| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Journal of Plankton Research Vol.25 no.9 pp.1021-1034, 2003
© Oxford University Press 2003
Structural–functional relationships in the pelagic community of the eastern tropical Atlantic Ocean
1 MSRC, State University of New York at Stony Brook, NY 11794-5000, USA, 2 Plymouth Marine Laboratory, Prospect Place, West Hoe, Plymouth PL1 3DH, UK and 3 Institute of Biology of the Southern Seas, Sevastopol 99011, Ukraine
* Corresponding Author: spiontkovski{at}notes.cc.sunysb.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Sampling was carried out in the eastern tropical Atlantic during July–September, 1987 on four transects at latitudes 6, 12, 18 and 24°W, from 6°S to 4°N. A total number of 81 stations were occupied where physical, chemical and biological measurements were taken. Temperature, salinity, chlorophyll a, phytoplankton, primary production (PP), bacterial production, micro-, meso-, macrozooplankton and mesopelagic fish and squid were sampled in the upper 120 m layer. The biomass size spectrum of the plankton community was relatively flat within the divergence zone but the slope of the size spectrum increased towards the convergence zone meaning an increased contribution of the small-sized organisms. Higher turnover rates (the ratio of PP to the total community biomass) were observed in regions where the size spectra sloped more steeply. PP exceeded the community metabolism by three to five times in regions where the size spectra were sharply sloped. In regions with flat-type spectra (with slopes from -0.2 to 0) a balance between PP and community metabolism was observed. The variation coefficient and the dispersion index (the variance to mean ratio) indicated an increase of the spatial variability of biomass in the sequence: ‘phytoplankton–mesozooplankton–macrozooplankton–micronekton’. At the higher trophic level of the zooplankton community, between the biomass of squid and gelatinous zooplankton, a logarithmic linear relationship was derived. Squid actively consumed gelatinous organisms in regions of high concentrations of their prey, although large copepods, euphausiids, amphipods, flying fish, and myctophid fish also contributed to their diet.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
The eastern part of the tropical Atlantic Ocean is known to be a region of high biological productivity surrounded by areas of relatively low productivity. The high productivity of the pelagic community provides the food for the large stocks of sardinella and tuna, which play key roles in the fishery (Shiohama et al., 1965
; Binet, 1997). This region has been the focus of research for several national and international expeditions over the last decades (Corcoran and Mahnken, 1969; Greze, 1971; Hizardet al., 1977; Le Borgne, 1977; Herbland and Voiturez, 1979; Herbland et al., 1983; Postel, 1990; Morel, 1996), although the majority have not investigated the whole pelagic community. These investigations were primarily concerned with phyto- and mesozooplankton studies. From this point of view, the three Ukrainian expeditions, in 1981, 1985 and 1987, are perhaps among the most comprehensive for this region (Zuyev et al., 1988a,b). However, the data from these cruises were published in Russian and the results are not widely known.
In this paper, we utilize the data collected on the expedition of R/V ‘Professor Vodyanitsky’ (July–September, 1987), which were based on several macroscale meridional transects of oceanographic stations, with simultaneous physical, chemical and biological sampling (Zuyev et al., 1990). This strategy is especially effective in equatorial regions, where a complicated interlacing of macroscale divergence and convergence zones and currents take place. Using data from this expedition we have attempted to analyse changes of the major integrative structural and functional characteristics of the pelagic community in the macroscale convergence and divergence zones and to quantify links between these characteristics. Under the terms integrative structural and functional characteristics we refer to the size spectrum of a community, primary and secondary production, community biomass and metabolism, and to the ratios between certain of these characteristics.
Elements of the integrative (holistic) approach in studies of complex biological systems, in particular the aquatic ecosystems, have been monitored for decades (Elton, 1927; Lindeman, 1942; Odum, 1971; Mann, 1982; Gomes and Haedrich, 1992; Sprules and Goyke, 1994; Gaedke, 1995). This approach became widely accepted due to the simplicity of monitoring general trends in the structure and functioning of marine eco-systems, in particular, the relationship between size structure and production (or stocks) at the higher trophic levels (Moloney and Field, 1985; Haedrich, 1986; Mann, 1988). Integrative estimates can play a key role when structural and functional parameters of ecosystems from different biogeographic provinces are compared. Indeed, a statement of necessity to monitor the key ecosystem variables can be found in a number of widely accepted concepts of biological partitioning of the ocean, such as biogeo-chemical provinces (Platt and Sathyendranath, 1988; Longhurst et al., 1995), the Large Marine Ecosystems concept (Sherman et al., 1990, 1993; Constanza, 1992; Sherman and Solow, 1992), and their synthesis (Paulyet al., 2000).
|
METHOD |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Field survey
Sampling was carried out from July to September 1987 on four transects along 6, 12, 18 and 24°W, from 6°S to 4°N. The total number of stations with physical, chemical and biological measurements was 81. The spatial resolution between stations was 50 km (6, 12, 18°W) or 90 km (24°W) within each of four meridional transects (Figure 1). However, only stations with complete sets of concurrent measurements were selected for further analysis for the current work. This limited the selection of stations to those conducted at night, because the majority of hauls and observations on abundance of surface myctophids and squid were taken during the night hours. This decreased the number of stations (subjected to principal components analysis) by 25. On the other hand, by selecting only ‘night stations’, we have removed the variability caused by the diel rhythms, which are pronounced in nearly all groups of pelagic communities in the tropical Atlantic Ocean (Greze, 1971; Piontkovski and Williams, 1995).
