ICES Journal of Marine Science: Journal du Conseil Advance Access published online on May 22, 2007
ICES Journal of Marine Science: Journal du Conseil, doi:10.1093/icesjms/fsm066
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Zooplankton abundance trends on Georges Bank, 1977–2004
NOAA, National Marine Fisheries Service, 28 Tarzwell Drive, Narragansett, RI, USA; tel: +1 401 7823244; fax: +1 401 7823201; e-mail: jkane{at}whsun1.wh.whoi.edu
Kane, J. 2007. Zooplankton abundance trends on Georges Bank, 1977–2004 – ICES Journal of Marine Science, 64.Interannual trends in Georges Bank zooplankton abundance are described and related to variations in environmental variables for the period 1977–2004. Total zooplankton counts increased to above average levels in 1989 and stayed over or close to average through 2004. This analysis identified a group of taxa including Centropages typicus, Metridia lucens, and Temora longicornis that had similar interannual patterns of abundance. All these taxa increased sharply in the early 1990s and remained high through 2001. Some taxa declined sharply in 2002, others have continued elevated through 2004. Total zooplankton counts in the past two years were also boosted by a substantial increase in the abundance of the copepod Calanus finmarchicus. Evidence is presented that these changes may be related to variations in Scotian Shelf inflow, which freshened water on the Northeast US continental shelf, perhaps increasing both primary production and the influx of zooplankton into the region. There was a positive correlation between the biomass of pelagic predators and the abundance of several zooplankton taxa, suggesting that bottom-up processes and advective supply are the key factors that regulate the Georges Bank foodweb.
Keywords: advection, Centropages typicus, Georges Bank, Metridia lucens, salinity, Temora longicornis, trends, zooplankton
Received 8 August 2006; accepted 11 April 2007.
 |
Introduction |
|---|
 Top Introduction Material and methodsResultsDiscussionReferences |
|---|
Multi-year zooplankton abundance data provide important information regarding the stability of marine ecosystems. Recent time-series studies have unearthed evidence that climate variability has had a profound effect on the distribution and structure of zooplankton communities across the Atlantic Basin. For example, since the late 1980s, the zooplankton population in the eastern North Atlantic Ocean has been slowly shifting to a warmer water community with a progressive increase in the presence of subtropical species (Beaugrand et al., 2002). Oceanographers have also established trans-Atlantic correlations between copepod population fluctuations and variability in the North Atlantic oscillation (NAO), the dominant climate mode in the North Atlantic Basin (Fromentin and Planque, 1996; Conversi et al., 2001; Greene and Pershing, 2001).
Zooplankton time-series data have also been used to demonstrate that overfishing can have serious implications for population levels of lower trophic levels. Stock assessment studies off eastern Canada have recently used zooplankton time-series data to imply that lowered predation from declining cod stocks had cascading effects throughout the foodweb (Frank et al., 2005). Monitoring and understanding the interactions between fish and zooplankton populations is a key concern in ecosystem-based fisheries management.
Georges Bank is a large, shallow region located off the northeastern US at the edge of the continental shelf (Figure 1). The primary source of the waters in the region is shelf water from the Gulf of Maine that enters its northern flank and moves in a clockwise direction around the eastern end of the Bank. Water then moves westwards along its southern flank until most of it exits into the Mid-Atlantic Bight. The region's unique topography and oceanographic conditions combine to promote highly productive phytoplankton and zooplankton populations that support high fish production (Sherman et al., 1987).
|
Recent studies have shown that the inflow of relatively cool water of low salinity from the Scotian Shelf into the Gulf of Maine increased during the 1990s (Smith et al., 2001; Mountain, 2004). This caused salinity to decline on Georges Bank and throughout the surface waters of the northeastern US continental shelf (Mountain, 2004). Pershing et al. (2005) analysed the Gulf of Maine continuous plankton recorder (CPR) zooplankton time-series from 1963 to 2002 and reported that the abundance of several taxa increased around 1990 and remained high through 2001. They speculated that the low salinity led to increased stratification that caused more primary production, leading to elevated levels of zooplankton.
