* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download Document
Arctic Ocean wikipedia , lookup
Marine microorganism wikipedia , lookup
Southern Ocean wikipedia , lookup
The Marine Mammal Center wikipedia , lookup
Future sea level wikipedia , lookup
El Niño–Southern Oscillation wikipedia , lookup
History of research ships wikipedia , lookup
Critical Depth wikipedia , lookup
Pacific Ocean wikipedia , lookup
Marine debris wikipedia , lookup
Indian Ocean wikipedia , lookup
Global Energy and Water Cycle Experiment wikipedia , lookup
Abyssal plain wikipedia , lookup
Marine biology wikipedia , lookup
Marine habitats wikipedia , lookup
Blue carbon wikipedia , lookup
Marine pollution wikipedia , lookup
Anoxic event wikipedia , lookup
Ocean acidification wikipedia , lookup
Effects of global warming on oceans wikipedia , lookup
Physical oceanography wikipedia , lookup
Ecosystem of the North Pacific Subtropical Gyre wikipedia , lookup
Background Section 1 Limnol. Oceanogr., 36(8), 1991, 1507-1526 0 1991, by the American Society of Limnology and Oceanography, Inc. Role of the marine biosphere in the global carbon cycle Alan R. Longhurst Biological Oceanography Division, Bedford Institute P.O. Box 1006, Dartmouth, Nova Scotia B2Y 4A2 of Oceanography, Department of Fisheries and Oceans, Abstract The geochemical disequilibrium of our planet is due mainly to carbon sequestration by marine organisms over geological time. Changes in atmospheric CO, during interglacial-glacial transitions require biological sequestration of carbon in the oceans. Nutrient-limited export flux from new production in surface waters is the key process in this sequestration. The most common model for export flux ignores potentially important nutrient sources and export mechanisms. Export flux occurs as a result of biological processes whose complexity appears not to bc accommodated by the principal classes of simulation models, this being especially true for food webs dominated by single-celled protists whose trophic function is more dispersed than among the multicelled metazoa. The fashionable question concerning a hypothetical “missing sink” for CO2 emissions is unanswerable because of imprecision in our knowledge of critical flux rates. This question also diverts attention from more relevant studies of how the biological pump may be perturbed by climatic consequences of CO, emissions. Under available scenarios for climate change, such responses may seem more likely to reinforce, rather than mitigate, the rate of increase of atmospheric CO,. There are other ways of saying it, but really the most urgent question for biological oceanography is whether the marine biosphere will mitigate or reinforce the increase of anthropogenic greenhouse gases in the atmosphere. This increase, as Fig. 1 shows, is an instantaneous step function on the time scale of natural changes of similar magnitude. A simple probe of the gross chemistry of the atmosphere, ocean, and sediments would reveal a state of geochemical disequilibrium. A reactive gas, oxygen, is abundant in both atmosphere and ocean, and carbon (with which it would combine on a lifeless but watery planet) is largely sequestered in Acknowledgments Though they won’t all agree with what I have written, I am particularly grateful to many colleagues and friends for spirited discussion of this problem and for providing me with charts for my voyage into unknown waters: especially Karl Banse, David Bird, Steve Calvert, John Cullen, Glen Harrison, Erica Head, Peter Jones, Paul Kepkay, Trevor Platt, and Jorge Sarmiento. biogenic carbonate formed below or along the margins of oceans. This geochemical disequilibrium must be maintained by biological processes against a background of geological episodes of subduction and vulcanism, during which C is cycled between biogenic carbonate rocks and the atmosphere (Owen and Rea 1985; Volk 1989b). Our probe might not sense the critical fact that the biologically mediated flux that maintains the geochemical disequilibrium is small compared with the transfer of CO2 across the ocean surface driven simply by diffusion and solubility. Figure 2 shows a new version of a familiar figure to make this point. We would surely see that the surface layer of the oceans is in equilibrium with the atmosphere, while the deeper water masses are oversaturated with dissolved CO,. From this we draw the basic concept of the “biological pump”: downward flux of C is maintained by gravitational settling of organic and carbonate C, fixed autotrophitally in the surface layer by plant cells. We 1507 Longhurst 1508 600 1 CO2 2050 + (atmosphere) 1990, 500 Glacial transition periods g 400 CL ..I lA 300 3 200 Most recent glacial period 100 Years -150000 Fig. 1. Estimates ofpCO,atm Pole data, with a projection to (-150,000 to +lOO) 100 for the last 150,000 yr from ice-core data and from modern Mauna Loa-South A.D. 2050 based on the present-day incremental rate. a.ssume that this flux is constrained by the supply of a limiting nutrient, usually supposed to be inorganic N as NO+ To predict the response of the ocean (and its biological pump) to a step function in the radiatively active gases in the atmosphere is difficult or even impossible, given our uncertainties about relevant processes and rates. Paradoxically, we are more confident about the role of marine biota in the remote past than in the present perturbation and, by constraining our predictions, this knowledge may help us to predict the role of marine biota in the immediate future (Berger et al. 1989). Riogeochemical evolution of the primitive atmosphere The CO,-rich atmosphere of the Proterozoic was progressively modified by the sequestration of C in carbonate rocks and organic sediments by stromatolite-building and other cyanobacteria (Rothschild and Mancinelli 1990) in shallow seas. By the early Cambrian, sufficient free oxygen was available both to oxidize reactive elements in the rocks (e.g. Fe to Fe oxides) and to produce a surplus sufficient for the evolution of protists and metazoa. The geochemical consequences of marine productivity dominate the geological record: terrestrial plants have endowed us with onl:y 3-7 x 1O3Gt C of coal, compared with 40 x lo6 Gt C of carbonate rocks from marine biota. However, the pCO,atm decline over the 60 x lo6 yr from the Cretaccous to the Pleistocene may be due to the activity of land plants. If the Cretaceous atmosphere resulted from active sea-floor spreading and vulcanism, and if Cretaceous oceanic circulation was as active as today (Manabe and Wetherald 1980), the subsequent decrease in pCO,atm may have been due to the evolution of the angiosperm forests (Volk 1989~). If, on the other hand, the C&-rich Cretaceous atmosphere originated in weak oceanic circulation and low plankton productivity, strengthening circulation and increasing marine productivity may have progressively reduced pCO,atm to Pleistocene levels. Role of the biosphere in Pleistocene glacial transitions-Because the rate of change in pCO,atm during Pleistocene glacial-interglacial transitions approaches our scale of interest, it may be profitable to examine how small changes in solar forcing (Berger et al. 1990) could be so amplified by changes in pCO,atm as to mediate transitions between glacial and interglacial climates (Genthon et al. 1987). Perhaps such a process might also amplify or reduce the effects of anthropogenic forcing of climate change in the coming centuries. 1509 Carbon cycle--&cdthe biota Flows as Gt C yr-1 Stocks as Gt C 0.3 f 0.3 0.3 - 11.0 & ii ATMOSPHERE 747 ( = 355 ppm) (annual 1 1 change +3.4 f 0.2 ppm) 54.3 f 9.5 LAND BIOTA I 639 (net 1.5 - 2.5) 80 f 23 1 + 186 FLUX L I DIC DOC - 0.2-0.5 DIC 1000 DOC 313 25.3-27.3 h f 3 PYCNOCLINE POC 7-9 PIG 0.75 f 0.3 OCEAN PRODUC ‘TIVITY OPTIONS INTERMEDIATE DEEP WATER AND DIC 37 Doc ?? DIC 36700 POC 4.7 DOC 1198 P/C 0.15 POC 0.04 Fig. 2. Summary of the global carbon budget, excluding carbonate and other rocks, modified and Bolin 1986. The levels of uncertainty indicated are mostly from the tabulations of Sundquist are explained in the text. Export flux is shown for different values for the global$ratio. A cornucopia of scenarios, having a rapid replacement rate, purports to explain such changes in pCO,atm and also the concomitant changes in terrestrial biosphere C: the sum of these two fluxes must have been absorbed by (or released from) the oceans during a glacial transition. It is generally supposed that the marine biosphere was in some way implicated (e.g. Barnola et al. 1987; Berger et al. 1987). At glacial onset, a flux of 93- 1,3 50 Gt C from terrestrial plants, litter, and soil oc- from Moore 1985; others curred (Faure 1990), to which must be added 2 11 Gt C from the atmosphere (to reduce it from 280 to 180 ppm), thus requiring a sink in the ocean of 304-l ,56 1 Gt C to maintain global balance. This amount is within observational constraints, for Sarnthein et al. (1988) computed a return of 580-720 Gt C from the ocean to the atmosphere during the transition to our own interglacial period. In recent years, thinking