|
divergence zones, 
convergence zones.Vertical profiles of temperature, salinity and water density were obtained to depths of 1500 m, with a spatial resolution of 1 m using the submersible CTD ‘ISTOK-5’. Appropriate authors in previous publications (Zuyevet al., 1990
) gave detailed descriptions of the methods used to sample phyto-, zooplankton and micronekton and to measure functional characteristics. We provide brief information on these methods. Structural characteristics
Biomass measurements of various groups of organisms were selected over several size intervals to analyse the community size spectra (Table I). Phytoplankton biomass values were obtained by a direct count of cells in aliquots extracted from bottle samples collected from 4 to 5 depths within the photic layer. Due to raw estimates used, there were no corrections for the shape, to derive biomass values for the community size spectra.
|
Mesozooplankton samples were collected by Juday Nets (145 µm mesh, 80 cm diameter) towed vertically, from 100 m to the surface at a speed of 0.7–1.0 m min-1. Usually, 50–75 m3 of water was filtered during the collection of each sample. Two types of trawls collected macrozooplankton. The BMN trawl had a 1 m2 square mouth with a mesh size of 750 µm. Tows were taken from 120 m to the surface taking ~20–30 min. The RMT trawl (40 m2 square mouth) was used simultaneously with BMN sampling, to obtain more abundant material for studies of macrozooplankton size structure and mesopelagic fish.
To analyse size spectra at the level of a single group, macrozooplankton crustaceans from the RMT trawl were chosen as an example. Organisms were ranked over the following size classes: 10–15, 15–20, 20–25, 25–30, 30–35, 35–40, 40–45, 45–50 and 50–55 mm. The higher boundary of each size class was used to plot the biomass size spectra over the meridional transects.
Squid were counted visually at the surface at the night stations during the 1–2 h drift of the vessel. Individuals were counted inside the 5 m light zone created by a halogen lamp (1 kW) on the sea surface. Jigger rods and manual nets were used to catch squid for weight and stomach content analysis.
Functional characteristics
Bacteria production was measured using isotopes of 3H (thymidine) and 14C. Water from 3 to 8 depths in the upper 150 m layer was sequentially filtered through ‘Synpor’ filters with mesh sizes of 2.5, 0.4 and 0.23 µm. Samples were stored in 250 ml in ambient conditions. Primary production (PP) was measured using the 14C isotope, in samples collected from 6 to 7 depths within the euphotic zone and incubated for 4–5 h in conditions similar to those in situ. Neutral light filters were used to imitate the natural photon flux density.
The production (P) and the metabolism (R) of microzooplankton, mesozooplankton, macrozooplankton and micronekton groups of organisms were calculated using appropriate and earlier published equations, derived from experiments carried out in tropical regions. Characteristics of the specific production and the organism’s energy equivalents used in calculations of functional characteristics are given in Table II. Due to the lack of direct measurements, the bacterioplankton metabolism was assumed to be equal to double the value of PP, which is similar to that in open ocean communities of the tropical Pacific Ocean (Vinogradov and Shushkina, 1987).
|
Statistical analysis
PCA and regression analyses were used to investigate the characteristics of community size spectra and its links with the structural–functional characteristics. Linear regressions were used in the size spectra approximations (as a function of body length or equivalent spherical diameter) and log-transformation was applied.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Hydrology of the region
Currents of the eastern tropical Atlantic form a system of interacting latitudinally oriented geostrophical fluxes of alternative directions (Figure 1
). The summer period, the time of data collection, is characterized by a seasonal uplift of the thermocline, accompanied by general cooling of the mixed layer (Philander and Pacanowski, 1981; Houghton, 1989).
The northern, central and southern limbs of the South Equatorial Current (the South Passat Current in Russian literature) dominate the transport of the western water mass during this period. The southern limb is the most powerful and occupies a large area to the south of 2°S. This limb, in fact, is an extension of the Benguela Current and transports the eastern tropical waters to the west (Zuyev et al., 1990). The northern and central limbs of the South Equatorial Current enter the Bay of Guinea and ensure the western transport of the equatorial water mass. The three limbs of the South Equatorial Current have a total width of over 1000 km and provide a water mass transport of ~92 sverdrups, which is comparable with the Gulf Stream intensity (Sokolov, 1977). The South Equatorial Current is more pronounced from summer to autumn.
The transport of the eastern water masses to the region is represented by the Interpassat Counter Current, the Equatorial Counter Current (the Lomonosov Current) and the South Equatorial Counter Current (Southern Underpassat Counter Current in Russian literature). Within the eastern water mass transport the Lomonosov Current is the most developed. It transports with an intensity from 15 to 50 sverdrups. The current is spread over the upper 200 m layer and has a typical width from 200 to 350 km and a velocity from 75 to 200 cm s-1 (Kort, 1975; Hanaichenko, 1980).
The system of interacting geostrophic currents and a system of the Passat (trade) winds form several macroscale divergence and convergence zones. The northern and southern divergence zones are well developed on the edges of the Equatorial Counter Current. The convergence zone separates them over 1°S, from 12 to 24°W, approximately. The convergence effects are well pronounced in the temperature and salinity fields at depths from 50 to 150 m. The divergence zone at 4–5°S is, perhaps, caused by the macroscale south equatorial cyclonic gyre, due to the western curl of the South Equatorial Current and their interaction with the Equatorial Counter Current. Several other less pronounced zones were also noted (Figure 1) and their description can be found in the literature (Belevich, 1975; Zuyev et al., 1990).
