ICES Journal of Marine Science: Journal du Conseil Advance Access originally published online on October 3, 2007
ICES Journal of Marine Science: Journal du Conseil 2007 64(9):1617-1626; doi:10.1093/icesjms/fsm138
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Interannual to diurnal variability in the near-surface scattering layer in Drake Passage
1 Scripps Institution of Oceanography, La Jolla, CA 92093-0230, USA
2 British Antarctic Survey, Natural Environment Research Council, High Cross, Madingley Rd, Cambridge CB3 OET, UK
Correspondence to T. K. Chereskin: tel: +1 858 534 6368; fax: +1 858 534 9820; e-mail: tchereskin{at}ucsd.edu
Chereskin, T. K., and Tarling, G. A. 2007. Interannual to diurnal variability in the near-surface scattering layer in Drake Passage. – ICES Journal of Marine Science, 64: 1617–1626.Backscattering strength was estimated from 127 shipboard surveys with an acoustic Doppler current profiler (ADCP) made during Drake Passage transits from 1999 to 2004. The backscattering strength is used to determine the characteristics of the near-surface scattering layer, which south of the Southern Antarctic Circumpolar Current Front (SACCF) is dominated by Antarctic krill (Euphausia superba). Diel vertical migration in the upper 150 m was the dominant variability observed in any single transect. When averaged over depth, there was a well-defined annual cycle in backscattering strength, with a factor of four increase from a late-winter minimum to a spring-summer maximum over a period of four months, followed by a more gentle decline during late summer and autumn. In addition, there were significant differences in scattering strength north and south of the Polar Front (PF) on both seasonal and interannnual time-scales. The average summer maximum to the north of the PF was more than twice the maximum to the south, but the winter minima were about the same. On interannual time-scales, scattering strength south of the PF displayed a negative linear trend primarily attributable to a fourfold decrease in backscattering strength south of the SACCF. No significant long-term trend in the scattering strength north of the SACCF was observed.
Keywords: acoustic backscatter, Drake Passage, long-term changes, regional variations, seasonal variations, Southern Ocean, vertical migration
Received 11 August 2006; accepted 9 July 2007; advance access publication 3 October 2007.
 |
Introduction |
|---|
 Top Introduction MethodsResultsDiscussionReferences |
|---|
Beginning with the study of Flagg and Smith (1989), many investigators have utilized the echo intensity measured by acoustic Doppler current profilers (ADCPs) to examine the distribution and variability of scattering layers in the ocean (e.g. Heywood et al., 1991; Zhou et al., 1994; Ashjian et al., 1998; Brierley et al., 2006). For ADCPs operating at frequencies between 150 and 300 kHz, the primary scatterers are zooplankton, and in these studies, significant correlations were found between backscattering strength estimated from the ADCP and zooplankton abundance estimated from net tows. Although coincident net tows are required to relate scattering strength to biomass directly, the backscatter may be useful in its own right to characterize the scattering layer and to guide biological sampling. The data can give valuable insights into depth distributions, vertical migration behaviours, and even life cycles of dominant backscatterers (Heywood, 1996; Tarling et al., 2001; Cottier et al., 2006).
This study presents a unique time-series of backscattering strength estimated from an uncalibrated, hull-mounted shipboard ADCP installed on the Antarctic supply vessel ARSV "Laurence M. Gould" (LMG) in September 1999 (Chereskin et al., 2000). The LMG is the principal supply vessel for Palmer Station, Antarctica, and it crosses Drake Passage 2–4 times per month (Figure 1). As of January 2005, 127 transects of ocean currents and acoustic backscatter had been collected. Unfortunately, no routine net sampling is conducted from the LMG within Drake Passage, although a number of long-term biological-sampling programmes have been conducted in the coastal waters surrounding the west Antarctic Peninsula from the LMG and other vessels (e.g. Smith et al., 1995; Hofmann et al., 2002; Hewitt et al., 2003; Quetin and Ross, 2003). Because the LMG transducer is uncalibrated, the backscattering-strength time-series presented in this study is relative to an unknown constant, as in the study by Heywood et al. (1991). However, as all cruises used the same transducer, operating under uniform conditions, the relative changes in backscatter over time and space are well-resolved even if the absolute level cannot be determined. Measurements made over a 6-year period, 1999–2004, are used to characterize the scattering layer in Drake Passage, its annual cycle, and interannual variability.
|
|
Methods |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
The vessel sails from Punta Arenas, Chile, and follows a fairly set track over the Patagonian shelf from the Straits of Magellan to the tip of Tierra del Fuego; within Drake Passage the tracks form a fan, spreading south and east (Figure 1). The typical steaming speed is 5 m s–1, and the Drake crossing (1000 km) takes about two days, crossing the Polar Front (PF) about mid-passage. Though irregular in space and time, the LMG dataset provides a time-series of highly resolved (<5 km spatial resolution) backscattering strength sampled continuously along the 127 tracks shown in Figure 1. The ship crosses Drake Passage in all seasons, but the temporal coverage is slightly denser in the austral winter because the LMG does quick turnarounds to Palmer Station (four crossings per month), whereas in summer, the ship tends to stay south for up to one month to support research in the Peninsula area (two crossings per month). Biannually, there is a one-month break for a shipyard period, usually in August.