on these fluxes has evolved rapidly. From reliance on the 1510 Longhurzt relationship between ocean temperature and pCO,ocean (e.g. Broecker 1982), recent scenarios are more likely to invoke increased ocean productivity during glaciations to draw down pCO,atm. Candidate mechanisms have ranged from increased production of biogenic carbonate by corals (Berger 1982), through changes in the Redfield ratio of sinking material (Broecker and Pcng 1982; Broecker and Takahashi 1984) to increased nutrient supply from continental shelves exposed during glaciations (McElroy 19 8 3; Broecker and Peng 1984; Walsh 1988). Current thinking, however, converges on the effects of altered ocean circulation on the supply of nutrients to the photic zone, especially in low latitudes where most new production occurs (Barber and Chavez 1986), but perhaps not in the Arabian Sea monsoon region (Sarkar et al. 1990). Sediment cores suggest that during Pleistocene glaciations the ocean was characterized by stronger upwelling centers (Muller and Suess 1979; Boyle 1986; Mix 1989a), while subtropical gyres were more stratified and less productive (Sarnthein et al. 1987; Sarnthein and Winn 1990). Such data suggest 38% more productivity at low latitudes in the Atlantic (Mix 19893), as well as an increase in the tropical Pacific (Pedersen et al. 1988), for a significant increase globally (Sarnthein and Fenner 1988). Other scenarios have emphasized highlatitude processes. In the North Atlantic, as the polar front approached the mcteorological intertropical convergence zone, increased wind stress would favor pulsed production and export flux, resulting in a drawdown of pCO,atm. At the same time, formation of d.eep water masses may have ceased at high latitudes in the North Atlantic (Oppo and Fairbanks 1987), with consequences for Southern Ocean alkalinity and photic zone NO3 and Si03 (Sarmiento and Toggweiler 1984; Broecker and Peng 1989; Knox and McElroy 1984). It has also been suggested that plant growth in our present interglacial Southern Ocean may be Fe-limited because of low aeolian input of this element, a flux thought to have been greater during glaciations (Martin and Fitzwater 1988; Martin 1990) though sediment cores suggest reduced, rather than enhanced, export flux in Ant- arctic seas during glaciations (Boyle 1988; Mortlock et al. 199 1). The question of the role of Fe in the Southern Ocean is, of course, one of the central subjects of this symposium. Mechanism of the biological pump-For ’ simplicity, we might characterize the biological pump as having three components, each representing a different ecological process: a rotary pump which circulates material in the microplankton food web of the photic zone, an Archimedian pump by which flux of fecal and aggregated material occurs continually under gravity, and a reciprocating pump by which diel migrants actively carry material down at dawn, to rise again at dusk to feed. The usual concept of the biological pump does not accommodate the consequences of the complex interactions between plant growth and the alkalinity of seawater for flux of CO, across the sea surface. Further, it usually simply assumes that sinking flux of POC is a measure of the rate of N03d:riven production in the photic zone. Although this balance may be confirmed by 15N uptake measurements (e.g. Knauer et al. 1990), the simple model ignores active flux of DON and respiratory C by diel migrants, vertical advection and diffusion of JDOM, and variable organic-inorganic C and C : N : P ratios of the flux. These omissions have had an important impact on our thinking. Many (most?) biological oceanographers would probably still accept the simple paradigm that weakened equatorial upwelling should reduce the biological uptake of C and thus enhance the normal rate of CO, outgassing there. Conversely, most might be surprised at the concomitant concept that reduced upwelling of CO,-rich water would actually result in reduced outgassing of CO, from a warmerthan-normal surface water (e.g. Siegenthaler and Wenk 1984). Actually, it must be the balance between these two processes that cletermines flux across the sea surface. Nutrient sources and carbon sinks--In simple models, photosynthesis is the only autochthonous source of particulate C and the limiting nutrient is usually supplied in our models only by eddy flux from below, although diverse intermittent mechanisms (Gargett 199 1) should also be accommo- Carbon cvcle and the biota dated: wind- or buoyancy-driven winter overturn, seasonal upwelling driven by nonlinear local or remote forcing, cross-shelf eddy transfer (e.g. Houghton et al. 1987), event-scale wind mixing (Eppley and Renger 1988), vorticity induced by eddy shear (Pollard and Regier 1990), wind-stress curl, and isopycnal upwelling. Only by invoking all such mechanisms can we explain the spatial variability of sea-surface chlorophyll. The concept of a single source 0” fixed N may not be sustained much longer. The supply of fixed N by the cyanobacteria Trichodesmium is likely to be an important source of export flux, regionally perhaps comparable to NO,-driven production, and representing true net sequestration of CO, (Bird pers. comm.). Dense blooms of Trichodesmium are widespread in moderate latitudes and supply fixed N at rates significant relative to algal demand and sometimes more rapidly than NO3 is supplied by vertical mixing (Carpenter et al. unpubl.; Karl unpubl.). Further, the endosymbiotic cyanophyte Richelia releases fixed N to sustain diatom blooms in central gyral waters (Venrick 1974) and may also be a significant source of N. Also usually ignored is atmospheric deposition, especially leeward of continental masses. On the eastern continental shelf of North America, this may contribute 20-30% of biologically available N (Paerl et al. 1990). Although Legendre and Gosselin (1989) recently calculated that aeolian input of NH3 in the North Atlantic could exceed upward transport of NO, by a factor of three or more, this has now been shown to be incorrect, and - 10% is more likely (Laws 199 1; Galloway et al. 1985). Nor can we ignore isopycnal advection of properties, since variability of plankton abundance in the California Current (and perhaps other eastern boundary currents) is principally determined by the variable transport of nutrient-rich water from higher latitudes, not by local coastal upwelling (Bernal and McGowan 198 1). The simple concept of the biological pump does not acknowledge vertical advection of dissolved organic matter. If the inverse, linear relationship between NO3 and DON in the upper kilometer observed by Sugimura and Suzuki (1988) is correct, then downward mixing of DON (and 1511 DOC) must be an important component of the export flux from the photic zone (Toggweiler 1989). We also need to know how much of the total annual particle flux for each ocean basin comprises seasonal or intermittent blooms which may sink almost directly to the sediments. Such flux is caused by rapid growth of diatoms outstripping the capacity of herbivore consumption (Riemann 19 8 9; Lampitt 19 8 5; Prezelin and Alldredge 19 8 3; Takahashi 1986; Nair et al. 1989) or else by the rapid development of salp populations capable of consuming the standing crop of small cells and producing rapidly sinking fecal material (Bathmann 1988; Wiebe et al. 1979). Finally, we must separate flux from micro- and mesoplanktonic food webs (Michaels and Silver 1988). Microplankton produce relatively small fecal material, including abundant minipellets from protists (Gowing and Silver 1985) with low rates of sinking, and even more abundant submicron particles from flagellates (Koike et al. 1990). On the other hand, mesoplankton produce larger, faster sinking fecal material and also interzonal diel migrants may actively transport material downward (Longhurst and Harrison 1989). Balance between micro- and mesoplankton consumption of primary production Classically, mesoplankton were thought to dominate grazing of phytoplankton, but actually they rarely consume more than a small fraction of daily primary production. Since the relative size of phytoplankton cells is progressively smaller offshore and toward low latitudes, grazers must follow the same trend. Mesoplankton consumption in oceanic situations is almost always ~20% of daily production (Tsuda et al. 1989). In intermittent upwelling systems, as off Peru, open-ocean consumption by mesozooplankton may be that low (3-8%, Dagg et al. 1980). During the JGOFS 1989 North Atlantic spring bloom study, mesoplankton removed only l-2% of the daily primary production (Morales et al. 199 1). Of course, at the end of any algal bloom, consumption may briefly exceed production. In shallow shelf seas, mesozooplankton consumption is somewhat higher than in the open ocean, usually lo-40% of daily 1512 Longhu& production (Dagg et al. 1982; Chervin et al. 198 1; Pace et al. 1987; Longhurst 1983) but may approach a balance with primary production, as in Puget Sound (60-70% daily, Welschmeyer and Lorenzen 1985a, b), or even exceed it, as in