Taxonomic structure of pelagic community
Copepods were the dominant mesozooplankton group accounting for 197 species in the studied region (Zuyevet al., 1990). However, only 23 of these species contributed to 75% of mesozooplankton abundance. The list of abundant species included: Calanus minor, Undinula vulgaris, Rhincalanus nasutus, Calocalanus pavo, Calocalanus pavoninus, Delius nudus, Clausocalanus furcatus, Acartia clausi, Scolecithrix brady, Scolecithrix dana, Pleuromamma abdominalis, Pleuromamma gracilis, Lucicutia flavicornis, Heterorhabdus papilliger, Euchaeta marina, Oithona plumifera, Oncaea venusta, Oncaea mediterranea, Oncaea conifera, Oncaea media, Corycaeus speciosus, Microsetella rosea and Macrosetella gracilis. These species were noted throughout the studied area.
The macrozooplankton group was represented by 126 species. Crustaceans, siphonophores and tunicates contributed up to 65% of the species diversity (Zuyevet al., 1990). Cephalopods (29 species), euphausiids (21 species) and amphipods (26 species) were the other major groups of macrozooplankton. About 94% of the species, such as Diacria trispinosa, Liocranchia reinhardti, Salpa fusiformis, Euphausia tenera, Euphausia americana, Sergestes atlanticus and Pyrosoma atlantica were widespread oceanic tropical forms. About 82% of macrozooplankton species were noted as rare, and only six species (such as E. tenera, S. fusiformis, L. reinhardti, D. trispinosa, Onychoteuthis banksi and Pterigioteuthis gemata) could be considered as abundant species. The macrozooplankton crustaceans accounted for 66 species: 21 of these were euphausiids, 17 decapods, 26 amphipods, 1 mysid and 1 stomatopod. Among euphausiids, Thysanopoda tricuspidata, E. americana, E. tenera and E. gibba were noted throughout the area studied. The pelagic Sergestidae, Peneidae, Oplophoridae, Pandalidae and the larval stages of crabs, shrimps and lobsters represented decapods. Shrimps of genera Sergestes and Gennadas were dominant forms in the upper 100 m layer. Siphonophores were represented by eight species of genera Chelophyes, Bassia, Abylopsis, Abyla, Dyphyes, Eudoxoides and Hippopodius. Cephalopods were represented mainly by species with broad latitudinal trans-oceanic distributions, such as L. reinhardi, O. banksi, Abraliopsis atlantica and species with narrow tropical distributions (Sthenoteuthis pteropus and Ornithoteuthis antillarum).
Micronekton fish were represented by 135 species from 14 orders (Zuyev et al., 1990). An enhanced species diversity was observed among the mesopelagic Myctophidae (37 species), Melanostomatidae (10 species), Paralepididae (10 species), Gonostomatidae and Astronestidae (6 species each). About 80% of fish fauna were species typical for the global tropical ocean: Vinciguerria nimbaria, Gonostoma elongatum, Benthosema suborbitale and Myctophum nitidulum. The nektonic squid were represented mainly by three species: Thysanoteuthis rhombus, O. banksi and Sthenoteuthis pteropus. The latter one was dominant, by numerical abundance and total biomass.
Peculiarities of macroscale spatial distribution
The typical values and macroscale variability of the physical and biological characteristics are represented as appropriate statistical estimates for the four meridional transects in Table III. These data should not be considered from the point of view of the zonal changes (for example from the west to the east), because the parameter values are averaged over currents with different orientation. Data are useful to monitor the amplitudes of changes on a scale of the whole eastern region. It should be noted that the variation coefficients are very high, particularly for the biomass of micronekton organisms. Thus, spatial changes of biomass of the gelatinous plankton achieve two orders of magnitude even when averaging has been applied along the macroscale transects.
|
Community size spectrum
The size spectra of the community biomass differ in regions of macroscale divergence and convergence zones (Figure 2). There is a tendency for spectra to be relatively flat within the first zone but the spectrum slope increases from the divergence to convergence zone. This implies that the biomass of organisms, expressed in energy units, diminishes in proportion with the organism size in the convergence zones.
|
The averaged size spectra have the following regression approximations. In the divergence zones, Y = 1.08 - 0.06X (P = 0.32); in the convergence zones, Y = 1.40 - 0.30X, (P = 0.06), where Y is the logarithm of biomass (kilocal m-2), and X is the logarithm of the size of organisms (mm). For the zooplankton crustaceans, 1 kilocal. = 80 mg C (Vinogradov and Shushkina, 1987
). The average size spectra exhibited no differences within the Equatorial Counter Current and the South Equatorial Counter Current and therefore, were approximated by the same regression: Y = 1.42 - 0.19X (P = 0.05). The differences in the convergence/divergence slopes of spectra are understandable when the total zoo-plankton biomass composition is taken into consideration: the averaged spectra have greater variability in themacrozooplankton size range. Within this group, which is represented mainly by crustaceans, siphonophores, doliolids and small mesopelagic fish, the gelatinous zooplankton displays considerable peaks of biomass in the divergence regions. For example, in the south subequatorial divergence region (4–5°S) the biomass of gelatinous zooplankton exceeded 1000 mg m-3 in the 24°W transect and 2000 mg m-3 in the 18°W transect. This was two orders of magnitude higher than that of the background biomass outside the divergence zones.
Fish inhabiting the surface layer at night were also numerous. They were represented by the vertically migrant species, such as Vinciguerria nimbaria, Myctophum nitidulum, Myctophum asperum, Myctophum affine and the non-migrating flying fish, such as Exocoetus volitans and Hirundichthys affinis.
