990: Mass sedimentation of Phaeocysris poucherii in the Barents Sea. Mar. Ecol. P r o g . Ser. 66. 183-195. Wassmann. P. 1991: Dynamics of primary production and sedi- mentation in shallow fjords and polls of western Norway. Oce- anogr. M a r . Biol. Ann. R e v . 29.87-154. Wiborg. K. F. 1955: Zooplankton in relation to hydrography in the Norwegian Sea. Fiskdrr. S k i . Ser. H a o u n d e r s . 11. 66 pp. Zenkevitch. L. A . 1963: B i o l o g y o f r h e s e a s o f the USSR. Georgc ringens Pp 22-240 i n Suess. E & Thiede. J (eds ) Coarlal Allen & Unwin Ltd , London