summer in the Celtic Sea (Joint and Williams 198 5). Oceanic microplankton, on the other hand, frequently consume 70-90% of daily primary production (Welschmeyer and Lorenzen 1985a). In coastal and high latitudes, experiments return quantities usually between 10 and 50% of standing stock or 25-100% of potential production daily (Gifford 1988; Paranjape 1987, 1990; Burkill et al. 1987). Role of mesoplankton in particulate exportflux- If mesoplankton are not the principal consumers we thought in the past, what is the importance of their apparently fastsinking fecal material in oceanic export flux? Can we relate this material directly to new production as the biological pump concept requires (e.g. Michaels and Silver 1988), since mesoplankton fecal pellets will usually not represent more than a small fraction of the total consumption in the overlying photic zone? The concept that sinking fecal pellets contribute to vertical flux dates from Lohmann (1902) and is confirmed by many trap observations (e.g. Lorenzen and Welschmeyer 1983; Sasaki et al. unpubl.), but some anomalous observations compel our attention: crustacean fecal pellets having sinking rates that should take them to the bottom may instead be consumed or broken up by other organisms while held in suspension by vertical turbulence in coastal areas (Smetacek 1980); there is also a remarkably close similarity between open-ocean profiles of pellets and of copepods, suggesting little or no vertical pellet flux (Krause 198 1; Bathmann et al. 1987). A similar case can be made for profiles of crustacean exuviae (Sameoto 1986; Longhurst et al. 1984). Depth proJiles ofPOCproduction and consumption -The biological pump is usually modeled as flux between two boxes: a surface photosynthetic and a deep aphotic box where sinking C is respired and NO3 regenerated. Whether this geochemical shorthand is supportable can be determined only from a knowledge of the profiles of light, temperature, nutrients, primary production, heterotrophic respiration, and sources of POC. Because profile characteristics are determined principally by physical processes, we can generalize plankton profiles for regions of the ocean having characteristic profiles for density (Fig. 3). Depths of maximum algal and growth rate coincide when these are shallow but are progressively separated in profiles where the maxima are deeper. In such cases, the depth of maximum productivity is shallower than the chlorophyll maximum, which lies at or just above the nutricline (Cullen 1982; Longhurst and Harrison 1989). Heterotrophic bacterial biomass follows algal biomass and productivity, reflecting the utilization of DOM released directly from plant cells and heterotrophs (Sherr and Sherr 1988). :New observations of oceanic DOM profiles (Sugimura and Suzuki 1988) show steep, biologically driven gradients to -500 m, raising the possibility of a flux-driven connection between the “zooplankton remineralization loop” discussed by Banse (1990a) and the microbial loop of the photic zone. Fr,om the depth distribution of trophic groups of mesoplankton we can conclude at what depths fecal POC, excretory DOC, and respiratory C are released and at what depths diRerent fractions of POC are consumed (Fig. 4); results are consistent with most partic:le flux from the photic zone being consumed and remineralized within the subsequent 200-300 m (e.g. Martin et al. 1987; Bishop et al. 1986). Active export flux by vertical migrantsDiel migrations, both within the photic zone and between the photic zone and the mesopelagial, have been well described so we can predict which genera will participate in each kind of migration (Fig. 5). Diel migration in coastal and seasonal situations can be: considered modifications of the openocean paradigm. Why diel migration should occur was a Holy Grail for planktologists for many decades. Only recently have we begun to ask, rather, what are its consequences for vertical flux of material? (e.g. Longhurst et al. 1989, 1990; Longhurst and Harrison 1988, 1989). Interzonal migrants feed in the photic zone at night and rest in colder, dark water by day at 300-800 m. 1513 Carbon cycle and the biota BIDSTAT - E. Trop. Pacific 1.0 Chl a , INO3 R I 2.5 zoo 2.0 ,F EASTROPAC ,Chla , ,NO3 , - 20”s , , ,NO3 zoo It-L3 2.0 , 0 zw DAY % GROUPS 50 NIGHT 10 0 % GROUPS 50 100 40 1.0, Southern Ocean Chl 02 21 4.0 Costa Rica Dome 1 .o , ,Chla ,NO3 a 45”s , 0.0 , 2.8 200 II I 4.0 I Y BIOMASS, NIGHT Fig. 4. Relative contribution of functional feeding groups to the day and night zooplankton biomass profile in the eastern tropical Pacific Ocean (GP-gelatinous predators; RP-raptorial predators; H-herbivores; O-omnivores; D and DETR-detritivores; FILT - filter-feeders; PRED - predators). Fig. 3. Cartoons to show characteristic profiles of chlorophyll, NO3 primary production (Pt), and zooplankton biomass in relation to the structure of the water column, particularly comparing several HNLC regions having significant mixed-layer nitrate (Chl a as mg m--j; NO, as pg-atms liter-l; Pt as mg C m-3 d-l. Aphotic zone, shaded; thermocline, striped). The guts of copepods, which are usually -70% of these migrants, are emptied prior to, or early in, their descent. The gut contents of only the larger nekton are carried down with them. Both plankton and nekton migrants release respiratory C and excretory NH4 at depth, usually at 400-600 m. Their feeding pattern ensures that this dissolved material originates in the surface layers. At seven stations in low latitudes (Longhurst et al. 1990), the mean flux of respiratory C from diel migration was equivalent to -30% of sinking flux across the thermocline. From the global biomass of zooplankton in regions where diel migration is significant, respiratory flux is 0.14-0.27 Gt Cyr-I, whichisg-17%ofSundquist’s(1985) estimate for total sinking flux of C at 200 m between these latitudes, extrapolated from Betzer et al. (1984). To this must be added the flux from gut contents of nekton and the loss of C (by respiration and predation) from populations of deep over-wintering oceanic copepods. Work in progress in the North Atlantic suggests that two-thirds of the migrating biomass do not survive the winter and represent active transport from the photic zone to depths of -600-800 m. The comparable flux of metabolic nitrogen as NH4 downward out of the photic zone as a result of diel migration requires upward revision of the f-ratio (relating NO,-driven to Longhurst 25 g-175 c zQ) 200 P Cd 225 275 300 325 r 3 STYLOUiElFlON THYSANOESSA NEMATOBRACHION NEMATOSCELIS ~cmJsocALANus CALOCAIANUS MK=YI\IDCERA PARACALANUS TEMoRoPlA ACARTIA ACROCAIANUS PSEUDOCALANUS OITHONA NANNOCAIANUS UNDINUIA, ETC ’ 350 375 Fig. 5. Depth distribution of trophic groups, day and night, excluding interzonal migrants. From the left, in sequence, the groups are: cpipelagic euphausiids, small Calanoid copepods, a large herbivorous Calanoid and pteropods, large epipelagic Calanoid copepods, large mesopelagic Calanoid copepods, an interzonal migrant Calanoid and, on the right, ostracods. (Dl, D2-top, bottom of thermocline; SCM and shading-subsurface chlorophyll maximum.) 4) total production) of from 5 to 65%; the relation between N flux due to diel migrants and sinking particle flux is positively correlated with the depth of the mixed layer (Longhurst et al. 1989), the ratio being 10% for mixed layers 40-50 m deep and 20% at 1 lo-120 m. Balance between production-consurnption: Excess nutrients-If we are to understand high N03, low chlorophyll (HNLC) regions, we must be able to quantify the balance between photosynthesis and consumption, both regionally and seasonally. .Seasonally pulsed production of phyto- plankton, exceeding the consuming capacity of herbivores (e.g. Billet et al. 1983), is easily observed, but herbivore constraint on the numerical growth of plant cells is harder to demonstrate. In grazing-constrained HNLC situations, we might expect to find unusually high pheopigment : chlorophyll ratios (Thomas 1979), especially where metazoan grazers are abundant. Although the dynamits of pheopigments in the water column are: complex (reingestion, photodegradation), these processes should not differ signiftcantly in normal and in HNLC regions. The HNLC region of the eastern tropical Carbon cycle and the biota 1515 Pacific lies consistently to the south of the must ask to what extent their output is really equator in the EASTROPAC data, but me- reliable. The basic assumption for sizescaled models is that material passes in a soplankton biomass, chlorophyll, and pheopigments all take their highest values along linear fashion from small to large biota and or to the north of the equator. The Pheo: *for network analysis that biota can be arranged into a few compartments, comprisChl a ratio is close to unity over the whole ing subsets of zooplankton, phytoplankton, region (Thomas 1979; Beers and Stewart 197 l), suggesting that the NO3 anomaly does and bacteria. The consequences of modeling not result from mesoplankton grazing pres- all metazoan biota as just a few interacting