The basic integrative functional characteristics of a community are associated with patterns of the biomass size spectrum. This can be demonstrated by the relationship between the PP to total biomass (PP/Btot) ratio and the community spectrum slope in different regions. The PP/Btot characterizes the turnover rate of phytoplankton through a community biomass. It indicates the intensity of energy flow per biomass unit (Margalef, 1967; Odum, 1971; Vinogradov and Shushkina, 1987) and has the dimensionality of ‘day-1’. The macroscale structure of the pelagic community is constructed in a way that the highest turnover rates are in regions where the size spectra are mainly sloped (Figure 3a). In a double logarithmic scale this relationship can be approximated by a linear regression. Size spectra with maximal slopes were observed in the convergence regions, where the Lomonosov and the South Passat currents interacted. The positive slopes of spectra were not statistically significant and perhaps could be considered as the ‘zero-sloped’ spectra, i.e. similar to that in the divergence zones.
|
An attempt was made to derive the relationship between the PP/Rtot ratio and the community biomass spectrum slope (Figure 3b
). The ratio of PP to total community metabolism (PP/Rtot) may represent the balance between production and destruction processes in a pelagic community (Vinogradov and Shushkina, 1987). The data in the chart have a noticeable scatter, therefore we treat it as a possible (tentative) trend only, which indicates that in a community with high-sloped spectra (with slopes from -0.7 to -0.4) the PP should exceed the respiration and should cause energy losses from three- to five-fold. In communities with the flat-type size spectra (with slopes from -0.2 to 0) a balance between PP and community metabolism might be observed. Size spectrum at the level of single groups: the macrozooplankton size range
Size spectra of macrozooplankton crustaceans were studied along transects, 6, 12 and 18°W (Figure 4). In all three cases, size spectra from the south subequatorial divergence region were different in comparison with those from other regions. Spectra from the divergence zone had a well-pronounced bimodal type. The size group 20–25 mm formed the second peak of biomass. The amplitude of this peak increased in a longitudinal (western) direction, along the main water mass transport (i.e. from 6°, through 12°, to 18°W transects). In a latitudinal (i.e. perpendicular) direction, within jets of the South Passat Current from south to north, the amplitude of the second biomass peak decreased from the divergence to convergence region. The Euphausiids, Thysanopoda tricuspidata,E. americana and E. gibboides were the major components of the crustaceans in the divergence zone. However, the percentage of Euphausia and Thysanopoda species in the total biomass of macroplankton crustaceans changed considerably from the south subequatorial divergence towards the north (Table IV). Thysanopoda were dominant throughout the Gulf of Guinea (transect 6°W) and were numerous in the divergence zone. The dominance in the divergence zone was very pronounced along the open ocean transects, i.e. 12 and 18°W. Thysanopoda contributed up to 96% of total crustaceans in these areas, but the percentage decreased considerably towards the north, where Euphausia became more abundant.
|
|
Macroscale spatial variability at the community level
Measurements taken on these transects of various components of the community enabled trends of variability to be estimated on a scale of the eastern tropical Atlantic. The biomass of phytoplankton (direct measurements), meso-, macroplankton and micronekton were used for these comparisons. The mesoplankton group was represented by the biomass of the net zooplankton, and the macrozooplankton group was represented by the total biomass of macroplanktonic crustaceans. The total biomass caught by mesopelagic trawl was used to quantify the micronekton fish. The variation coefficient and the dispersion index (the variance to mean ratio) were used to assess the spatial variance. Both characteristics showed the same trend of increasing spatial variability in the sequence: phytoplankton–mesozooplankton–macrozooplankton–micronekton organisms (Table V). The same trend is seen in spatial variability as the function of the size of organisms.
|
Relationships between the parameters
To evaluate any links between the community and the water mass, data were subjected to principal components analysis (Table VI). Four principal components explained ~75% of the total variability. The first and the most powerful component mirrored the trophic link between various size fish and depth of occurrence and seston biomass. The second component could also be interpreted in terms of trophic interactions between macroplankton crustaceans, gelatinous organisms and squid. The key element in the third component was the link between the thermocline (i.e. the temperature vertical gradient) and PP, which were positively associated. The fourth principal component seemed to support the information in the third component. Not only were functional characteristics, such as PP, associated with the structure of the water masses (the thickness of the thermocline), but also the biomass of phyto- and zooplankton and their abundant groups. Trends within this link are controversial for the phytoplanktonic and zooplankton components.
|
Structural and functional characteristics were also used to explore the relationships between basic integrative ratios, characterizing a community on the macroscale, from hundreds to thousands of km. One of these is the ratio of the community total metabolism to PP (Rtot/PP). Another is the ratio of the PP to community total biomass (PP/Btot). Rtot and Btot were calculated from groups of organisms in Table I
and their values summarized. The relationship between ratios could be approximated by the asymptotic-type curve (Figure 5a). As mentioned above, the PP/Btot ratio indicates the intensity of matter or energy flow per biomass unit. In terms of a given relationship it can be shown that when the imbalance between the community metabolism and the rate of PP was high, the intensity of matter flux per unit of biomass was at a minimum. When the intensity of the matter flow increased these imbalances decreased non-linearly. In situations when the PP was ~0.1 of the community biomass a balance between metabolism and autotrophic production might be seen. The ratio of PP to total biomass achieved a plateau of minimal values (~0.05) when the community total metabolism exceeded the PP by 1.5–2 times.
|
The community metabolism per unit of biomass (Rtot/Btot) was also associated with the balance between the production and metabolism characterized by Rtot/PP ratio. On the scale of the area studied and within the observed ecological situations Rtot/Btot was linearly inclined with an increase of the Rtot/PP ratio (see Figure 5b
).