The transducer, a 153.6 kHz RD Instruments (RDI) narrowband ADCP, is installed in a sonar pod that deflects the ice and bubble-laden boundary layer. The ADCP is nominally configured to transmit a 16-m pulse along four acoustic beams orientated at 30° with respect to the vertical and equally spaced in azimuth. As part of the signal processing for currents, the received echo intensity is maintained at constant amplitude by an automatic-gain-control (AGC) loop (Flagg and Smith, 1989; Chereskin and Harding, 1993). The recorded dataset comprises 300-s averaged profiles of currents and 4-beam averaged AGC levels. The depth of the transducer is 6 m, the "blank-after-transmit" interval is 8 m, and the depth bin is 8 m. In all, 50 x 8-m depth bins were collected. The first depth is centred at 26 m, and the maximum depth is centred at 418 m, although the typical maximum depth of good data is less, 
300 m.
Backscattering strength (Sv) in dB referenced to [m x 4
]–1 was calculated as a function of depth and latitude for all Drake Passage transects using the following equation (RD Instruments, 1998; Urick, 1983):
|
| (1) |
The AGC or echo intensity (E) was measured in counts (0–256) with Er, the reference level or noise count, estimated from examining the constant level measured at the end of the profile in deep water, Er = 18. Echo-intensity counts relative to the noise level were converted from internal logarithmic units to dB using the conversion factor Kc = 0.436 dB per count. This factor depends on the temperature of the electronics chassis, which is located in a climate-controlled laboratory maintained at 19°C. The conversion factor varies by only 0.34% per °C, and hence a variation in temperature of a few degrees is a relatively small source of error. Ocean sound speed, c, was calculated as a function of transducer temperature Tx (°C), which is part of the recorded dataset, and a constant salinity appropriate for the surface layer of Drake Passage (34.15). The echo intensity was corrected for spherical spreading and absorption; the attenuation of sound, 
, was calculated as a function of Tx at 153.6 kHz, a salinity of 34.15, and a pH of 8 appropriate for seawater (Ainslie and McColm, 1998). P is the pulse length (16 m). Two variables that would be determined from a system calibration are K1, the power (in W) transmitted into the water, and K2, a dimensionless system noise factor. RDI no longer provides factory calibrations for narrowband systems, and therefore nominal values K1 = 10 and K2 = 2 were used here. Hence, the calculated Sv is relative rather than absolute. Ks = 4.17 x 105 is a frequency-dependent, system constant (RD Instruments, 1998). The slant range, R, to depth bin n was calculated from
|
| (2) |
the angle with respect to vertical. D/4 shifts the depth-bin centre, because the AGC is measured during the last quarter of the sampling bin (RD Instruments, 1998). The scale factor c/1475.1 corrects for the constant sound speed used by the firmware. Whenever averages of Sv were computed for this study (e.g. in latitude, depth, or time), the Sv estimates were first converted to linear units, then averaged and converted back to dB. A despiking algorithm was applied before averaging to remove a few extremely large spikes (about 25 standard deviations from the mean) that occurred in six sections and were large enough to bias the section mean. The spikes were localized in space by being near the coast, and could also have been removed by truncating the section a little earlier. Heywood et al. (1991) discussed a bias that they observed in the echo intensity that was correlated to ship speed. They observed that the noise level increased and the surface-signal level decreased when the ship was underway relative to on station, and they used only station data. This study is based predominantly on underway data because the LMG rarely occupies stations during Drake Passage transits. A comparison of LMG underway and on-station data did not show a correlation between echo intensity and ship speed.
The amplitude of the bottom echo from the repeated transects across the Patagonian shelf were used to establish a baseline for transducer stability, assuming that the true bottom amplitude at a fixed location is time-invariant. Water depth along the shelf transit is 70 m and varies gradually. Backscatter strength was estimated for the bin of maximum amplitude using Equation (1). Time-series were formed at a number of fixed locations and examined for mean, standard deviation, and trend. The time-series at 53.55°S is typical (Figure 2). The variability is due to non-exact repeating ship tracks and to the coarse resolution in determining the depth of the maximum amplitude from the 8-m water-track bin. The hypothesis of zero trend is valid at 99% confidence, and only 8% of the variance is explained by a linear fit. The standard deviation is 1 dB, and the standard error in the mean is 0.1 dB. We conclude that the transducer is stable to 1 dB.