sure. Further, a meridional section across compartments are difficult enough to prethe region (Chavez et al. 1990) shows no dict, even though functional feeding groups can be identified for mesoplankton, as we correlation between microheterotroph abundance and surface NO,; an earlier me- have discussed above. But what of the proridional section of Protist abundance (Beers tist plankton, the principal photic zone consumers? Can we treat them as reducible to and Stewart 197 1) does show higher abuna few feeding groups like metazoa? dance at some stations south of the equator, Network models: Dispersed autotrophy and but this is not a clear result. Thus, neither meso- nor microheterotroph abundance is heterotrophy among protists -Although bacteria and protists are the major comunequivocally associated with high nearsurface NO3 in available observational data, ponent of respiration in the ocean, we are so suggestions that the anomaly is due to only now learning that to compartmentalize grazing constraint (e.g. Toggweiler 1990) are their trophic functions involves uncertainnot supported by the EASTROPAC data. ties that are much greater than for metazoa The same conclusion was reached by because the functions of photosynthesis, Thomas (1979) for the region where superpredation, and heterotrophy are dispersed ficial HNLC water is advected from the disamong all Protist groups, and many individual cells frequently perform more than tant coastal upwelling region, rather than originating in equatorial upwelling. one of these functions. It is now apparent that all groups of functionally photosynModels for C flux analysis thetic “algae” include forms (sometimes Two generalizations are available for abundant) that also attack and consume models of the marine biosphere: by simple other protists: diatoms appear to be the only network analysis to a few interacting biotic exceptional group. Conversely, most groups compartments (e.g. Volk and Hoffert 1985; of “heterotrophic” protists, with the posFasham et al. 1990), and by allometric ar- sible exception of tintinnids, also include guments to size-scaled linear biomass spec- abundant forms which harbor photosyntra (e.g. Platt 1985; Echevarria et al. 1990). thetic symbiotic algal cells or photosynSuch models are capable of delivering outthetic plastids. put within observational constraints, alThus, the classical distinction between though it is difficult to be sure that they do protozoa and algae is no longer tenable so for the right reasons, and the arguments (Sleigh 199 l), and we should abandon this that follow suggest that most currently used misleading classification. However, our models actually simulate trophic functions ecological interpretation of function among dispersed among taxa, rather than the purprotists must be sensitive to the Great Diported function-specific taxonomic units. vide between procaryotes (metabolically diBecause such models seem to work, we must verse and adaptable to changing ambient assume that our generalizations about biochemistry through DNA recombination by logical processes are sufficiently robust to plasmid exchange) and eucaryotes (strucdeliver realistic output despite inappropriturally diverse but metabolically less flexiate assumptions concerning the nature of ble through DNA exchange at gametogenthe simulated compartments. esis). Nevertheless, these two main classes of The autotrophic symbionts accommomodels require simplifications that cannot dated by “protozoan” taxa whose dimenbe supported by ecological analysis and we sions permit such a relationship are diverse: 1516 Longhurst cyanobacteria, chrysomonads, prymnesiomonads, cryptomonads, and dinoflagellates (Taylor 1987). To further complicate the issue, heterotrophic dinoflagellates may themselves harbor symbionts, using a variety of mechanisms to do so. Symbionts may or may not dominate the physiology of the host by an exchange of molecules partly satisfying the nutrient demands of both symbiont and host. The opposite case, of chlorophyll-bearing, “phytoplanktonic” protists that regularly consume other living cells, is also widespread. The dinoflagellate genus Ceratium is a classical component of autumnal blooms and of the oceanic phytoplankton, yet includes species capable of consuming diatoms larger than themselves, with a variety of engulfing mechanisms (Taylor 1989). The same is true for the small, abundant genus Protoperidinium, which is capable of daily consuming > 20 large diatom cells to support a growth rate of 1.7 divisions daily (Jacobsen 1987). Many genera of nanophytoflagellates (including coccolithophores) ingest bacteria by a variety of techniques. These chlorophyllbearing flagellates are among the principal consumers of bacteria in lakes (Bird and Kalff 1986) and in the marine pelagial (D. F. Bird pers. comm.). In the Sargasso Sea, phagotrophic (mixotrophic) nanoflagellates may constitute 50% of all photosynthetic nanoplankton and contribute significantly to consumption of bacteria (Caron and Lim unpubl.). Our knowledge of predatory “protophytes” and photosynthetic “protozoa” is quite insufficient to generalize their effect on our measurements of (for example) heterotroph respiration or size-fractioned photosynthesis. Plastid-bearing ciliates in both neritic and oceanic environments may represent significant proportions of large cell photosynthesis, reaching -25% in both temperate and Arctic environments. Even so, because of their fragility on polycarbonate filters, the ciliate fraction of the largecell component of size-fractioned photosynthesis may be underestimated (Putt 1990). The functional complexity of the pelagic protistan ecosystem approaches the spatial and functional complexity of coral reef organisms within their carbonate matrix. The Protist ecosystem exists in a high viscosity matrix which permits spatial complexity through the development of microzones within which single Protist cells interact chemically with ambient organic and inorganic molecules. Their symbionts and the bacteria associated with their cell wall share th.eir microzone and participate in its chemistry. The complexity of this part of the biosphere stems also from the interaction within. it between the metabolically labile procaryotes, the metabolically complex eucaryotes, and the structurally diverse small metazoa (Longhurst 1990). Attempts to simplify and compartmentalize even the major trophic relationships in this protistan ecosystem result in complex networks (Fig. 6), which recall the unstructured food-web model of Isaacs (1972). As discussed by Banse (1990b), the complexity of metazoan food webs may be more apparent than real, but it is perhaps in the protistan food web, if anywhere, that we can expect to find a lack of biomagnification because of the complexity of trophic links. Biomass spectrum models: Variability of predator-prey ratios -Though size-scaled plankton biomass spectra usually have a slope close to that predicted by allometric models, it is only recently that they have been made to accommodate events such as loss from the linear biomass spectrum, which occurs during a spring bloom when autotrophic production overwhelms the capacity for consumption of the relevant larger size classes and algal cells sink unconsumed (e.g. Rodriguez et al. 1987). Further assumptions concerning the relative sizes of predators and prey are inherent in linear biomass models. A prey-predator linear ratio approaching what is often (but probably incorrectly) assumed for terrestrial ecosystems is also proposed for marine biota (e.g. Fenchel 19 8 8). Available generalizations suffer the same defect as classical terrestrial models, only more acutely: this ratio fits chase-and-grab predators (hawks-sparrows, fish-shrimps) but other important, feeding relationships (caterpillars-trees, salps-picoplankton) fall far outside the 0.05-O. 15 range usually assumed, with many B 1.O. Extending -an earlier analysis (Longhurst 11390) a review of > 100 pairs of prey and ISa V ‘_._.._ __..:._.:_. ._ _‘.... : .:.., ::.:.y j : v VIRUSES 2.0 pm 0.2 mm PARTICULATE ORGANIC MATTER 20 pm 2.0 mm Fig. 6. Size-structured compartment flux formulation based on discussions at a NATO Advanced Studies Institute in 1988. To the original has been added the flux of submicron particles from flagellate feeding (Koike et al. 1990) and also the role of marine viruses in lysing bacteria during algal blooms (Bratbak et al. 1990; Proctor and Fuhrman 1990). HEIEFIOTROPHS .:. _:__: :. .> ::.... I .:_ MIXOTFWHS + LOMOLwr DOM HIMOLWT DOM lEzz!l AUTOTROPHS ( H _ / j 1 Longhurst 1518 61 4 PROTISTA CETACEA o 2 0 -2 -4 -6 -6 10 -10-c -8 1 -4 -6 -2 y = 0.626x - 2.13, r2 = 0.551 2 0 6 4 6 10 In(x) OF PREDATOR LENGTH (mm) loo g2 E E E E 5 : PROTISTA CETACEA 10. 