The National Marine Fisheries Service has conducted broad-scale plankton surveys of the Georges Bank region since 1977. The sampling protocol has remained virtually unchanged throughout (Kane, 2003), providing a time-series of observations needed to define long-term trends in abundance. Previous Georges Bank studies using this database have generally focused on dominant species for relatively short periods (Sherman et al., 1983; Kane, 1993; Meise and O'Reilly, 1996). Here, I describe the interannual abundance variability of all the common taxa that constituted the region's zooplankton community for the years 1977–2004. Multivariate analysis was used to test whether the underlying structure of the zooplankton community changed over the 28 years. Abundance variability was examined in relation to fluctuations in environmental indices measured during the same period.
|
Material and methods |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Zooplankton data
Plankton samples from Georges Bank were collected seasonally from 1977 to 2004 on two types of cruises: (i) broad-scale surveys dedicated to plankton, where sampling was carried out at standard or randomly selected stations spaced approximately 8–35 km apart (Figure 1), and (ii) trawl and dredge surveys where plankton stations were selected from a stratified random plan at locations uniformly distributed over the region. Samples were collected with a 61-cm bongo frame fitted with a net of 0.333 mm mesh, towed obliquely to a maximum depth of 200 m or 5 m from the bottom and back to the surface. A flowmeter was suspended in the centre of the Bongo frame to measure the volume of water filtered during the tow. Specimens were preserved in 5% formalin. In the laboratory, samples were reduced to approximately 500 organisms by subsampling with a modified box splitter. Zooplankton were sorted, counted, and identified to the lowest possible taxa at the Polish Plankton Sorting and Identification Centre in Szczecin, Poland. Abundance is expressed as number per 100 m3. The results from 4939 samples were used in the study.
Survey cruises did not cover the region at the same time each year, so direct comparisons between annual survey abundance levels are confounded by the seasonal cycle of zooplankton populations. To reduce the bias caused by sampling variability and to allow comparison between years, the average annual cycle in abundance of each taxon was computed by fitting a spline-curve function to the time-series bi-monthly mean values of abundance, calculated with data from complete surveys whose midpoint fell within the two-month bin. This generates the expected abundance on any day of the year. Survey means were then subtracted from the projected abundance on the median day of that particular cruise. This created a data set of anomalies from the seasonal cycle, which were then averaged over each year to produce an annual abundance index. This study was limited to the 23 taxa whose time-series mean was at least 500 per 100 m3 or percentage occurrence 
25%. All data were log-transformed (base 10) prior to analysis.
Environmental data
The interannual variability of several environmental variables was examined and analysed for linkages to the zooplankton community. Surface temperature was measured at every station throughout the time-series, with a stem thermometer from a surface bucket sample from 1977 to 1998, and thereafter via a thermister attached to the vessel. Comparison of these two measurements has shown that simultaneous readings from both techniques are not significantly different. Annual anomalies were calculated using the same methods described earlier for zooplankton abundance indices. Annual surface layer (0–30 m) temperature and salinity anomalies for the northwestern Georges Bank were also analysed (see Mountain, 2004). Measurements were restricted to this area (Figure 1) because it represents the Georges Bank water mass, unaffected by the contamination effects of the frequent slope water incursions along the southern flank of Georges Bank.
The phytoplankton colour index measured on CPR surveys of the nearby Gulf of Maine (Figure 1) was used for this work as a proxy measure of Georges Bank primary production. This is appropriate because satellite-derived surface chlorophyll values from 1998 to 2005 in the two regions exhibited similar trends (www.nefsc.noaa.gov/omes/OMES/spring2006/adv2.2.html). The value is a qualitative indicator of phytoplankton biomass (Reid et al., 1998), and its annual anomaly was calculated as described above for zooplankton.