|
The major fronts of the Antarctic Circumpolar Current (ACC), the Subantarctic Front (SAF), the PF, and the Southern ACC Front (SACCF) are potentially important in structuring zooplankton communities in Drake Passage, because they are water-mass boundaries and are associated with strong, vertically coherent velocity jets. Orsi et al. (1995), hereafter referred to as O95, determined the front locations using climatological hydrographic data (Figure 1). Within Drake Passage, high-resolution repeat observations made from the LMG of upper-ocean temperature using expendable bathythermographs (XBTs) (Sprintall, 2003), and surface-layer currents from the ADCP (Lenn et al., 2007) have been used to refine the mean front locations. The mean SAF as determined from XBT and ADCP observations lies north of the O95 location and closely follows the 1000-m bathymetric contour of the Patagonian shelf break (Lenn et al., 2007). Hence, the bulk of our data lie south of the SAF. The mean PF as determined from XBT and ADCP data is similar in orientation and location to O95 but on average lies
50 km to the north of it (Lenn et al., 2007). To examine differences in backscattering strength north and south of the PF, the O95 PF location was determined for each transect. On cruises where XBT observations were made, we also determined the synoptic PF location. The SACCF determined from ADCP observations is in good agreement with the O95 location (Lenn et al., 2007). Most of the LMG tracks lie west of the Shackleton Fracture Zone, where O95 indicates a distinct northward meander (Figure 1) that complicates finding a unique SACCF crossing on some transects. To examine differences in backscattering strength north and south of the SACCF, the data were binned into a pair of regions enclosed by dashed lines in Figure 1 encompassing roughly equal numbers of observations. |
Results |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Figure 3 shows a single section from a northbound crossing in March 2000, and Figure 4 shows a typical sequence from the crossings made in 2001 (Table 1). The intersection of the mean O95 PF with each transect is indicated (Figures 3
–6). Also indicated is the time of day, to assess the importance of diel vertical migration (DVM), a known behaviour in which zooplankton descend to greater depths to escape predation during daylight and ascend to feed at night (e.g. Bollens et al., 1991; Wilson and Firing, 1992; Ashjian et al., 1998; Lenn et al., 2003). Daytime and night-time were determined from calculating the times of sunrise and sunset at 65°W, 60°S.
|
|
|
|
|
A clear example of a daytime descending layer is seen during lg0003 (Figure 3). North of the PF, the daytime layer has an Sv level of about –87 dB and descends from 26 m to a depth of 150 m. A descending daytime layer is also observed to the south of the PF, but there the Sv level is weaker, about –94 dB. By night, the overall backscatter level is higher, but distinct layers are not readily discernible. The maximum profiling depth is greater at that time of day because of deep scatterers that rise to lie within the ADCP's range. Note, for example, the very pronounced night-time scattering layer south of the PF, in the depth range 150–350 m (Figure 3). In general, the diurnal variation in Sv is one of the most striking features of the observations, and the features described for lg0003 can be seen in most other transects (Figure 4).
Seasonal modulation of the diurnal cycle is also readily apparent. The day/night difference is most notable in winter, when nights are long and the daytime minimum in Sv level routinely reaches the minimum level contoured, –115 dB (Figure 4). In summer, in addition to shorter nights, the overall mean Sv level is higher. The night-time deep-scattering maximum also seems to have seasonal variability. It first appears on the southern boundary of Drake Passage in late summer (e.g. lg0003 in Figure 3 and lg0103 in Figure 4) and subsequently is seen at progressively more northerly latitudes through autumn and winter, and is in evidence on the northern boundary of Drake Passage in spring (lg0108a in Figure 4). This deep night-time signal occurs in all sampled years.
Depth-averaged Sv was examined to define the boundaries between regions with different characteristic Sv levels (Figures 5 and 6). Sv was averaged from the first recorded bin, at 26 m, to 154 m. The cut-off depth is a depth of consistently good returns that includes DVM of the surface scattering layer, except for the upper 26 m, but excludes the deeper scattering layer that is only sampled at night. Comparison of the depth-averaged signal with a fixed level of –100 dB (Figures 5 and 6) illustrates that the overall signal level is higher in summer than other seasons. The most significant changes in Sv level in any given transect occur at sunrise and sunset. Also apparent is the tendency for higher Sv in the coastal regions at both the northern and southern boundaries of Drake Passage.
To examine the annual cycle, monthly averages of Sv, averaged in depth to 154 m and in latitude across the Drake Passage, were calculated. Figure 7 compares the annual cycles for the individual years (2000–2004) and the 6-year mean (1999 only partially sampled). Errors in the mean are standard errors. A spring transition is readily apparent, with Sv increasing by an average of 6 dB from August to November, typically within 2 months. The spring-summer maximum erodes gradually through summer and autumn, reaching a minimum in late winter (August). July 2001 had the lowest monthly Sv level; in 2004, the first half of the year had normal levels, but the spring maximum was anomalously low, by about 2 dB.