0 0 O0 1 -0 *,.O 0 0 O 0 .Ol - &” 0 0 0 0 0 0 0 f8’ 0 -0 00 ocp ’ O: Q ““Og O o. 0 0 4 0” 00 0 0:o”Q?a 0 OeO 0 -0 .OOl - 0 0 0 0 0 l.OOOE-4 -8 . -6 0 -4 -2 0 7 4 6 8 -----TO In(x) OF PREDATOR LEiNGTH ratio for 100 cases from the Fig. 7. The feeding ratio (in longest dimensions) shown as a prey/predator marine ecosystem, including all major taxa. Standard regression of ratio and predator length-scaled ratio scaled by predator length shown. predators (from picoplankton to whales) in the pelagic ecosystem, selected to cover all kinds of feeding mechanisms, gives the results shown in Fig. 7. Prey really are usually smaller than predators, but the linear ratio is 0.65 (+ 1.62!) in these selected samples. The high variance suggests that we should place little reliance on the actual value of this ratio, and it is unlikely that increasing the data base would reduce the scatter. For metazoa, the ratio covers five orders of magni tude; high values ( 1.2-9.09) occur among protists, copepods, ctenophores, heteropods, fish, and cetacea, while the lowest values (0.0001-0.001) occur in fish, salps, and pteropods. Taxonomic position (except for protists) and relative abundance do not appear capable of predicting the ratio-even in the most approximate manner. Su.ch results suggest that strict linearity in biomass spectra must be to some extent an illusion. Because such models can have realistic output, we can only assume that there is internal balance between the consequences of shortcuts up the size-structured food chain and reverse flow back along it. Response of the biological pump to forcing at d@erent time scales After this short excursion into our uncertainty about the mechanism of the biological pump and how to model it, we can return to the question of how the ocean, including its biota, may mitigate or reinforce the rate of increase of greenhouse gases in the atmosphere. More fashionably (e.g. Moore and Bolin 19&6), this question is of- Carbon cycle and the biota 1519 geochemical budgets require, most are of an ten posed as “what part of the CO, emissions has the ocean-already assimilated?” or accuracy that would cause them to be re“where is the missing CO, sink?” It is not jected by an experimental scientist. From Fig. 2 we could select values to support consurprising that answers are contradictory tradictory conclusions concerning the given the low accuracy and precision of our amount of the imbalance (or even its exisestimates of the relevant fluxes, although tence) and the sink(s) to which the missing this problem often seems not to be recogCO, must have gone. As it happens, Detnized. The quest for the missing CO, sink de- wiler and Hall ( 198 8) have already done this: using high values for oceanic uptake rives from our inability to balance presentand low values for fossil fuel and forest day sources and sinks. Increases in the airclearance, they showed that it is possible to borne fraction of CO, emissions and estimates of oceanic uptake fall short by - 1.2 balance the budget within 0.3 Gt C yr-‘, Gt C yr-’ of the canonical estimates of 6.3 which is a negligible quantity. If we reverse their argument and use high emissions and Gt C yr-’ emitted from all anthropogenic sources (Broecker et al. 1979; Detwiler and low ocean uptake, the missing CO, sink can Hall 1988). The atmospheric increase is, of become 2.3 Gt C yr- l. Updating their argument and using the range of estimates now course, based on measurements with rather small error terms, but this is not the case available for flux from the terrestrial biosphere (e.g. Houghton et al. 1987; Woodwell for ocean uptake. Computation of flux into et al. 1978), we can obtain an excess of the ocean is based on box diffusion models, sources over sinks ranging from 0.2 to 13.1 parameterized by uptake rates for passive Gt C yr-l . In view of such great uncertaintracers into the ocean (radon, tritium, bombties, strong justification is needed to work 14C), assumptions concerning ocean ventilation rates, transfer velocities of CO, across with computations based on values chosen the surface, and rates of vertical mixing (e.g. simply because they lie centrally in the scatBroecker et al. 1979; Oeschger et al. 1975); ter of estimates. A recent study (Tans et al. 1990) illuserror estimates are usually not given for such trates this difficulty. A global data base of models, which converge on oceanic uptake of 1.5-2.5 Gt C yr- l. With median values, oceanic mixed-layer pC0, is used together with assumptions of transfer velocity of CO2 then, - 1.2 Gt C yr- 1 cannot be accounted for and represents the missing COZ. across the sea surface to suggest that observed differences in atmospheric pC0, beBalance could be obtained if the terrestrial biosphere was a net sink of C (Tans et tween the northern and southern hemial. 1990; Siegenthaler and Oeschger 1987). sphere require a mid-northern latitude sink However, terrestrial ecologists concur that of 1.5-2.0 Gt C yr-’ for which it is suggested there is presently a net flux of CO, from the that the terrestrial biosphere is the best canterrestrial biosphere to the atmosphere at a didate. No statistics are provided for the median rate of 3.25 Gt C yr-’ (Woodwell pCO,ocean data base, but it is known that et al. 1978; Houghton et al. 1987; Detwiler this is highly variable on the scale of days and Hall 1988). That this result is contrary to weeks; in fact, because this variability is to what is apparently required to balance driven by growth of phytoplankton, pC0, global geochemical models is acknowledged is probably at least as patchy as the chlo(e.g. Houghton et al. 1987; Mellilo et al. rophyll field seen in satellite imagery. Re1990), although no convincing solution to gional computations of pC0, in the mixed this paradox has been advanced. These cal- layer, therefore, can be no better than the culations accommodate soil weathering, notoriously unreliable regional estimates of forest clearance and regrowth, and growth chlorophyll or primary production based on stimulation by increasing atmospheric ship data. Even supposing that the transfer pCO,, for which there is little direct evivelocity assumptions used in this study are dence at present (Detwiler and Hall 1988; correct, the level of uncertainty of the asKohlmaier et al. 1987). sociated pCO,ocean data base seems quite Of the many estimates for flux terms that inadequate with which to challenge the can- 1520 Longhurst on of terrestrial ecology concerning the sign of C flux with the atmosphere. What are we to make of all this? Probably we should admit that the fashionable question (where has the apparently unaccounted-for CO2 gone to?) is not very profitable, as are global geochemical models in which interplay across the range of estimates for each flux is not permitted. Unless the flux from the terrestrial biosphere is really at the very lowest end of all current estimates, we should admit the possibility that our estimate of ocean uptake may be far too low. New thinking on this flux is appropriate. Since most scenarios for change in pCO,atm during glacial-interglacial transitions require some perturbation of the biological processes of the photic zone, the first point that we ought to explore is whether the biological pump is presently in steady state (if, indeed, a steady state can be envisaged given the lack of concordance of the time scales relevant to the rotary, the arch.imedean, and the reciprocating pumps). And, secondly, how do we expect it to be perturbed as climate changes induced by greenhouse gases begin to manifest themselves? The significance of these to points is easily overlooked. Unless we overturn our basic concept of phytoplankton limitation by inorganic nutrients and accept some form of C limitation, the biological pump cannot respond to increasing pCOZ in surface waters unless nutrient limitation in the photic zone is relaxed. Only if we think that this (or the opposite case) has already occurred, because of changing global weather and circulation patterns, could we claim that the marine biosphere has already modified the rate of increase of atmospheric COz. Is the biological pump already perturbed?