Climate variability was indexed with the winter phase of the NAO, an index based on the difference in normalized sea level pressure between Lisbon, Portugal, and Stykkisholmur/Reykjavík, Iceland, from December through March (Hurrell, 1995).
The combined total biomass of Atlantic herring (Clupea harengus) and Atlantic mackerel (Scomber scombrus) in Georges Bank and Gulf of Maine waters (Overholtz et al., 2004; Northeast Fisheries Science Center, 2006) was used as an indicator of pelagic predator abundance through the time-series. The annual collective standing stock of these two common planktivores was log-transformed and normalized to one unit of variance prior to analysis.
Data analysis
Interannual changes in species diversity were assessed using the Shannon diversity index (Shannon, 1948):
|
|
Statistical analysis of abundance data employed a variety of multivariate techniques, using the PRIMER 6.1.5 software package (Clarke and Warwick, 2001). Initially, the biotic relationship between any two years was represented by the Bray–Curtis index (Bray and Curtis, 1957), which measures the similarity (or dissimilarity) in species composition. The triangular matrix of similarity between each pair of years was then classified into groups using two techniques: (i) hierarchical agglomerative cluster analysis, and (ii) non-metric multi-dimensional scaling (MDS). Clusters of years that were found to be statistically significant (p < 0.05) by the similarity profile test (SIMPROF) and also isolated by low stress (<0.20) MDS ordinations were judged to be years with similar zooplankton community structure. Taxa responsible for dissimilarity between these clusters of years were identified using the SIMPER procedure (Clarke, 1993).
Patterns in the environmental data were initially examined using the ordinations produced by principal component and cluster analysis (Clarke and Warwick, 2001). Zooplankton community structure was then linked to environmental variables using the BIO–ENV procedure (Clarke and Ainsworth, 1993). This analysis compares ordinations from biotic and abiotic configurations and selects the subset of environmental variables that provides the best match, as ascertained by the resultant Spearman's rank correlation of similarities. The latter coefficients were also calculated between individual annual abundance anomalies and environmental variables, to determine which had similar trends.
|
Results |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Zooplankton trends
The 23 individual taxa examined here in sum constituted 94.9% of the total zooplankton enumerated in the time-series (Table 1). The overall diversity of the Georges Bank zooplankton community was relatively stable during the study period. The change in the Shannon diversity index during the period was minimal, with no long-term trend (Figure 2). In contrast, multivariate analysis showed that the community abundance structure of Georges Bank zooplankton changed considerably during the time-series. Ordination of annual anomalies by both clustering and MDS multivariate techniques indicated that there were four distinct groups of nearly consecutive years of different abundance levels: 1977 and 1978 (Group 1); 1980–1990 (Group 2); 1979 and 1991–2001 (Group 3); and 2002–2004 (Group 4) (Figure 3). MDS produced an ordination with low stress (0.11), indicating that the data were well represented in two-dimensional space. The SIMPROF procedure showed that the abundance structure of these four groupings of years were significantly different (p < 0.01).
|
|
|
The first two groups were isolated primarily by low abundance, whereas groups 3 and 4 were segregated out by high density of certain taxa. Group 1 differed from the others because of low abundance of several taxa that had negative abundance anomalies in these years (Table 2). Group 2 was isolated from Groups 3 and 4 by the low density of Oithona spp., hyperiid amphipods, Temora longicornis, and Appendicularia (Table 2). The abundance of several taxa in Group 3 separated those years out from the others, with T. longicornis contributing great dissimilarity to all pairings (Table 2). Although high densities of Appendicularia and Oithona spp. were major factors separating Group 4 from groups 1 and 2, the decline in numbers of three common copepods, Centropages typicus, Metridia lucens, and T. longicornis (Table 2), separated the years 2002–2004 from the years of Group 3.