|
Synoptic PF crossings were examined for the subset of cruises (26) where XBT observations were made, and to avoid confusing geographical variability with DVM, the crossing did not coincide with local sunrise or sunset. No significant gradient in Sv was observed. However, despite the lack of any abrupt change at the PF, there were quantifiable differences in Sv to the north and south of it (Figures 8 and 9). Averages of depth-averaged Sv north of the O95 PF were higher during summer in all years compared with averages of Sv south of it (Figure 8a). Because of potential bias from the diurnal cycle, the calculation was repeated using only night-time data (Figure 8b) (note that although this removes bias attributable to diel migration, it also reduces the spatial sampling by 50%). The patterns are robust, with the northern Drake Passage consistently higher than the southern Drake Passage. Sv averaged over the northern Drake Passage is
5 dB higher than the southern Drake Passage during summer (Figure 8). The only times when northern Drake Passage Sv levels drop to levels within 1–2 dB of those observed south of the PF are at the end of winter, when levels everywhere are at a seasonal low (Figure 9). The average winter minima north and south of the PF are about the same, differing by an amount comparable with the standard error in the mean (Figure 8).
|
|
In order to examine interannual variability, anomaly time-series were calculated. Four such series were considered: north/south of the PF using all and night-time-only data in turn. In each case, the anomaly was relative to the 6-year monthly means of the respective time-series. South of the PF, the backscatter anomaly has a negative trend, significant at 99% confidence, of –0.04 dB per month, corresponding to a twofold (3 dB) decline over 6 years (Figure 10). The percentage of the variance in the anomalies explained by the linear fit is 28% (r2 = 0.28). North of the PF, not shown, the sign of the slope was not significantly different from zero.
|
The data south of the PF were binned further into a pair of regions on either side of the SACCF (Figure 1). Two series were considered: north and south of the SACCF, respectively, using all data. In each case, the anomaly is defined relative to the 6-year monthly mean of the region. The results indicate that there is a significant (99% confidence, r2 = 0.30) negative trend south of the SACCF of –0.08 dB per month, corresponding to a 6 dB decline over 6 years (Figure 11b). North of the SACCF the fit is not significant (Figure 11a). Hence, the decline south of the PF can be further isolated to south of the SACCF.
|
|
Discussion |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
Backscattering strength estimated from 127 shipboard ADCP surveys made during all seasons over a 6-year period was used to examine variability in the scattering layer in the upper 300 m of Drake Passage. Each Drake Passage crossing took about two days, and DVM was the dominant variability observed in any single transect. A shallow scattering layer descended to
150 m during the day. At night, there was an increase in overall backscatter level and an increase, to about 300 m, in the depth range profiled, consistent with the ascent and dispersion of scatterers from below the daytime-profiling range. There was a well-defined annual cycle in depth-averaged backscatter, with the average scattering strength increasing by a factor of four from a late-winter minimum to a spring-summer maximum over a period of four months, followed by a more gentle decline during late summer and autumn. In addition, there were significant differences in scattering levels north and south of the PF. Backscattering strength in the northern Drake Passage was consistently higher than that in the southern Drake Passage in all months, especially in summer. The average summer maximum backscatter level north of the PF was more than twice the maximum level south of the PF, but the winter minima were about the same. Interannual variability was also observed, with a fourfold decrease in backscatter south of the SACCF over six years.
Species composition
A comprehensive net-sampling campaign was carried out in the Atlantic sector of the Southern Ocean by the "Discovery" expeditions in the first part of the last century (Kemp and Bennet, 1929). From those samples, Mackintosh (1934) categorized the following species as "abundant": the calanoid copepods, Calanus acutus, C. propinquus, and Rhincalanus gigas, the euphausiids Euphausia superba and Thysanoessa spp., the thecosome pteropod Limacina balea (now called L. retroversa australis), and the Chaetognatha. Other species that fell under the "numerous" category were Euphausia frigida and L. helicina, and the hyperiid amphipod Themisto gaudichaudii. Euphausia triacantha and E. vallentini were noted as being abundant in patches to the north of the PF. There have been very few recent net-catch surveys carried out in the north of the Drake Passage. One of the most comprehensive carried out in the southern Drake Passage in recent times was by Ward et al. (2004), who found patterns of abundance in accord with those reported by Mackintosh (1934), with calanoid copepods being the most abundant organisms, followed by Limacina helicina. Most other recent net-sampling surveys in Drake Passage have only covered the very southern extremity (Piatkowski, 1989; Siegel, et al., 2002; Hewitt et al., 2003), where E. superba frequently dominates zooplankton biomass.
Crustaceans <5 mm will have a weak target strength at 150 kHz (<–110 dB, Stanton et al., 1994; Greene et al., 1998), making it unlikely that the calanoid copepods contribute to backscattering patterns observed in the present study. Euphausiids are larger (10–60 mm) and have a higher target strength at 150 kHz (Buchholz et al., 1995; Tarling et al., 2001). Although thecosome pteropods such as L. helicina and L. retroversa australis are comparatively small (10–20 mm; van der Spoel and Dadon, 1999), their hard aragonite shells make them a very efficient scatterer of sound (Stanton et al., 1994). Measured as echo-energy per unit biomass, the scattering strength of thecosomes was 60 times that of a euphausiid at 200 kHz (Stanton et al., 1994). Per individual, a euphausiid would have a target strength at 150 kHz of around –80 dB, whereas that of a thecosome pteropod would be around –90 dB (Tarling et al., 2001). The respective target strengths and abundances of euphausiids and thecosome pteropods make them the most likely contributors to the backscatter patterns observed in this study.