- Fisheries scientists need knowledge of interannual variation in fish stocks, so information on decadal-scale variability in ocean circulation, and the distribution of biota is available for many parts of the ocean. It is certainly premature to conclude that any of the trends seen in these data reflect the effects of increased greenhouse gases, but they do indicate how rapidly ocean biota can respond to changes in wind patterns resembling those likely to occur as radiative effects strengthen. The most profound quasi-periodic event in ocean circulation is El Niiio/Southem Oscillation (ENSO). During the 1982-l 983 event, pCO,atm ceased to increase, only to rebound immediately afterward to above the accustomed 1.5 ppm yr-l (Volk 1989a; Keeling and Revelle 1985). This global perturbation must have involved transfers between anomalous sources and sinkschanges which interacted to produce the observed signal in the atmosphere. The reduced upwelling and deeper mixed layer over the eastern tropical Pacific Ocean during the 1982-1983 ENS0 caused a reduction of primary production to 29% in the equatorial upwelling, normally responsible for 20-50% of global new production, and a reduction to 1O”h of the normal area of the Peru-Chile upwelling region (Barber and Chavez 1986; Walsh 1988). Such a decrease in export flux might be expected to result in increased outgassing of CO, at the anomalously warm sea surface (Siegenthaler 1990). However, measurements (Feeley et al. 1987) and simulations (Volk 1989a) suggest that during 1982 the equatorial ocean ceased to be a net source for CO2 because weaker upwelling reduced the near-surface supply of CO,-rich deeper water. Weaker upwelling and reduced plant growth interacted to determine transfer rates of CO2 across the sea surface. Since the effect of ENS0 drought conditions reduces plant growth in monsoon regions, the observed changes in the atmospheric pC0, during the 19 82- 19 8 3 ENS0 requires major intermittent sinks during an ENS0 event in higher latitude oceans or in the nontropical terrestrial biosphere. Events on a longer time scale may also have important effects on biological production. Decadal increases in wind strength and wave height in the northern hemisphere ma.y explain decadal increase in chlorophyll values in the North Pacific gyre through increased vertical transport of nutrients (Venriclk et al. 1987; Carter and Draper 1988). Anomalous and sustained northerly winds in the northeast Atlantic resulted in a decadal-scale surface salinity anomaly that was advected oceanwide in the thermohaline circulation. This event has been invoked to explain changes in the abundance and tim- Carbon cycle and the biota ing of diatom blooms and a reduction by half of high latitude primary production due to increased mixed-layer stability during its passage (Cushing 1990). Potential responsesof the biological pump to an anthropogenic greenhouse climateOur predictions for the response of the marine biosphere to a changed climate can be no better than the available predictions for how ocean circulation will change with the climate. The main outlines, though not regional details, of changes in circulation have been predicted (Manabe and Stouffer 1980): weaker meridional thermal gradients will decrease polar ice cover, decrease the strength of the tradewinds and-at least during a period of transition-strengthen the flow of low-salinity surface water from the polar regions. In some situations, increased thermal gradients may develop between continental mass and adjacent ocean so that coastal upwelling may be enhanced (e.g. Bakun 1990). It is difficult to predict the balance of the various biological responses to these changes, which may be of opposite sign in low and high latitudes. Can we get any help by reference to the response of the marine biosphere to changes in wind-stress patterns so as to mediate glacial transitions? Can we assume that after a period of rapid and perhaps chaotic response to the present anomalous rate of increase of atmospheric pCO,, the biogeochemical balance will attain a quasi-steady state resembling the much slower rate of change at a glacial transition? Output from a two-dimensional model of Baes and Killough (1986) appears to confirm, in fact, that biological uptake of C by the marine biosphere is capable of important modification of atmospheric CO2 levels: an ocean with no biological uptake induces a doubling of atmospheric COZ, and a doubling of nutrient supply to the phytoplankton almost halves atmospheric COz. Much smaller changes in atmospheric CO, are induced by halved or doubled circulation rates in the absence of biological processes. There are also suggestions concerning other potential changes likely to occur if the anthropogenic greenhouse effect strengthens. Ice cover in the Arctic decreased - 5% between 1979 and 1987 (recalling the con- 1521 current global retreat of glaciers) and a further decrease of the magnitude suggested by climate scenarios is computed to increase levels of primary production over presently ice-covered Arctic shelf areas by an order of magnitude, or to about - 1 Gt C yr-’ (Walsh 1989). This process, by sequestering C to the sediments, would reduce the rate of increase in atmospheric pC0, levels and may already have started to take effect, if observations of reduced Arctic ice cover in recent years are correct. On the other hand, any oceanic primary production processes presently driven by wind stress at the sea surface and consequent renewal of photic zone nutrients (equatorial upwelling, event-scale mixing, and perhaps spring blooms driven by winter overturn) will become weaker and reduce the present rate of C sequestration: the socalled phytoplankton accelerator effect in relation to increasing pCO,atm. However, as noted above, there is evidence that in recent decades, when the first effects of increased greenhouse gases may be expected to have taken effect, these processes may have become stronger rather than weaker in midhigh latitudes of the northern hemisphere. Conclusions Several conclusions can now be drawn concerning our original question: will the marine biosphere mitigate or reinforce the rate of increase of anthropogenic pCO,atm? Our first conclusion must be that the output from formal biogeochemical models deployed for this purpose should be accorded less deference than has become customary, given our extreme uncertainty about the rates of almost all critical fluxes and even of the size of the stocks between which flux occurs. Where such models do not accommodate reasonable error terms for each flux and stock, their out.put should be regarded with great caution. The second conclusion concerns the paradox that highly aggregated simulation models of biochemical processes in the pelagial should give output within observational constraints, although their formulation departs so strikingly from our emerging understanding of flow within protistan food 1522 Longhurst webs. Perhaps the complexity of simulated processes is, in reality, lost as noise in the overall signal representing the fluxes simulated, or perhaps, since we know approximately what constraints to place on outputs, we may unconsciously parameterize the models accordingly. Output may be correct, but if network flow is incorrect, predicted responses to novel conditions may be fallacious. Thus, a way must be found to subsume the dispersion of function in procaryotes and eucaryotes (still very largely undescribed, but becoming more complex with each new finding) into a few interacting compartments, representing dispersed fLmclions rather than groups of taxa, as is the current practice. If we do not defer to geochemical models to answer our question, can we obtain any help from scenarios of the role of biota in pCO,atm changes during glacial transitions? Can these scenarios be a guide to biotic response to other global perturbations? ILf this is so, and if the changed greenhouse climate is to be dominated by a reduced meridional thermal gradient (reduced meridional wind stress), then we can at least guess what the sign of the biotic response will be. It is more likely that C uptake by lthemarine biosphere will weaken rather than strengthen and consequently will enhance the rate of increase of pCO,atm rather than mitigating the effect. Hopefully, in the foreseeable future, it will be possible to upgrade this guess (through the step of being a plausible scenario) to a predictive model on which we can place some confidence. References BAES, C. F., JR., AND G. G. KILLOUGH. 1986. Changes and biological processes in CO, ocean models, p. 329-347. In J. R. Trabalka and D. E. Reichle [eds.], The changing carbon cycle. Springer. BAKUN, A. 1990. Global climate change and intensification of coastal ocean upwelling. Science 247: 198-201. BANSE, K. 1990a. New views on the degradation and disposition of organic particles as collected by sediment traps in the open sea. Deep-Sea Rcs. 37: 1177-l 195. .-. 1990b. On pelagic food web interactions in large water bodies, p. 556-579. In M. M. Tilzer and C. Serruya [eds.], Large lakes-ecological structure and function. Springer. BARBER, R. T., AND F. P. CHAVEZ. 1986. Ocean variability in relation to living resources during the 1982-83 El Nifio. Nature 319: 279-285. BARNOLA, J. M., D. RAYNAIJD, Y. S. KOROTKEVICH, AND C. LORIUS. 1987. Vostok ice core provides 160,000-year record of atmospheric CO,. Nature 329: 408-4 13. BATHMANN, U. V. 1988. Mass occurrence of Salpa fusiformis in the spring of 1984 off Ireland: Implications for sedimentation processes. Mar. Biol. 97: 127-135. --, T. T. NOJI, M. Voss, AND R. PEINERT. 1987. Copepod faecal pellets: Abundance, sedimentation and content at a permanent station in the Norwegian Sea. Mar. Ecol. Prog. Ser. 38: 45-5 1. BEERS,J. R., AND G. L. STEWART. 197 1. Micro-zooplankters in the plankton communities of the upper waters of the eastern tropical Pacific. DeepSea Res. 18: 861-883. BERGER, A., H. GALLI?E, T. FICHEFET, I. MARSIAT, AND C. TRICOT. 1990. Testing the astronomical theory with a coupled climate-ice sheet model. Palaeogeogr. Palaeoclimatol. Palaeoecol. 89: 125-14 1. BERGER, W. H. 1982. Increase of carbon dioxide in the atmosphere during deglaciation: The coral reef hypothesis. Naturwissenschaften 69: 87-88. --, K. FISCHER, C. LAI, AND G. WV. 1987. Ocean productivity and organic carbon flux. Part 1. Overview and maps of primary production and export production. Univ. Calif. San Diego. SIO Ref. 8730. --, V. S. SMETACEK, AND G. WEFER. 