|
Species associations were examined in greater detail by calculating pairwise correlation coefficients between abundance anomalies (Table 3) and plotting time-series trend lines (Figure 4). A group of eight taxa was strongly correlated with each other in terms of abundance (p < 0.01): Centropages typicus, M. lucens, T. longicornis, Thecosomata, Oithona spp., Appendicularia, Hyperiidae, and Coelenterata. Abundance of all eight increased in the early 1990s, then stayed at above average levels through at least 2001, with the latter four remaining high through 2004 (Figure 4). The abundance patterns of Euphausiacea, Brachyura, and Polychaeta were also generally above average during the second half of the time-series (Figure 4), positively correlated with some of the latter eight taxa (Table 3). The dominant copepods, Centropages hamatus and Pseudocalanus spp. were more variable and only weakly correlated to other taxa that increased, but both were also usually above average during the 1990s (Figure 4).
|
|
This sudden increase of several species combined to push the annual anomaly of total zooplankton counts above or close to average from 1989 through 2004 (Figure 4). Total counts were exceptionally high from 1996 through 2001 and again in 2004. In contrast, low abundance of several taxa during the 1980s brought overall counts to below average levels from 1980 to 1988 (Figure 4), causing those years to be grouped in the multivariate analysis. The first two years of sampling were also marked by low abundance of many taxa, but were separated out by the high abundance of Calanus finmarchicus, Pseudocalanus spp., and chaetognaths (Figure 4). Sandwiched between these low-year clusters was one year where the densities of several taxa were above average, causing 1979 to be grouped with the 1990s. Abundance anomalies of the dominant copepods Centopages typicus, Pseudocalanus spp., and M. lucens were all positive that year, along with those of several minor taxa (Figure 4).
Abundance patterns during the two extended high- and low-abundance periods were not seasonally restricted. Total zooplankton counts were elevated year-round during the high-abundance period (Group 3), and low throughout below-average years (Group 2) (Figure 5). The seasonal cycles of the five species whose abundance contributed the most dissimilarity between groups (Table 2) were also seasonally coherent between year groups (Figure 5), although they have very different climatological annual cycles. Metridia lucens abundance is high from late autumn through early spring, and that of T. longicornis during summer. Centropages typicus peaks during autumn, and hyperiid amphipods pulse in late spring and early autumn. Oithona spp. numbers are similar year-round (Figure 5).
|
Environmental variables
The surface and surface layer (0–30 m) temperature trend lines were similar (r = 0.66, p < 0.01), both showing mostly positive anomalies from 1990 through 2003, with an extended period of elevated values from 1999 to 2002 (Figures 6a and 6b). Salinity and phytoplankton colour values were negatively correlated (r = –0.37, p = 0.05), displaying nearly opposite sawtooth patterns, with major fluctuations in the late 1990s (Figures 6c and 6d). Surface waters were fresher throughout most of the 1990s and into the new millennium, with a minimum in 1998. The annual phytoplankton colour anomaly was high for the three-year periods from 1991 to 1993 and from 1997 to 1999, the latter period having the maximum values. The only extended period where annual phytoplankton anomalies were below average was from 2002 to 2004. The NAO index was in a predominantly positive phase during the time-series (Figure 6e) that was only weakly correlated (r in all cases <0.15) to the other environmental variables. Pelagic fish biomass slowly rose through the time-series, with a sharp rise in 2000 that has been followed by steady, modest increases (Figure 6f). This pattern was positively correlated to surface layer temperature (r = 0.47, p = 0.01), and negatively related to salinity (r = –0.41, p = 0.03).
|
The ordination from principal component analysis (PCA) of the environmental data showed no clear or common pattern (Figure 7). The first two components accounted for just 56.9% of the total variation, indicating that the analysis gives an inadequate and possibly misleading representation of interannual differences between the variables. Cluster analysis did not support the PCA results, grouping and isolating different years. For example, the years 1996 and 2004 were clustered out together as significantly different (p < 0.05), whereas the PCA ordination placed them far apart and positioned 2004 tightly associated with a large group of other years (Figure 7).
|
The BIO–ENV procedure indicated that the zooplankton community structure was not strongly correlated to the variability of the environmental variables. The greatest correlation between the similarity rankings of species abundance and combinations of environmental variables was 0.23, using only the matrix based on trends in salinity and pelagic biomass. The analysis was also applied to specific group subsets (e.g. copepods only, top 10 most abundant taxa, only taxa with correlated trends) to examine whether different aggregations of zooplankton could be better related to environmental heterogeneity. However, correlations between these groupings were similar and lower than the analysis that included all taxa.