Vertical distribution and DVM
DVM in scattering layers was frequently observed in the present study. Euphausiids such as Thysanoessa spp. and E. triacantha are noted for their pronounced DVM behaviour. The former migrates vertically between 150 m and the surface over a diel cycle (Hardy and Gunther, 1935; Piatkowski et al., 1994), whereas the latter may reside as deep as 500 m by day before migrating to the surface layers at night (Baker, 1959; Piatkowski et al., 1994). The DVM behaviour of E. superba is more erratic, with some swarms showing a regular DVM cycle and others hardly ever migrating vertically (Hardy and Gunther, 1935; Everson, 1983). Godlewska (1996) reports that E. superba can migrate vertically between the surface and 250 m, and Demer and Hewitt (1995) reported that the biomass of E. superba peaked at the two depths of 12.5 m and 62.5 m. Hardy and Gunther (1935) provide one of the only considerations of vertical migration of L. helicina, which occupied depths of around 100 m by day, and around 20 m at night. With respect to the acoustic patterns observed in the present study, it is likely that the scattering layer migrating from the surface layer at night to around 100–150 m by day (Figure 3) consists of a combination of euphausiids and pteropods. The euphausiid E. triacantha may be partly responsible for some of the scattering layers ascending from greater depths.
Some of the transects, especially those carried out in winter, showed evidence of a deep scattering layer, usually below 300 m. Myctophids have frequently been identified as the main contributor to these layers (Piatkowski et al., 1994; Pusch et al., 2004), which appear over a large part of the transects, demonstrating that this group is both abundant and widespread throughout this region of the Southern Ocean. The species composition of these layers is likely to change over the course of the transect, with Protomyctophum tenisoni and Gymnoscopelus fraseri being typically north of the PF, P. anderssoni, P. parallelum, P. bolini, Electrona carlsbergi, and Lampanyctus achirus near the PF, and E. antarctica, G. braueri, and G. opisthopterus to the south of the front (McGinnis, 1982). As with the euphausiids, myctophids are noted for their DVM behaviour, migrating from daytime depths of below 300 m to just below the surface layers at night (Piatkowski et al., 1994). The depth of scattering layers has frequently been related to the locations of particular isolumes in the water column (Frank and Widder, 2002). The greater tendency of this layer to be observed during winter may be a function of the decreasing levels of irradiance bringing the optimal isolume closer to the surface.
North–south trends in backscatter
Depth-averaged backscatter tended to be higher in the north of the transect than the south, but note that the deep-scattering layer was excluded from this calculation. This trend agrees with patterns observed in the "Discovery" expedition net catches taken along a similar transect line, with the relative quantity of plankton being between 20 and 100 times higher in the northern part than in the south (Mackintosh, 1934). Ward et al. (2004) found that copepods as well as Limacina and Thysanoessa were particularly abundant towards the north of the region, although they did not sample beyond the PF. The abundance of E. triacantha also peaks around the PF (Baker, 1959), which may increase backscattering to the north even further. At the northern side of the PF, it is likely that euphausiids and thecosome pteropods become even more abundant, especially E. vallentini and L. retroversa australis. The abundance of the former is demonstrated by the fact that it is the main component in the diet of many Subantarctic higher predators (Ridoux, 1988). The latter reaches concentrations of between 67 and 1000 individuals m–3 (Boltovskoy, 1971a, b; Dadon, 1990a, b), orders of magnitude greater than typical concentrations of the dominant congener to the south of the PF, L. helicina (Ward et al., 2004).
Seasonal trends in backscatter
A prominent feature in the backscattering patterns in all of the years studied in the present investigation was the large seasonal oscillation in biomass, with a peak in summer and a trough in winter. This is likely to result from two major processes: first, recruitment and mortality of individuals in populations over the course of the year; and second, seasonal change in vertical distribution.
In the first case, euphausiids in this region display multiyear life cycles, meaning that the adults persist for a number of years and population numbers are relatively stable. However, the numbers will oscillate over the course of the year as recruitment into the adult population peaks during the productive summer and mortality increases during the challenging winter (Siegel, 2000a, b). Limacina normally has a 1-year life cycle, with overwintering juveniles developing into adults to spawn the new generation before dying in autumn (Gannefors et al., 2005). This means that the population will increase in both abundance and individual body size over the course of summer. The combination of these effects will increase backscatter in line with our observations.
The second major process, seasonal change in vertical distribution, occurs when certain species enter a state of diapause and descend to depths beyond 1000 m during the winter (Schnack-Schiel, 2001). At present, most examples of this behaviour have been established for calanoid copepods, although Southern Ocean euphausiids are likely to be exemplars also.
Reports of seasonal horizontal migrations have mostly been restricted to E. superba. Here, spawning females migrate to the outer shelf to spawn in summer and return to the shelf to spend the winter under ice (Siegel, 1988, 2000a; Nicol, 2006). This would mean that they would no longer be observed along the respective survey tracks during winter, so decreasing backscatter along its southern fringes.