Of note, at high trophic levels of the community a relationship between the biomass of squid and gelatinous zooplankton was found (Figure 6). Perhaps this link has a causal relationship as well as a statistical one. Analysis of the contents of the squid stomachs showed that the squid were actively consuming gelatinous organisms wherethey were abundant, although pelagic euphausiids, amphipods, large copepods, flying and myctophid fish were also a considerable part of the squids’ diet in the eastern tropical Atlantic (Zuyev et al., 1988a,b). In turn, this forms a basis for a relationship between characteristics of the community size structure and an index of the fullness of squid stomachs (Figure 7). This index is given in non-dimensional units. We had only 10 samples taken in strong divergence and convergence zones, together with analysis of squid stomachs. Therefore, the chart gives a possible pattern of the relationship, which enables us to hypothesize that squid might be sensitive to the structure of the community energy spectrum. Squid exhibited maximal values of stomach fullness in regions with mainly flat energy spectra, i.e. in the macroscale divergence regions, and minimal values of stomach fullness in regions with steeper spectra.
|
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
The PP measurements from the studied area are characteristic of mesotrophic regions with values from 100 to 500 mg C m-2 day-1 and in localized eutrophic zones from 500 to 1000 mg C m-2 day-1 (Zuyev et al., 1990
). The so called ‘typical tropical structure’ of water masses (Herbland and Voituriez, 1979) dominated throughout the region except for the local divergence and convergence zones, where the vertical profiles of chlorophyll a did not exhibit well developed deep maxima in the thermocline layer (Finenko et al., 1991). If we consider our results in terms of spatial–temporal drift of a pelagic community, from divergence to convergence zones, it could be stated that the community responded to both dynamic processes by changing the biomass size spectrum. These changes are seen in spectra of single ecological groups, such as macrozooplankton (see Figure 4) and changes also take place at the level of the total ‘rough’ community spectrum (see Figure 2).
Over a wide range of spatial and size scales, the biomass spectrum of a pelagic community is shown to be relatively flat (Sheldon et al., 1972), which might be explained in terms of relationships between the body size and metabolism intensity (Kerr, 1974). If the organisms contributing to the spectrum are believed to be trophically linked (Platt and Denmann, 1978; Silvert and Platt, 1980), the slope characterizes the efficiency of biomass transfer through the spectrum, from small to larger sized groups of organisms. In this case, then the steeper the spectrum, the larger the amount of small-sized organisms supports a relatively small biomass of larger ones (Gaedke, 1995).
The biomass spectra obtained allow us to suggest that the spectrum slope is a fairly dynamic parameter. In the eastern tropical Atlantic, due to the contribution of the large-sized macroplankton organisms, community size spectra from the divergence zone, represented in energy units (i.e. the energy stocked in a body), were less sloped than spectra from convergence zones. This does not agree with the concept that divergence zones are sites of community formation, where microplankton dominate and convergence zones are where higher trophic levels are more abundant, as described for the equatorial currents of the Indian and the Pacific Oceans (Vinogradov and Voronina, 1964; Vinogradov and Shushkina, 1987). It should be noted, however, that in the eastern tropical Atlantic, zones of localized open ocean upwellings, with the subsequent divergence of waters, were thicker than those observed in the equatorial and subequatorial currents of the Indian Ocean. In the Indian Ocean the thickness of the mixed layer was often 5–20 m (Plotnikov, 1986), whereas it was 40–65 m in the divergence zones of the eastern tropical Atlantic.
Dominance of the large size macrozooplankton is also evident at the level of single ecological groups. From the size spectra of macrozooplankton it can be seen that, from divergence to convergence zone, the large-size groups of crustaceans are replaced by smaller ones in the southern part of the studied area (see Figure 4).
The complexity of structural relationships within the pelagic community can be classified into patterns of statistical links most contributing to the variability of the macroscale community. Using principle components analysis, we have shown that in the framework of the studied parameters, ~50% of the macroscale variability can be explained by trophic links acting at the higher trophic levels of the community. First, there are linkages between the seston biomass (as the food base), the biomass of mesopelagic migrating and the non-migrating fish inhabiting the surface layer. Secondly, there are linkages between the gelatinous organisms, euphausiids and squid. Additionally, the relationships between PP, phytoplankton biomass, mesozooplankton biomass and characteristics of the thermocline can explain 75% of the variability of the community. Therefore, macroplankton trophic interactions are obviously important in the regulation of macroscale variability of the pelagic ecosystem of the eastern tropical Atlantic. This also means that previous investigations conducted without assessments of the macrozooplankton have given distorted estimations of macroscale structure of pelagic community and the associations between their components in these regions.
A great deal of literature has been directed to the problem of how fluctuations at low trophic levels are transmitted up the pelagic food webs (Silvert and Platt, 1980; Cushing, 1982; Sinclair et al., 1986; Runge, 1988). Our data suggest that the spatial variability of the biomass of a pelagic community is organized in such a way that variation and dispersion increase as one progresses through the community size spectrum, from phyto- through meso-, macroplankton and finally to micronekton organisms. The same trend is observed for the tropical pelagic communities of the Indian Ocean on meso- and macroscale (Piontkovski et al., 1985, 1995). Mackas and Boyd (Mackas and Boyd, 1979) have also reported that zooplankton is distributed more heterogeneously than phytoplankton on a scale from tens to hundreds of kilometres, in temperate waters, and the same qualitative trend was noted for the plankton communities of the tropical Atlantic Ocean (Piontkovski et al., 1997).