Pairwise correlation analysis between environmental variables and individual abundance anomalies was more enlightening, revealing several significant relationships between measures. Many taxa were positively correlated (p < 0.05) to pelagic biomass values (Table 4). Oithona spp., Appendicularia, Echinodermata, and Clausocalanus arcuicornis were strongly correlated (p < 0.01), mirroring the steady, increasing trend displayed by herring and mackerel stocks (Figures 4 and 6). The dominant copepods, M. lucens, C. finmarchicus, and T. longicornis were negatively correlated to salinity, usually increasing in numbers during years when salinity declined. Large numbers of C. finmarchicus were also found when the upper surface layer was cool, whereas Centropages typicus and Centropages hamatus flourished when the water wa
warmer (Table 4). There were no significant correlations between any of the zooplankton taxa with either phytoplankton colour or the NAO index.
|
|
Discussion |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
The Georges Bank zooplankton community underwent a major shift in abundance levels during the 1990s. The density of the region's most abundant mesozooplankton species (Centropages typicus) began to rise in the late 1980s, then steadily increased throughout the 1990s into 2001. The numbers of two other common copepod species, M. lucens and T. longicornis, began to increase in the early 1990s and also remained above average through 2001. Abundance of all three species declined sharply in 2002 and then remained at low levels through 2004.
Several less dominant taxa also increased in the 1990s (e.g. Appendicularia, Oithona spp., and Hyperiidae) and have remained at above average levels through 2004. Others, including Chaetognatha, Echinodermata, and the copepods Paracalanus parvus and Clausocalanus arcuicornis have all increased in abundance in the later years of the time-series. After more then a decade of approximately average density, the numbers of the dominant, large copepod C. finmarchicus increased sharply in 2003 and reached a time-series maximum in 2004. Despite the decline of the key species that fuelled the expansion in the 1990s, these later increases pushed total zooplankton counts to high levels in 2004.
CPR surveys in Gulf of Maine waters north of Georges Bank also reveal a dramatic increase in the abundance of some of these same taxa during the 1990s that ended with a rapid decline in 2002 (Pershing et al., 2005). Those authors describe a community shift away from C. finmarchicus dominance towards a group of smaller taxa that showed a large proportional increase in winter abundance. Northeast of the study area, on the Scotian Shelf off eastern Canada, there is evidence that the zooplankton community also increased significantly in the 1990s. There, winter counts of dominant taxa increased and the phytoplankton colour index was significantly higher year-round (Sameoto, 2001). The results of this analysis and others indicate a large-scale change in zooplankton abundance within Northwest Atlantic shelf waters during the 1990s. However, the increase in the Georges Bank zooplankton community was not seasonally restricted, nor did it end in 2002. The abundance of some dominant taxa declined, but that of others increased or continued at high levels through 2004.
The abundance increases in the early 1990s coincided with a large-scale freshening of shelf waters that was caused by increased inflow from the Scotian Shelf (Mountain, 2004). The numbers of many taxa were clearly significantly correlated to salinity trends, indicating that salinity played a role in the abundance pulse. Moreover, after several years of high abundance in waters of low salinity, the numbers of three important copepod species declined sharply in 2002, when the annual salinity anomaly became positive for the first time in 11 years (Figure 6). Pershing et al. (2005) proposed that low salinity in the Gulf of Maine during the 1990s increased stratification sufficiently to enhance primary production, later triggering increased secondary production. Although the Gulf of Maine phytoplankton index anomaly was above average in several years during the 1990s, interannual abundance variability of Georges Bank zooplankton taxa was not significantly correlated, perhaps because it is a poor indicator of spatial and temporal phytoplankton production on Georges Bank. Several CPR surveys in the North Atlantic report a large increase in phytoplankton biomass throughout the region since the late 1980s (Reid et al., 1998). The phytoplankton colour index for the Canadian Atlantic increased significantly in 1991, and it continued high through the end of its data series in 1998 (Sameoto, 2001). These high levels were similar to the indices recorded on a CPR transect crossing the southern flank of Georges Bank from 1991 to 1998 (Sameoto, 2001), showing that primary production was high during that period. Increased food availability likely was a key factor that allowed increased zooplankton abundance on Georges Bank during the 1990s.