Interannual decline in backscatter
The present study found there to be a significant decline in backscatter south of the PF over the six years of observation, peaking in 1999 and dropping to the lowest levels in 2006. Most notably, the largest decline was seen south of the SACCF, where the biomass is dominated by E. superba (Siegel, 1988). Euphausia superba goes through peaks and troughs of biomass over a 6–8-year cycle (Hewitt et al., 2003; Quetin and Ross, 2003), believed to result from the influence of environmental factors on larval recruitment (Quetin and Ross, 2003). At the Peninsula, for instance, Siegel (1988) put forward the theory that recruitment of larvae was large when the extent of ice reached its maximum. Quetin and Ross (2003) found that larval recruitment was good in years when conditions were around their average for the region. Patterns in the timing and extent of ice formation have been correlated with El Niño cycles (Kwok and Comiso, 2002), although the exact mechanisms remain unclear (Turner, 2004).
In respect of the pattern observed in the present study, it is possible that 1999 corresponds with the start of a 6–8-year recruitment cycle, such that a decline in krill biomass is observed over the period of study. Hewitt et al. (2003) documented a peak in krill density in 1997/1998 and then a decline through 1999. However, they also observed krill biomass to increase in 2000 and again in 2004, contrary to present observations. This suggests either that the LMG transect is not sampling the same krill population as that described by them or that the significant decline in backscatter observed over the six-year study period is being caused by different organisms. At present, it is impossible to distinguish between these two scenarios.
It is notable in this regard that populations of planktivorous higher predators (e.g. Adelie penguins, Pygoscelis adeliae) at nearby islands have been declining over a number of years (Forcada et al., 2006). Such planktivorous predators have the capacity to switch between prey items to buffer against the episodic cycles of abundance in their prey species over shorter time-scales, ensuring that their population sizes remain relatively stable (Forcada et al., 2005). Long-term declines in population size point to a more significant environmental shift. The fact that both planktivores and acoustic backscatter have declined over similar periods suggests that the wider zooplankton community is currently in a phase of decline in this region of the Southern Ocean. This is especially true in regions where Antarctic krill dominates, for example south of the SACCF. Such a decline may be a result of recent warming trends in the surface waters of this region (Meredith and King, 2005) as well as the changing ice dynamics (Vaughan et al., 2003).
|
Acknowledgements |
|---|
The National Science Foundation (NSF) Office of Polar Programs (OPP) sponsored TKC and this research through grants OPP-9816226 and OPP-0338103. The contribution of GAT was carried out as part of the FLEXICON project, within the DISCOVERY 2010 programme at British Antarctic Survey. Thanks go to Eric Firing for his comments and suggestions on an earlier draft, to Joel Gast for advice on the backscatter calculation, to Janet Sprintall for providing XBT data, and to the captain and crew of the ARSV "Laurence M. Gould" and Raytheon Polar Services Corporation for their excellent technical and logistic support.
|
References |
|---|
|
TopIntroductionMethodsResultsDiscussionReferences |
|---|
-
Ainslie M. A., McColm J. G. A simplified formula for viscous and chemical absorption in sea water. Journal of the Acoustical Society of America (1998) 103:1671–1672.[CrossRef][Web of Science]
Ashjian C. J., Smith S. L., Flagg C. N., Wilson C. Patterns and occurrence of diel vertical migration of zooplankton biomass in the Mid-Atlantic Bight described by an acoustic Doppler current profiler. Continental Shelf Research (1998) 18:831–858.[CrossRef][Web of Science]
Baker A. de C. Distribution and life history of Euphausia triacantha. "Discovery" Reports (1959) 29:309–340.
Bollens S. M., Frost B. W., Thoreson D. S., Watts S. J. Diel vertical migration in zooplankton–field evidence in support of the predator-avoidance hypothesis. Hydrobiologia (1991) 234:33–39.[Web of Science]
Boltovskoy D. Contribucíon al conocimiento de los pteropodos thecosomados sobre la plataforma continental bonaerense. Revista del Museo de la La Plata Seccíon Zoologia (1971) a 11:121–136.
Boltovskoy D. Pteropodos thecosomados del atlantico sudoccidental. Malacologia (1971) b 11:121–140.
Brierley A. S., Saunders R. A., Bone D. G., Murphy E. J., Enderlein P., Demer D. A., Conti S. G. Use of moored acoustic instruments to measure short-term variability in abundance of Antarctic krill. Limnology and Oceanography: Methods (2006) 4:18–29.[Web of Science]
Buchholz F., Buchholz C., Reppin J., Fisher J. Diel vertical migrations of Meganyctiphanes norvegica in the Kattegat: comparison of net catches and measurements with acoustic Doppler current profilers. Helgoländer Wissenschaftliche Meeresuntersuchungen (1995) 49:849–866.
Chereskin T. K., Firing E., Hummon J. High-resolution ADCP measurements across Drake Passage. EOS Transactions (2000) 83:47.