A linear relationship evaluated between the biomass of squid and that of gelatinous macroplankton (represented mainly by phyrosomids) suggests a new type of trophic link between these groups of organisms in the eastern tropical Atlantic. This was first noted at the qualitative level (Zuyev et al., 1990), when huge aggregations of Phyrosoma atlantica, from 2 to 3 g m-3, were reported from upwellings in the open ocean. The biomass exceeded the background values in adjacent waters by one to two orders of magnitude. Perhaps, phyrosomids consuming phytoplankton in the upwelled waters can achieve these observed high aggregations fairly quickly due to their high specific diel production, which is ~1.2 (Zaika, 1983). We have now outlined a quantitative framework of the relationship between gelatinous zooplankton and squid (see Figure 6). High reproduction of gelatinous zooplankton at sites of open ocean upwellings and the fact that they are actively consumed by squid, an abundant pelagic predator of the eastern tropical Atlantic, indicates an important role of gelatinous animals in the energy and matter transformation within tropical ecosystems. This has been previously underestimated. On the other hand, the relationship between the slope of the size spectrum and the fullness of squid stomachs point out that these abundant predators are fairly sensitive to the energetic structure of the whole pelagic community. This is not surprising, as half of the community size spectrum contains organisms that are preyed upon by squid (Zuyev et al., 1988a,b).
Statistical links evaluated between the integrative characteristics perhaps enable one to monitor the relationship between structure and function at the level of the whole pelagic community. It was shown that basic ratios reflecting the turnover rate of PP through a community biomass (PP/Btot) and the balance between production and metabolism within a community (PP/Rtot) depend upon size structure. The regressions observed for both of these ratios makes it possible to assess the functional characteristics of a community on the basis of its size structure. On the other hand, basic structural–functional characteristics, mirroring the ratio betweenproduction and metabolism, are also linked statistically. Thus, the turnover rate of PP through a community biomass (PP/Btot) and the metabolism rate per community biomass (Rtot/Btot) could be presented as functions of the Rtot/PP ratio. This mirrors a degree of balance between the total production and the metabolism of a community (see Figure 5). As a result, any studied integrative characteristics of structure or functional processes in a region could be reconstructed, if just one of them is known. Such an opportunity is attractive in light of the indicators for a long-term monitoring of fishery-stressed regions, such as the eastern tropical Atlantic Ocean. The monitoring could be based on remotely acquired data from past (CZCS) imagery and from recently launched satellites (SeaWiFS). Knowing the chlorophyll a or the PP, the above relationships enable the reconstruction of the spectra of biomass and the major structural–functional ratios of the pelagic community to be made.
Our results imply a high variability in the majorstructural–functional characteristics of the pelagic community, which is on a scale from tens to hundreds of kilometres. This is important from the point of view of the concept of spatial–temporal succession of a community in which the ratio of community metabolism to PP (log Rtot/PP) is interpreted as the ‘index of maturity’ of a community, whereas the inverse ratio (PP/Rtot) is considered as the ‘coefficient of maturity’ (Vinogradov and Shushkina, 1987). This interpretation was used to analyse community succession drifting from the upwelling to the open ocean regions. However, in the eastern tropical Atlantic with its complicated layered system of interacting currents in various directions and cross-circulations, a similar interpretation of the above ratios is hardly applicable. A ‘pie’ of local convergence and divergence zones caused by interacting and meandering currents leads to high variability of all basic integrative ratios in the mesoscale. If we assume the succession concept, i.e. that, drifting with a current, the community passes through different stages of development, from the ‘initial’ (at the upwelling) through to the ‘developed’ to the ‘mature’ stage and that each of these stages can be characterized by definite values of above mentioned ratios, we should obtain spatially random changes of these ratios. For example, the PP/Rtot ratio changes from 0.1 to 5.0 covering the whole range of stages of a community, from the ‘young’ (where PP/Rtot = 5.0–1.6) to the ‘mature’ (where PP/Rtot = 0.1–0.01). This means that mesoscale variability of the structural–functional ratios is fairly high and exceeds that of the macroscale pattern.
From the point of view of ecosystem typology, we have attempted to analyse the relationship between the structure and the function in a biogeochemical province (the eastern tropical Atlantic Ocean). We hope to extend this approach to other biogeochemical provinces of the Atlantic Ocean, such as the Western tropical Atlantic, the South Atlantic gyre and the others, following a modern biogeochemical partitioning of the oceans (Longhurstet al., 1995; Pauly et al., 2000). These comparisons should enable an ecological classification of relationships between structural and the functional parameters to be established, and possibly to develop further the concepts in a way enabling us to answer the question: can we use a set of basic structural–functional relationships as a sensitive indicator of the biogeochemical provinces? Are these relationships different over the provinces, or, could the provinces be grouped into ones with similar trends of structural–functional relationships (i.e. not just the chlorophyll–environment linkages used to develop a current typology of marine ecosystems)?
|
Acknowledgments |
|---|
We are grateful to colleagues from IBSS: S. Tsarin, V. Nikolsky, G. Zuyev, V. Lyschina, V. Skryabin, L. Stelmah, I. Prusova, and others, for data provided. This work was supported by the NSF grant DEB-0203622, INTAS grant 00-INFO-0059 and the DETR grant 162/8/251.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODRESULTSDISCUSSIONREFERENCES |
|---|
Belevich, R. R. (1975) Studies of the spatial-temporal variability of the water heat structure of the central Atlantic troposphere. Rep. Oceanogr. Inst. (Trydi Okeanograficheskogo Instituta), 124, 89–103 (in Russian).
Binet, D. (1997) Climate and pelagic fisheries in the Canary and Guinea currents 1964–1993: the role of trade winds and the southern oscillation. Oceanol. Acta, 20, 177–190.
Constanza, R. (1992) Toward an operational definition of ecosystem health. In Constanza, R., Norton, B. G. and Haskell, B. D. (eds), Ecosystem Health: New Goals for Environmental Management. Island Press, Washington, DC, pp. 134–147.