Many of the common zooplankton species on Georges Bank do not have overwintering stages, so their populations require annual advective input. Durbin et al. (2003) found evidence that Scotian Shelf water was the likely source of several zooplankton taxa found in the Gulf of Maine during winter, specifically the dominant copepods Centropages typicus, Pseudocalanus spp., M. lucens, and Oithona spp. Therefore, increased inflow of Scotian Shelf water during the 1990s may also have contributed to increased zooplankton abundance on Georges Bank by injecting more taxa into the system. As zooplankton abundance was elevated on the Scotian Shelf during the 1990s (Sameoto, 2001), it was likely that higher than normal numbers were entrapped within the increased flow that eventually circulates or washes onto Georges Bank. Zooplankton stocks could also be transported directly onto Georges Bank from the Scotian Shelf. A cold, low-salinity surface layer periodically crossed the North East Channel during the late 1990s and onto Georges Bank (Wiebe et al., 2002), so it is certainly possible that the Scotian Shelf region is an important source of the zooplankton seed populations that colonize Georges Bank. Interannual variation in the advective processes transporting them could have a major influence on their annual population levels, and on the higher trophic levels that feed on zooplankton.
In the temperate North Atlantic, C. finmarchicus is generally known as a winter/spring cold-water species that gives way to warm-water species whose abundance peaks in summer/autumn. The results of this study confirmed that situation on Georges Bank, abundance anomalies of C. finmarchicus being negatively correlated to temperature trends, whereas Centropages typicus and Centropages hamatus were both positively related to temperature. Therefore, abundance trends are reversed, cool years enhancing C. finmarchicus production and warm ones being more favourable for the two species of Centropages. As the production of these three species is sensitive to temperature fluctuations, they are excellent candidates to be named sentinel species as harbingers of climate warming. Monitoring the future abundance trends of these species could, therefore, be critical in assessing the impact of climate variability.
The interactions of predators (top-down) and resource availability (bottom-up) control the community structure of aquatic foodwebs. It is generally believed that much of the variation found in marine systems is bottom-up regulated, whereas top-down processes are more critical in freshwater habitats where small and simple foodwebs exist. Only recently has evidence been presented that suggests that top-down processes might explain foodweb variability in continental shelf ecosystems (Worm and Myers, 2003), and Frank et al. (2005) reported that the collapse of benthic fish stocks in the eastern Scotian Shelf caused cascading effects down through the food chain. They found that the biomass of benthic invertebrates and pelagic fish increased sharply after the depletion of cod and other large predators in the benthic fish community, hypothesizing that this drop led to lowered levels of C. finmarchicus, the dominant herbivorous copepod and preferred food item of the now-elevated fish prey populations.
In contrast, the current analysis suggests that bottom-up processes control the abundance and species composition of the pelagic foodweb on Georges Bank. Pelagic fish biomass and abundance of several zooplankton taxa were positively correlated, indicating that both populations depend on factors that regulate productivity. The spawning biomass of the two major planktivores on the Northeast shelf, Atlantic herring and Atlantic mackerel, declined as a consequence of overfishing in the 1960s and 1970s. Fishing pressure was greatly reduced during the 1980s, and stocks began to increase slowly and reached record high levels in 2004 (Overholtz et al., 2004; Northeast Fisheries Science Center, 2006). As their increase parallels the increase in zooplankton levels, it appears their recovery was fuelled, at least in part, by the high density of prey during the 1990s, indicating predominant bottom-up controls on the Georges Bank foodweb.