Chereskin T. K., Harding A. J. Modelling the performance of an acoustic Doppler current profiler. Journal of Atmospheric and Oceanic Technology (1993) 10:41–63.[CrossRef][Web of Science]
Cottier F. R., Tarling G. A., Wold A., Falk-Petersen S. Unsynchronized and synchronized vertical migration of zooplankton in a high Arctic fjord. Limnology and Oceanography (2006) 51:2586–2599.[Web of Science]
Dadon J. R. Annual cycle of Limacina retroversa in Patagonian waters. American Malacological Bulletin (1990) a 8:77–84.[Web of Science]
Dadon J. R. Distribution and annual cycle of Limacina retroversa Fleming 1823 in Patagonian waters. Report of the Western Society of Malacologists (1990) b 22:2.
Demer D. A., Hewitt R. P. Bias in acoustic-biomass estimates of Euphausia superba due to diel vertical migration. Deep-Sea Research I (1995) 42:455–475.[CrossRef]
Everson I. Variations in vertical distribution and density of krill swarms in the vicinity of South Georgia. National Institute of Polar Research Memoirs (1983) 27(Special Issue):84–92.
Flagg C. N., Smith S. L. On the use of the acoustic Doppler current profiler to measure zooplankton abundance. Deep-Sea Research I (1989) 36:455–474.
Forcada J., Trathan P. N., Reid K., Murphy E. J. The effects of global-climate variability in pup production of Antarctic fur seals. Ecology (2005) 86:2408–2417.[CrossRef][Web of Science]
Forcada J., Trathan P. N., Reid K., Murphy E. J., Croxall J. P. Contrasting population changes in sympatric penguin species in association with climate warming. Global Change Biology (2006) 12:411–423.[CrossRef][Web of Science]
Frank T. M., Widder E. A. Effects of a decrease in downwelling irradiance on the daytime vertical distribution patterns of zooplankton and micronekton. Marine Biology (2002) 140:1181–1193.[CrossRef]
Gannefors C., Boer M., Kattner G., Graeve M., Eiane K., Gulliksen B., Hop H., et al. The Arctic sea butterfly Limacina helicina: lipids and life strategy. Marine Biology (2005) 147:169–177.[CrossRef]
Godlewska M. Vertical migrations of krill (Euphausia superba Dana). Polskie Archiwum Hydrobiologii (1996) 14:9–63.
Greene C. H., Wiebe P. H., Pershing A. J., Gal G., Popp J. M., Austin T. C., Bradley A. M., et al. Assessing the distribution and abundance of zooplankton: a comparison of acoustic and net-sampling methods with D-BAD MOCNESS. Deep-Sea Research II (1998) 45:1219–1237.[CrossRef]
Hardy A. C., Gunther E. R. The plankton of the South Georgia whaling grounds and adjacent waters, 1926–1927. "Discovery" Reports (1935) 11:1–456.
Hewitt R. P., Demer D. A., Emery J. H. An eight-year cycle in krill-biomass density inferred from acoustic surveys conducted in the vicinity of the South Shetland Islands during the austral summers of 1991/92 through 2001/02. Aquatic Living Resources (2003) 16:205–213.[CrossRef][Web of Science]
Heywood K. J. Diel vertical migration of zooplankton in the Northeast Atlantic. Journal of Plankton Research (1996) 18:163–184.
Heywood K. J., Scrope-Howe S., Barton E. D. Estimation of zooplankton abundance from shipborne ADCP backscatter. Deep-Sea Research I (1991) 38:677–691.
Hofmann E. E., Klinck J. M., Costa D. P., Daly K. L., Torres J. J., Fraser W. R. U. S. Southern Ocean Global Ocean Ecosystems Dynamics Program. Oceanography (2002) 15:69–74.
Kemp S., Bennet A. G. "Discovery" investigations: objects, equipment and methods. "Discovery" Reports (1929) 1:1–151.
Kwok R., Comiso J. C. Southern Ocean climate and sea-ice anomalies associated with the Southern Oscillation. Journal of Climate (2002) 15:487–501.[CrossRef][Web of Science]
Lenn Y-D., Chereskin T. K., Glatts R. C. Seasonal to tidal variability in currents, stratification and acoustic backscatter in an Antarctic ecosystem at Deception Island. Deep-Sea Research II (2003) 50:1665–1683.[CrossRef]
Lenn Y-D., Chereskin T. K., Sprintall J., Firing E. Mean jets, mesocale variability and eddy-momentum fluxes in the surface layer of the Antarctic Circumpolar Current in Drake Passage. Journal of Marine Research (2007) 65:27–58.[CrossRef][Web of Science]
Mackintosh N. A. Distribution of the macroplankton in the Atlantic sector of the Antarctic. "Discovery" Reports (1934) 9:65–160.
McGinnis R. F. Biogeography of lanternfishes (Myctophidae) south of 30°S. In: Biology of the Antarctic Seas XII, 35—Pawson D. L., ed. (1982) Washington, DC: American Geophysical Union. 110.
Meredith M. P., King J. C. Rapid climate change in the ocean west of the Antarctic Peninsula during the second half of the 20th century. Geophysical Research Letters (2005) 32. doi:10.1029/2005GL024042.