Corcoran, E. F. and Mahnken, C. V. W. (1969) Productivity of the tropical Atlantic ocean. Proc. Symp. Oceanogr. and Fish Resour. Tropic. Atlantic. Abidjan, October 20–28, 1966, UNESCO, Paris, pp. 57–67.
Cushing, D. H. (1982) Climate and Fisheries. Academic, New York, 373 pp.
Elton, C. (1927) Animal Ecology. Sidwick and Jackson, London, 207 pp.
Finenko, Z. Z., Stelmah, L. V. and Jorov, C. V. (1991) The impact of the frontal zones on the distribution of chlorophyll ‘a’ in the eastern part of the tropical Atlantic (Vliyanie frontalnyih zon na raspredelenie hlorofilla a v vostochnoi chasti tropicheskoi Atlantiki). Ecology of the Sea (Ekologiya Morya), 39, 17–22 (in Russian).
Gaedke, U. (1995) A comparison of whole-community and ecosystem approaches (biomass size distributions, food web analysis, network analysis, simulation models) to study the structure, function and regulation of pelagic food webs. J. Plankton Res., 17, 1273–1305.
Gomes, M. C. and Haedrich, R. L. (1992) Predicting community dynamics from food web structure. In Rowe, G. T. and Pariente, V. (eds), Deep Sea Food Chains and the Global Carbon Cycle. Kluwer Academic Publishers, Dordrecht, Vol. 360, pp. 277–293.
Greze, V. N. (1971) Plankton and Biological Productivity of the Tropical Atlantic (Plankton i prodyktivnost tropicheskoi Atlantiki). Naukova Dumka, Kiev, 278 pp (in Russian).
Haedrich, R. L. (1986) Size spectra in mesopelagic fish assemblages. Pelagic Biogeography. UNESCO Tech. Papers Mar. Sci., 49, 107–117.
Hanaichenko, N. K. (1980) In Pandit, V. (ed.), The System of Equatorial Counter Currents in the Ocean. Oxonian Press Pvt. Ltd, New Delhi,158 pp.
Herbland, A. and Voituriez, B. (1979) Hydrological structure analysis for estimating the primary production in the tropical Atlantic Ocean.J. Mar. Res., 37, 87–101.
Herbland, A., Le Borne, R., LeBouteiller, A. and Voituriez, B. (1983) Structure hydrologique et production primaire dans l’Atlantique tropical oriental. Oceanogr. Tropical, 17, 15–25.
Hizard, P., Citeau, S. and Voituriez, B. (1977) Equatorial undercurrent influences on enrichment processes of upper water in the Atlantic ocean. Report of the International Workshop on the GATE Equatorial Experiment. February 28–March 10, 1977, Miami, pp. 1–10.
Houghton, R. W. (1989) Influence of local and remote wind forcing in the Gulf of Guinea. J. Geophys. Res., 94, 4816–4828.
Kerr, S. R. (1974) Theory of size distribution in ecological communities. J. Fish. Res. Bd Can., 31, 1859–1862.
Kort, V. G. (1975) Studies of the oceanic currents. Oceanolog. Res. (Okeanologicheskie issledovaniya), 27, 5–25 (in Russian).
Le Borne, R. (1977) Etude de la production pelagique de la zone equatoriale de l’Atlantique a 4o W. 3. Respiration et excretion d’azote et de phosphore du zooplancton. Cah. ORSTOM Oceanogr., 15, 349–362.
Lindeman, R. L. (1942) The trophic–dynamic aspect of ecology. Ecology, 23, 399–418.[CrossRef][Web of Science]
Longhurst, A., Sathyendranath, S., Platt, T. and Caverhill, C. (1995) An estimate of global primary production in the ocean from satellite radiometer data. J. Plankton Res., 17, 1245–1271.
Mackas, D. L. and Boyd, K. (1979) Spectral analysis of zooplankton spatial heterogeneity. Science, 204, 4388, 62–64.
Mann, K. H. (1982) Ecology of Coastal Waters. A Systems Approach. Blackwell, Oxford, 322 pp.
Mann, K. H. (1988) Towards predictive models for coastal marine ecosystems. In Pomeroy, L. R. and Alberts, J. J (eds), Concepts of Ecosystem Ecology. Ecological Studies, 67, Springer, New York, pp. 291–316.
Moloney, C. L. and Field, J. G. (1985) Use of particle-size data to predict potential pelagic-fish yields of some southern african areas. S. Afr. J. Mar. Sci., 3, 119–128.
Morel, A. (1996) An ocean flux study in eutrophic, mesotrophic, and oligotrophic situations: the EUMELI program. Deep-Sea Res. I, 43, 1185–1190.
Odum, H. T. (1971) Environment, Power, and Society. Wiley-Interscience, New York, 331 pp.
Pauly, D., Christensen, V., Froese, R., Longhurst, A., Platt, T., Sathyendranath, S., Sherman, K. and Watson, R. (2000) Mapping fisheries onto marine ecosystems: a proposal for a consensus approach for regional, oceanic and global integration. In Pauly, D. and Pitcher, T. J. (eds), Methods for Evaluating the Impacts of Fisheries on North Atlantic Ecosystems. Fisheries Centre Research Reports 8 (2), 13–22.
Philander, S. G. H. and Pacanowski, R. C. (1981) The oceanic response to cross equatorial winds with application to coastal upwelling in low latitudes. Tellus, 33, 201–210.[Web of Science]
Piontkovski, S. A. and Williams, R. (1995) Multiscale variability of tropical ocean zooplankton biomass. ICES J. Mar. Sci., 52, 643–656.