Zooplankton levels measured during the first decade of the MARMAP time-series are comparable with values recorded during earlier studies from 1913 to 1975 (Sherman et al., 1983), leading those authors to conclude that the zooplankton component of the shelf ecosystem was coherent, not undergoing any large-scale changes over several decades of the 20th century. In contrast, the current work shows a substantial increase in zooplankton abundance on Georges Bank during the early 1990s that persisted through 2004, providing increased levels of secondary production to higher trophic levels. Proof of this is seen in the recent strong year classes of yellowtail flounder (Limanda ferruginea) and haddock (Melanogrammus aeglefinus) in the region (Sosebee and Cadrin, 2006). This change is potentially linked to upstream processes that may be related to climate variability, indicating that advective linkages and bottom-up processes can combine to drive the large-scale patterns of zooplankton abundance on Georges Bank.
|
Acknowledgements |
|---|
I thank the numerous people who worked hard to collect and process the time-series of zooplankton and environmental measures used in this report. Special thanks are due to David Mountain for providing surface layer temperature and salinity anomalies, and to William Overholtz for providing pelagic fish biomass data. Jack Green, Carolyn Griswold, and Jon Hare provided valuable comments on an earlier draft of the manuscript.
|
References |
|---|
|
TopIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
-
Beaugrand G., Reid P. C., Ibanez F., Lindley J. A., Edwards M. Reorganization of North Atlantic marine copepod biodiversity and climate. Science (2002) 296:1692–1694.
Bray J. R., Curtis J. T. An ordination of the upland forest communities of southern Wisconsin. Ecological Monographs (1957) 27:325–349.[CrossRef]
Clarke K. R. Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology (1993) 18:117–143.[CrossRef][Web of Science]
Clarke K. R., Ainsworth M. A method of linking multivariate community structure to environmental variables. Marine Ecology Progress Series (1993) 92:205–219.[Web of Science]
Clarke K. R., Warwick R. M. Changes in Marine Communities: an Approach to Statistical Analysis and Interpretation. (2001) 2nd. Plymouth: PRIMER-E.
Conversi A., Piontkovski S., Hameed S. Seasonal and interannual dynamics of Calanus finmarchicus in the Gulf of Maine (Northeastern US shelf) with reference to the North Atlantic Oscillation. Deep-Sea Research II (2001) 48:519–530.[CrossRef]
Durbin E. G., Campbell R. G., Casas M. C., Ohman M. D., Niehoff B., Runge J., Wagner M. Interannual variation in phytoplankton blooms and zooplankton productivity and abundance in the Gulf of Maine during winter. Marine Ecology Progress Series (2003) 254:81–100.[Web of Science]
Greene C. H., Pershing A. J. The response of Calanus finmarchicus populations to climate variability in the Northwest Atlantic: basin-scale forcing associated with the North Atlantic Oscillation (NAO). ICES Journal of Marine Science (2001) 57:1536–1544.[CrossRef][Web of Science]
Frank K. T., Petrie B., Choi J. S., Leggett W. C. Trophic cascades in a formerly cod-dominated ecosystem. Science (2005) 308:1621–1623.
Fromentin J-M., Planque B. Calanus and the environment in the eastern North Atlantic. 2. Influence of the North Atlantic Oscillation on C. finmarchicus and C. helgolandicus. Marine Ecology Progress Series (1996) 134:111–118.[Web of Science]
Hurrell J. W. Decadal trends in the North Atlantic Oscillation: regional temperatures and precipitation. Science (1995) 269:676–679.
Kane J. Variability of zooplankton biomass and dominant species abundance on Georges Bank, 1977–1986. Fishery Bulletin US (1993) 91:464–474.