1989. Ocean productivity and palaeoproductivity-an overview, p. 1-31. In W. H. Berger et al. teds.], Productivity of the oceans. Wiley. BER.NAL, P. A., AND J. A. MCGOWAN. 1981. Advection and upwelling in the California Current, p. 381-399. Zn F. A. Richards [ed.], Coastal upwelling. Am. Geophys. Union. BETZER, P. R., AND OTHERS. 1984. Primary productivity and particle fluxes on a transect of the equator at 153W in the Pacific Ocean. Deep-Sea Res. 31: l-11. BILLET, D. S. M., R. S. LAMPITT, A. L. RIUCE, AND R. F. C. MANTOURA. 1983. Seasonal sedimentation of phytoplankton to the deep-sea benthos. Nature 302: 520-522. BIRD, D. F., AND J. KALFF. 1986. Bacterial grazing by planktonic lake algae. Science 231: 493-495. BISHOP, J. K. B., J. C. STEPIEN,AND P. H. WIEBE. 1986. Particulate matter distributions, chemistry and flux in the Panama Basin: Response to environmental forcing. Prog. Oceanogr. 17: l-59. BO’U’LE,E. A. 1986. Deep circulation, preformed nutrients and atmospheric CO,: Theories and evidence from oceanic sediments, p. 49-59. In K. J. Hsii [ed.], Mesozoic and Cenozoic oceans. Am. Geophys. Union. 1988. Cadmium: Chemical tracer of deep--. water paleoceanography. Paleoceanography 3: 47 l489. BR~TBAK, (i., M. HELDAL, S. NORLAND, AND T. F. THINGSTAD. 1990. Viruses as partners in spring bloom microbial trophodynamics. Appl. Environ. Microbial. 56: 1400-1405. Carbon cycle amndthe biota BROECKER,W. S. 1982. Glacial to interglacial changes in ocean chemistry. Prog. Oceanogr. 11: 15 l-l 97. -, AND T.-H. PENG. 1982. Tracers in the sea. Eldigio. -, AND 1984. The climate-chemistry connection. Gedphys. Monogr. 29, p. 327-336. -, AND . 1989. The cause of the glacial to interglacial atmospheric CO, change: A polar alkalinity hypothesis. Global Biogeochem. Cycles 3: 2 15-239. -, AND T. TAKAHASHI. 1984. Is there a tie between atmospheric CO, content and ocean circulation? Geophys. Monogr. 29, p. 314-326. -, H. J. SIMPSON,AND T.-H. PENG. 1979. Fate of fossil fuel carbon dioxide and the global carbon budget. Science 206: 409-4 18. BURKILL, P. H., R. F. C. MANTOURA, C. A. LLEWELLYN, AND N. J. P. OWENS. 1987. Microplankton grazing and selectivity of phytoplankton in coastal waters. Mar. Biol. 93: 581-590. CARTER, D. J. T., AND L. DRAPER. 1988. Has the north-east Atlantic become rougher? Nature 332: 494. CHAVEZ, F. P., K. R. BUCK, AND R. T. BARBER. 1990. Phytoplankton taxa in relation to primary production in the equatorial Pacific. Deep-Sea Res. 37: 1733-1752. CHERVIN, B., T. C. MALONE, AND P. J. NEALE. 198 1. Interactions between suspended organic matter and copepod grazing in the plume of the Hudson River. Estuarine Coastal Shelf Sci. 13: 169-183. CULLEN, J. J. 1982. The deep chlorophyll maximum: Comparing profiles of chlorophyll-a. Can. J. Fish. Aquat. Sci. 39: 791-803. CUSHING, D. H. 1990. Plankton production and yearclass strength in fish populations: An update of the match/mismatch hypothesis. Adv. Mar. Biol. 26: 249-293. DAGG, M., AND OTHERS. 1980. Grazing and excretion by zooplankton in the Peru upwelling system during April 1977. Deep-Sea Res. 27: 43-59. -, J. VIDAL, T. E. WHITLEDGE, R, L. IVERSON, AND J. J. GOERING. 1982. The feeding, respiration and excretion of zooplankton in the Bering Sea during a spring bloom. Deep-Sea Res. 29: 4563. DETWILER, R. P., AND C. A. S. HALL. 1988. Tropical forests and the global carbon cycle. Science 239: 42-47. ECHEVARR~A, F., AND OTHERS. 1990. The size-abundance distribution and taxonomic composition of plankton in an oligotrophic, high mountain lake (La Caldera, Sierra Nevada, Spain). J. Plankton Res. 12: 4 15-422. EPPLEY, R. W., AND E. H. RENGER. 1988. Nanomolar increase in surface layer nitrate concentration following a small wind event. Deep-Sea Res. 35: 11191125. FASHAM, M. J. R., H. W. DUCKLOW, AND S. M. MCKELVIE. 1990. A nitrogen-based model of plankton dynamics in the oceanic mixed layer. J. Mar. Res. 48: 591-639. FAURE, H. 1990. Changes in the global continental reservoir of carbon. Palaeogeogr. Palaeoclimatol. Palaeoecol. 82: 47-52. 1523 FEELEY, R. A., AND OTHERS. 1987. Distribution of chemical tracers in the eastern equatorial Pacific Ocean during and after the 1982-83 El Nifiol Southern Oscillation event. J. Geophys. Res. 92: 654516558. FENCHEL, T. 1988. Ecology: Concepts and limitations. Ecol. Inst. GALLOWAY, J. N., R. J. CHARLSON, M. 0. ANDRE& AND H. RODHE [EDs.]. 1985. The biogeochemical cycling of sulfur and nitrogen in the remote atmosphere. Reidel. GARGETT, A. E. 199 1. Physical processes and the maintenance of nutrient-rich euphotic zones. Limnol. Oceanogr. 36: 1527-l 545. GENTHON, C., AND OTHERS. 1987. Vostok ice core: Climatic response to CO, and orbital forcing changes over the last climatic cycle. Nature 329: 414-418. GIFFORD, D. J. 1988. Impact of grazing by microzooplankton in the northwest arm of Halifax Harbour, Nova Scotia. Mar. Ecol. Prog. Ser. 47: 249258. GOWING, M. M., AND M. W. SILVER. 1985. Minipellets: A new and abundant size class of marine fecal pellets. J. Mar. Res. 43: 395-418. HOUGHTON, R. A., AND OTHERS. 1987. The flux of carbon from terrestrial ecosystems to the atmosphere in 1980 due to changes in land use: Geographical distribution and global flux. Tellus 39B: 122-139. ISAACS, J. 1972. Unstructured food webs, and “pollutant analogies.” Fish. Bull. 70: 1053-1059. JACOBSEN, D. M. 1987. The ecology and feeding of thecate dinoflagellates. Ph.D. thesis, Woods Hole Oceanogr. Inst. WHOI-87- 10. JOINT, I. R., AND R. WILLIAMS. 1985. Demands of the herbivore community on phytoplankton production in the Celtic Sea in August. Mar. Biol. 87: 297-306. KEELING, C. D., AND R. REVELLE. 1985. Effects of El Niiio/Southern Oscillation on the atmospheric content of carbon dioxide. Meteoritics 20: 437450. KNAUER, G. A., D. G. REDALJE, W. G. HARRISON, AND D. M. KARL. 1990. New production at the VERTEX time-series site. Deep-Sea Res. 37: 11211134. KNOX, F., AND M. B. MCELROY. 1984. Changes in atmospheric CO,: Influence of the marine biota at high latitude. J. Geophys. Res. 89: 4629-4637. KOHLMAIER, G. H., H. BRQL, E. 0. SIR& M. PL~CHL, AND R. REVELLE. 1987. Modelling stimulation of plants and ecosystem response to present levels of excess atmospheric CO,. Tellus 39B: 155-l 62. KOIKE, I., S. HARA, K. TERAUCHI, AND K. KOGURE. 1990. Role of sub-micrometre particles in the ocean. Nature 345: 242-244. KRAUSE, M. 198 1. Vertical distribution of faecal pellets during FLEX ‘76. Helgol. Meeresunters. 34: 313-327. LAMPITT, R. S. 1985. Evidence for the seasonal deposition of detritus to the deep-sea floor and its subsequent resuspension. Deep-Sea Res. 32: 885897. 1524 Longhurst LAWS, E. A. 1991. Photosynthetic quotients in the open ocean. Deep-Sea Res. 38: 143-167. LEGENDRE, L., AND M. GOSSELIN. 1989. New production and export of organic matter to the deep ocean: Consequences of some recent discoveries. Limnol. Oceanogr. 34: 1374-l 380. LOHMANN, H. 1902. Die Coccolithophoridae. Arch. Protistenk. 1: 89-l 65. LONGHURST, A. R. 1983. Benthic-pelagic coupling and export of organic carbon from a tropical Atlantic continental shelf-Sierra Leone. Estuarine Coastal Shelf Sci. 17: 26 l-285. -1990. Pelagic ecology: Definition of pathways for material and energy flux, p. 263-288. Zn M. M. Denis [ed.], Oceanologie, actualite et prospective. Proc. Centennial Sta. Endoume, Marseille. -, AND OTHERS. 1989. NFLUX: A test of vertical nitrogen flux by diel migrant biota. Deep-Sea Res. 36: 1705-1719. ---, A. W. BEDO, W. G. HARRISON, E. J. H. HEAD, AND D. D. SAMEOTO. 1990. Vertical flux of respiratory carbon by oceanic diel migrant biota. Deep-Sea Res. 37: 685-694. --, AND W. G. HARRISON. 1988. Vertical nitrogen flux from the oceanic photic zone by diel migrant zooplankton and nekton. Deep-Sea Res. 35: 88 l-889. -‘-, AND -. 1989. The biological pump: Profiles of plankton production and consumption in the upper ocean. Prog. Oceanogr. 22: 47-123. ---, D. SAMEOTO, AND A. HERMAN. 1984. Vertical distribution of Arctic zooplankton in summer: Eastern Canadian archipelago. J. Plankton Res. 6: 137-168. LORENZEN, C. J., AND N. A. WELSCHMEYER. 1983. The in situ sinking rates of herbivore fecal pellets. J. Plankton Res. 5: 929-933. MCELROY, M. B. 1983. Marine biological controls on atmospheric CO, and climate. Nature 302: 328329. MANABE, S., AND R. 1. STOUFFER. 1980. Sensitivity of a global climate change model to an increase of CO, in the atmosphere. J. Geophys. Res. 85: 55295554. --, AND R. T. WETHERALD. 1980. On the distribution of climate change resulting from an increase in CO, content ofthe atmosphere. J. Atmos. Sci. 37: 99-l 18. MARTIN, J. H. 1990. Glacial-interglacial CO, change: The iron hypothesis. Palcoceanography 5: l-l 3. -., AND S. E. FITZWATER. 1988. Iron deficiency limits phytoplankton growth in the north-east Pacific subarctic. Nature 331: 34 l-343. ----, G. A. KNAUER, D. M. KARL, AND W. W. BROENKOW. 