Nicol S. Krill, currents and sea-ice: Euphausia superba and its changing environment. Bioscience (2006) 56:111–120.[CrossRef][Web of Science]
Orsi A. H., Whitworth T. III., Nowlin W. D Jr. On the meridional extent and fronts of the Antarctic Circumpolar Current. Deep-Sea Research I (1995) 42:641–673.[CrossRef]
Piatkowski U. Macroplankton communities in Antarctic surface waters—spatial changes related to hydrography. Marine Ecology Progress Series (1989) 55:251–259.[CrossRef][Web of Science]
Piatkowski U., Rodhouse P. G., White M. G., Bone D. G., Symon C. Nekton community of the Scotia Sea as sampled by the RMT-25 during austral summer. Marine Ecology Progress Series (1994) 112:13–28.[CrossRef][Web of Science]
Pusch C., Hulley P. A., Kock K-H. Community structure and feeding ecology of mesopelagic fishes in the slope waters of King George Island (South Shetland Islands, Antarctica). Deep-Sea Research I (2004) 51:1685–1708.[CrossRef]
Quetin L. B., Ross R. M. Episodic recruitment in Antarctic krill, Euphausia superba, in the Palmer LTER study region. Marine Ecology Progress Series (2003) 259:185–200.[CrossRef][Web of Science]
RD Instruments. Calculating absolute backscatter in narrowband ADCPs. (1998) Field Service Technical Paper 003 (FST-003).
Ridoux V. Subantarctic krill, Euphausia vallentini Stebbing, preyed upon by penguins around Crozet Islands (Southern Indian Ocean): population structure and annual cycle. Journal of Plankton Research (1988) 10:675–690.
Schnack-Schiel S. B. Aspects of the study of the life cycles of Antarctic copepods. Hydrobiologia (2001) 453:9–24.[CrossRef][Web of Science]
Siegel V. A concept of seasonal variation of krill (Euphausia superba) distribution and abundance west of the Antarctic peninsula. In: Antarctic Ocean and Resources Variability—Sahrhage D, ed. (1988) Springer, Berlin. 304.
Siegel V. Krill (Euphausiacea) demography and variability in abundance and distribution. Canadian Journal of Fisheries and Aquatic Sciences (2000) a 57:151–167.
Siegel V. Krill (Euphausiacea) life history and aspects of population dynamics. Canadian Journal of Fisheries and Aquatic Sciences (2000) b 57:130–150.
Siegel V., Bergstrom B., Muhlenhardt-Siegel U., Thomasson M. Demography of krill in the Elephant Island area during summer 2001 and its significance for stock recruitment. Antarctic Science (2002) 14:162–170.[CrossRef][Web of Science]
Smith R. C., Baker K. S., Fraser W. R., Hofmann E. E., Karl D. M., Klinck J. M., Quetin L. B., et al. The Palmer LTER: a long-term ecological research program at Palmer Station, Antarctica. Oceanography (1995) 8:77–86.
Sprintall J. Seasonal to interannual upper-ocean variability in Drake Passage. Journal of Marine Research (2003) 61:25–57.
Stanton T. K., Wiebe P. H., Chu D., Benfield M. C., Scanlon L., Martin L., Eastwood R. L. On acoustic estimate of zooplankton biomass. ICES Journal of Marine Science (1994) 51:505–512.
Tarling G. A., Matthews J. B. L., David P., Guerin O., Buchholz F. The swarm dynamics of northern krill (Meganyctiphanes norvegica) and pteropods (Cavolinia inflexa) during vertical migration in the Ligurian Sea observed by an acoustic Doppler current profiler. Deep-Sea Research I (2001) 148:1671–1686.
Turner J. The El N
no-Southern Oscillation and Antarctica. International Journal of Climate (2004) 24:1–31.[CrossRef]
Urick R. J. Principles of Underwater Sound. (1983) New York: McGraw Hill. 384.
van der Spoel S., Dadon J. R. Pteropoda. In: South Atlantic Zooplankton, 1, pp. 649–706—Boltovskoy D, ed. (1999) Backhuys, Leiden, The Netherlands. 1705.
Vaughan D. G., Marshall G. J., Connolley W. M., Parkinson C., Mulvaney R., Hodgson D. A., King J. C., et al. Recent, rapid, regional-climate warming on the Antarctic Peninsula. Climatic Change (2003) 60:243–274.[CrossRef][Web of Science]
Ward P., Grant S., Brandon M., Siegel V., Loeb V., Griffiths H. Mesozooplankton community structure in the Scotia Sea during the CCAMLR 2000 survey: January-February 2000. Deep-Sea Research II (2004) 51:1351–1367.[CrossRef]
Wilson C. D., Firing E. Sunrise swimmers bias acoustic Doppler current profiles. (1992) 39:885–892.
Zhou M., Nordhausen W., Huntley M. ADCP measurements of the distribution and abundance of euphausiids near the Antarctic Peninsula in winter. Deep-Sea Research I (1994) 41:1425–1445.[CrossRef]
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||