Kane J. Spatial and temporal abundance patterns for the late stage copepodites of Metrdia lucens (Copepoda: Calanoida) in the US northeast continental shelf ecosystem. Journal of Plankton Research (2003) 25:151–167.
Meise C. J., O'Reilly J. E. Spatial and seasonal patterns in abundance and age-composition of Calanus finmarchicus in the Gulf of Maine and on Georges Bank: 1977–1987. Deep-Sea Research II (1996) 43:1473–1501.[CrossRef]
Mountain D. G. Variability of the water properties in NAFO Subareas 5 and 6 during the 1990s. Journal of Northwest Atlantic Fisheries Science (2004) 34:103–112.[CrossRef]
Northeast Fisheries Science Center. (2006) 61. 42nd SAW Assessment Summary Report. Northeast Fisheries Science Center Reference Document 06–01.
Overholtz W. J., Jacobson L. D., Melvin G. D., Cieri M., Power M., Libby D., Clark K. Stock Assessment of the Gulf of Maine—Georges Bank Atlantic Herring Complex, 2003. (2004) 290. Northeast Fisheries Science Center Reference Document 04–06.
Pershing A. J., Greene C. H., Jossi J. W., O'Brien L., Brodziak J., Bailey B. Interdecadal variability in the Gulf of Maine zooplankton community, with potential impacts on fish recruitment. ICES Journal of Marine Science (2005) 62:1511–1523.
Reid P. C., Edwards M., Hunt H. G., Warner A. J. Phytoplankton change in the North Atlantic. Nature (1998) 391:546.[CrossRef]
Sameoto D. Decadal changes in phytoplankton color index and selected calanoid copepods in continuous plankton recorder data from the Scotian Shelf. Canadian Journal of Fisheries and Aquatic Sciences (2001) 58:749–761.
Shannon C. E. A mathematical theory of communication. Bell System Technical Journal (1948) 27:379–423.
Sherman K., Green J. R., Goulet J. R., Ejsymont L. Coherence in zooplankton of a large Northwest Atlantic ecosystem. Fishery Bulletin US (1983) 81:855–862.
Sherman K., Smith W. G., Green J. R., Cohen E. B., Berman M. S., Marti K. A., Goulet J. R. Zooplankton production and the fisheries of the northeastern shelf. In: Georges Bank—Backus R. H., ed. (1987) Cambridge, MA: MIT Press. 268–282. 593 pp.
Smith P. C., Houghton R. W., Fairbanks R. G., Mountain D. G. Interannual variability of boundary fluxes and water mass properties in the Gulf of Maine and on Georges Bank: 1993–97. Deep Sea Research (2001) 48:37–70.[CrossRef]
Sosebee K. A., Cadrin S. X. A Historical Perspective on the Abundance and Biomass of Northeast Demersal Stocks from NMFS and Massachusetts Inshore Bottom Trawl Surveys, 1963–2002. Northeast Fisheries Science Center Reference Document, 06–05 (2006) 200.
Wiebe P., Beardsley R., Mountain D., Bucklin A. U.S. GLOBEC Northwest Atlantic/Georges Bank program. Oceanography (2002) 15:13–29.
Worm B., Myers R. A. Meta-analysis of cod–shrimp interactions reveals top-down control in oceanic food webs. Ecology (2003) 84:162–173.[CrossRef][Web of Science]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  T. Kristiansen, C. Jorgensen, R. G. Lough, F. Vikebo, and O. Fiksen Modeling rule-based behavior: habitat selection and the growth-survival trade-off in larval cod Behav. Ecol., May 1, 2009; 20(3): 490 - 500. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Kane A comparison of two zooplankton time series data collected in the Gulf of Maine J. Plankton Res., March 1, 2009; 31(3): 249 - 259. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Kane and J. Prezioso Distribution and multi-annual abundance trends of the copepod Temora longicornis in the US Northeast Shelf Ecosystem J. Plankton Res., May 1, 2008; 30(5): 619 - 632. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||