1987. VERTEX: Carbon cycling in the northeast Pacific. Deep-Sea Res. 34: 267-285. MELLILO, J. M., T. V. CALLAGHAN, F. I. WOODWARD, E. SALATI, AND S. K. SINHA. 1990. Effects on ecosystems, p. 283-310. Zn J. T. Houghton et al. [eds.], Climate change-the IPCC scientific assessment. Cambridge. MICHAEL& A. F., AND M. SILVER. 1988. Primary production, sinking fluxes and the microbial food web. Deep-Sea Res. 35: 473-490. MIX, A. C. 1989a. Pleistocene productivity: Evi- dence from organic carbon and foraminifera1 species, p. 313-340. Zn W. H. Berger et al. [eds.], Productivity of the oceans-past and present. Wiley. --, 1989b. Influence of productivity variations on long-term atmospheric CO,. Nature 337: 54 l544. MOC)RE, B., AND B. BOLIN. 1986. The oceans, carbon dioxide and global climate change. Oceanus 30: 9-l 5. MORALES, C., A. BEDO, R. P. HARRIS, AND P. R. G. ‘TRANTER. 1991. Grazing of copepod assemblages in the north-east Atlantic: The importance lofthe small size fraction. J. Plankton Res. 13: 455,472. MOR.TLOCK, R. A., AND OTHERS. 199 1. Evidence for lower productivity in the Antarctic Ocean during the last glaciation. Nature 351: 220-223. MUL.LER, P. J., AND E. SUESS. 1979. Productivity, sedimentation rate and sedimentary organic matter in the oceans. 1. Organic carbon preservation. Deep-Sea Res. 26: 1347-1362. NAIF!, R. R., AND OTHERS. 1989. Increased particle flux to the deep ocean related to monsoons. Nature ,338: 749-75 1. OESC:HGER,H., U. SIEGENTHALER, U. SCHO’ITERER,AND .4. GUGELMANN. 1975. A box diffusion model to study the carbon dioxide exchange in nature. Telhs 27: 168-192. OPP~, D. W., AND R. G. FAIRBANKS. 1987. Variability in the deep and intermediate water circulation ofthe Atlantic Ocean during the past 25,000 Iyears: Northern hemisphere modulation of the Southern Ocean. Earth Planet. Sci. Lett. 86: l-l 5. OWEN, R. W., AND D. K. REA. 1985. Sea-floor hydrothermal activity links climate to tectonics: The ‘Eocene carbon dioxide greenhouse. Science 227: 166-169. PACE), M. L., G. A. KNAUER, D. M. KARL, AND J. H. MARTIN. 1987. Primary production, new production and vertical flux in the eastern Pacific Ocean. Nature 325: 803-804. PAERL, H. W., J. RUDEK, AND M. A. MULLIN. 1990. Stimulation ofphytoplankton production in coastal -waters by natural rainfall inputs. Mar. Biol. 107: 247-254. PARANJAPE, M. A. 1987. Grazing by microzooplankton in the eastern Canadian arctic in summer 1983. Mar. Ecol. Prog. Ser. 40: 239-246. --. 1990. Microzooplankton herbivory on the IGrand Bank: A seasonal study. Mar. Biol. 107: 321-328. PEDEIRSEN,T. F., M. PICKERING, J. S. VOGEL, J. N. SOUTHON, AND D. E. NELSON. 1988. The re:sponse of benthic foraminifera to productivity cycles in the eastern equatorial Pacific: Faunal and geochemical constraints on glacial bottom water Ioxygen levels. Paleoceanography 3: 15 7- 168. PLAIT, T. 1985. Structure of the marine ecosystem: its allometric basis. Can. Bull. Fish. Aquat. Sci. 213, p. 55-64. POLI,ARD, R. T., AND L. REGIER. 1990. Large variations in potential vorticity at small spatial scales in the upper ocean. Nature 348: 227-229. PRÉZELIN, B. A., AND A. ALLDREDGE. 1983. Primary Carbon cycle and the biota production of marine snow during and after an upwelling event. Limnol. Oceanogr. 28: 11561167. PROCTOR, L. M., AND J. A. FUHRMAN. 1990. Viral mortality of marine bacteria and cyanobacteria. Nature 343: 60-62. PUTT, M. 1990. Abundance, chlorophyll content and photosynthetic rates of ciliates in the Nordic seas during summer. Deep-Sea Res. 37: 17 13-1732. RIEMANN, F. 1989. Gelatinous phytoplankton detritus aggregates on the Atlantic deep-sea bed. Structure and mode of formation. Mar. Biol. 100: 533539. RODRIGUEZ, J., F. JIMBNEZ, B. BAUTISTA, AND V. RODRIGUEZ. 1987. Planktonic biomass spectra dynamics during a winter production pulse in Mediterranean coastal waters. J. Plankton Res. 9: 1183-l 194. ROTHSCHILD, L. J., AND R. L. MANCINELLI. 1990. Model of carbon fixation in microbial mats from 3,500 Myr ago to the present. Nature 345: 7 lo712. SAMEOTO, D. D. 1986. Influence of the biological and physical environments on the vertical distribution of mesoplankton and micronekton in the tropical Pacific. Mar. Biol. 93: 263-279. SARKAR, A., R. RAMESH, S. K. BHATTACHARYA, AND G. RAJAGOPALAN. 1990. Oxygen isotope evidence for a stronger winter monsoon current during the last glaciation. Nature 343: 549-55 1. SARMIENTO, J. L., AND J. R. TOGGWEILER. 1984. A new model for the role of the oceans in determining atmospheric PCoz. Nature 308: 621-624. SARNTHEIN, M., AND J. FENNER. 1988. Global windinduced change of deep-sea sediment budgets, new ocean production and CO, reservoirs. Phil. Trans. R. Sot. Lond. Ser. B 318: 487-504. -, AND K. WINN. 1990. Reconstruction of low and middle latitude export productivity, 30,000 years BP to present: Implications for global carbon reservoirs, p. 3 19-342. Zn M. E. Schlesinger [ed.], Climate-ocean interaction. Kluwer. -, J.-C. DUPLESSY, AND M. R. FONTUGNE. 1988. Global variations of surface ocean productivity in low and mid latitudes: Influence on CO, reservoirs of the deep ocean and atmosphere during the last 2 1,000 years. Paleoceanography 3: 36 l399. SHERR, E., AND B. SHERR. 1988. Role of microbes in pelagic food webs: A revised concept. Limnol. Oceanogr. 33: 1225-l 227. SIEGENTHALER, U. 1990. El Niiio and atmospheric CO,. Nature 345: 295-296. -, AND H. OESCHGER. 1987. Biospheric CO, emissions during the past 200 years by deconvolution of ice-core data. Tellus 39B: 140-l 54. AND T. WENK. 1984. Rapid atmospheric CO, vaiiations and ocean circulation, Nature 308: 624626. SLEIGH, M. A. 1991. A taxonomic review of heterotrophic protists important in marine ecology, p. 9-38. Zn P. C. Reid et al. [eds.], Protozoa and their role in marine processes. Springer. SMETACEK, V. S. 1980. Zooplankton standing stock, copepod faecal pellets and particulate detritus in 1525 Kiel Bight. Estuarine Coastal Mar. Sci. 11: 477490. SUGIMURA, Y., AND Y. SUZUKI. 1988. A high temperature catalytic oxidation method for the determination of non-volatile dissolved organic carbon in seawater by direct injection of a liquid sample. Mar. Chem. 24: 105-131. SUNDQUIST, E. T. 1985. Geological perspectives on carbon dioxide and the carbon cycle. Geophys. Monogr. 32, p. 5-59. TAKAHASHI, K. 1986. Seasonal fluxes of pelagic diatoms in the subarctic Pacific, 1982-1983. DeepSea Res. 33: 1225-125 1. TANS, P. P., I. Y. FUNG, AND T. TAKAHASHI. 1990. Observational constraints on the global atmospheric CO, budget. Science 247: 143 1-1438. TAYLOR, F. J. R. 1987. Heterotrophic nutrition, p. 225-267. Zn F. J. R. Taylor [ed.], The biology of dinoflagellates. Bot. Monogr. 2 1. Blackwell. TAYLOR, G. T. 1989. Variability in the vertical flux of micro-organisms and biogenic material in the epipelagic zone of a North Pacific central gyre station. Deep-Sea Res. 36: 1287-l 308. THOMAS, W. H. 1979. Anomalous nutrient-chlorophyll interrelationships in the offshore eastern tropical Pacific Ocean, J. Mar. Res. 37: 327-335. TOGGWEILER, J. R. 1989. Is the downward DOM flux important in carbon transport?, p. 84-107. Zn W. H. Berger et al. [eds.], Productivity of the oceans: Past and present. Wiley. 1990. Modeling workshop offers first look at ne’w simulation of equatorial Pacific. U.S. JGOFS News 2(2): 1, 11. TSUDA, A., K. FURUYA, AND T. NEMOTO. 1989. Feeding of micro- and macrozooplankton at the subsurface chlorophyll maximum in the subtropical North Pacific. J. Exp. Mar. Biol. Ecol. 132: 41-52. VENRICK, E. L. 1974. The distribution and significance of Richelia intracelIularis in the North Pacific central gyre. Limnol. Oceanogr. 19: 437-445. -, J. A. MCGOWAN, D. R. CAYAN, AND T. L. HAYWARD. 1987. Climate and chlorophyll a: Long-term trends in the central North Pacific Ocean. Science 238: 70-72. VOLK, T. 1989a. Effect of the equatorial Pacific upwelling on atmospheric CO, during the 1982-83 El Niiio. Global Biogeochem. Cycles 3: 267-279. -. 19893. Sensitivity of climate and atmospheric CO, to deep-ocean and shallow-ocean carbonate burial. Nature 337: 637-640. 1989~. Rise of angiosperms as a factor in long-term climatic cooling. Geology 17: 107-l 10. -, AND M. I. HOFFERT. 1985. Ocean carbon pumps: Analysis of relative strengths and efficiencies in ocean-driven atmospheric CO, changes. Geophys. Monogr. 32, p. 99-l 10. WALSH, J. J. 1988. On the nature of continental shelves. Academic. 1989. Arctic carbon sinks: Present and future. Global Biogeochem. Cycles 3: 393-4 11. WELSCHMEYER, N. A., AND C. J. LORENZEN. 1985a. Chlorophyll budgets: Zooplankton grazing and phytoplankton growth in a temperate fjord and the central Pacific gyres. Limnol. Oceanogr. 30: l21. 1526 -, Longhurst AND -. 198%. Role of herbivory in controlling phytoplankton abundance: Annual pigment budget for a temperate marine fjord. Mar. Biol. 90: 75-86. WIEBE, P. H., L. P. MADIN, L. R. HAURY, AND G. R. HARBISON. 1979. Diel vertical migration by Snl- pa aspersa and its potential for large-scale particulate organic matter transport to the deep-sea. Mar. Biol. 53: 249-255. WOODWELL, G. M., AND OTHERS. 1978. The biota and the world carbon budget. Science 199: 14 l146.