Piontkovski, S. A., Melnik, T. A., Sidko, A. F., Plotnikov, V. A. and Chistenko, V. M. (1985) Elements of spatial structure of pelagic communities in the surface layer of the Indian Ocean. Pol. Arch. Hydrobiol., 32, 385–393.
Piontkovski, S. A., Williams, R. and Melnik, T. A. (1995) Spatial heterogeneity, biomass and size structure of plankton of the Indian Ocean: some general trends. Mar. Ecol. Progr. Ser., 117, 219–227.
Piontkovski, S. A., Williams, R., Peterson, W. T., Yunev, O. A., Minkina, N. A., Vladimirov, V. L. and Blinkov, V. (1997) Spatial heterogeneity of the planktonic fields in the upper mixed layer of the open ocean. Mar. Ecol. Progr. Ser., 148, 145–154.
Platt, T. and Denmann, K. (1978) The structure of pelagic marine ecosystems. Rapp. P.-V. Reun. Cons. Int. Explor. Mer., 173, 60–65.
Platt, T. and Sathyendranath, S. (1988) Oceanic primary production: estimation by remote sensing at local and regional scales. Science, 241, 1613–1620.
Plotnikov, V. A. (1986) Temperature-salinity and dynamical structure of waters in the subequatorial regions of the Indian Ocean. In Petipa, T. S. (ed.), Ecological Systems in Active Dynamical Zones of The Indian Ocean (Ekologicheskie Sistemyi Aktivnyih Dinamicheskih Zon Indiiskogo Okeana). Naukova Dumka, Kiev, pp. 18–46 (in Russian).
Postel, L. (1990) The mesozooplankton response to coastal upwelling off West Africa with particular regard to biomass. Ber. Mar. Sci. Rep., 1, 127 pp.
Runge, J. A. (1988) Should we expect a relationship between primary production and fisheries? The role of copepod dynamics as a filter of trophic variability. Hydrobiologia, 167/168, 61–71.
Sheldon, R. W., Prakash, A. and Sutcliffe, W. H. Jr (1972) The size distribution of particles in the ocean. Limnol. Oceanogr., 17, 327–340.
Sherman, K. and Solow, A. R. (1992) The changing states and health of a Large Marine Ecosystem. ICES J. Mar. Sci., L 38, 1–31.
Sherman, K. L., Alexander, M. and Gold, B. D. (eds), (1990) Large Marine Ecosystems: Patterns, Processes and Yields. AAAS Press, Washington, DC, 242 pp.
Sherman, K. L., Alexander, M. and Gold, B. D. (eds), (1993) Stress, Mitigation and Sustainability of Large Marine Ecosystems. AAS Publishers, Washington, DC, 268 pp.
Shiohama, T., Myojin, M. and Sakamoto, H. (1965) The catch statistic data for Japanese tune long-line fishery in the Atlantic Ocean. Nakai Regional Fish Lab. Sanbarghidoru, Kochi, Japan, 21, 1–131.
Silvert, W. and Platt, T. (1980) Dynamic energy-flow model of the particle size distribution in pelagic ecosystems. In Kerfoot, W. C. (ed.), Evolution and Ecology of Zooplankton Communities, pp. 754–763.
Sinclair, M. G., Bugden, G. L., Tang, C. L., Therriault, J.-C. and Yeats, P. A. (1986) Assessment of effects of freshwater runoff variability on fisheries production in coastal waters. In Skreslet, S. (ed.), The Role of Freshwater Outflow in Coastal Marine Ecosystems. NATO ASI Ser., G7, pp. 139–160.
Sokolov, V. A. (1977) The circulation of waters of the tropical Atlantic. Reports of the Oceanographic Institute (Trydi Okeanograficheskogo Instituta), 135, 79–86 (in Russian).
Sprules, W. G. and Goyke, A. P. (1994) Size-based structure and production in the pelagia of lakes Ontario and Michigan. Can. J. Fish. Aquat. Sci., 51, 134–149.
Vinogradov, M. E. and Shushkina, E. A. (1987) The Functioning of the Plankton Communities of the Ocean Epipelagic (Funktcionirovanie Planktonnyih Soobchestv Epipelagiali Okeana). Moskow, Nauka, 240 pp (in Russian).
Vinogradov, M. E. and Voronina, N. M. (1964) The distribution of macroplankton in the northern part of the Indian ocean. Oceanology, 2, 119–125.
Zaika, V. E. (1983) The Comparative Productivity of Hydrobionts (Sravnitelnaya Productivnost Gidrobiontov). Naukova Dumka, Kiev, 206 pp (in Russian).
Zuyev, G. V., Oven, L. S. and Ovcharov, O. P. (1988a) Macroplankton and Nekton of the Tropical Atlantic (Makroplankton I Nekton Tropicheskoi Atlantiki). Naukova Dumka, Kiev, 218 pp (in Russian).
Zuyev, G. V., Nikolski, V. N. and Ovcharov, O. P. (1988b) The Assessment Of The Fish And The Squid Stocks (Otcenka Zapasov Ryib I Kalmarov). Agropromizdat, Moscow, 107 pp (in Russian).
Zuyev, G. V., Ovcharov, O. P. and Nikolski, V. N. (1990) In Zuyev, G. V. (ed.), The Productivity of the Equatorial Atlantic (Produktivnost Ekvatorialnoi Atlantiki). Naukova Dumka, Kiev, 228 pp (in Russian).
Received on April 19, 2001
; accepted on June 6, 2003
CiteULike 
Connotea 
Del.icio.us What's this?