|USFDC Home | USF Electronic Theses and Dissertations||| RSS|
This item is only available as the following downloads:
Characterization of the Western An tarctic Peninsula Ecosystem: Environmental Controls on the Zooplankton Community by Marina Marrari A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy College of Marine Science University of South Florida Major Professor: Kendra L. Daly, Ph.D. Chuanmin Hu, Ph.D. Gary Mitchum, Ph.D. Luis Garca-Rubio, Ph.D. Meng Zhou, Ph.D. Date of Approval: June 30, 2008 Keywords: Southern Ocean, Marguerite Bay, SeaWiFS chlorophyll, krill, Euphausia superba Copyright 2008, Marina Marrari
DEDICATION to my husband, Christian my daughter, Valentina and my parents Juan and Marcela
ACKNOWLEDGEMENTS I would like to thank my advisor, Dr. Kendra Daly, for making herself permanently available and for her continued su pport. Collaborati on with Dr. Chuanmin Hu was crucial for this research, and to hi m I feel greatly indebt ed. I also extend a special thank you to my committee, Drs Gary Mitchum, Luis Garca-Rubio and Meng Zhou. I am grateful to the many people w ho participated in SO GLOBEC cruises, collected and processed data : M. Zhou, Y. Wu, R. Dorland, D. Mertes, and J. Smith collected MOCNESS samples, J. Zimmerma n, A. Timonin and T. Semenova analyzed zooplankton samples, K. Arrigo and G. van Dijken supplied sea ice images, T. Bolmer provided bathymetry data, E. Chapman made i ce edge data available. Data from the LTER Program provided by Drs R. Smith, R. Ross L. Quetin and M. Vernet. J. Smith and R. Conmy helped with collection and an alysis of CDOM samples. Financial support provided by the Fulbright Commission, Seasp ace Fellowship, Carl Riggs Fellowship, and Parrot Heads in Paradise Fellowship. F unding also provided by NSF and US NASA. Data from the LTER data archive supported by NSF (OPP-9011927). A very special thank you to my friends in Argentina for their long distance encouragement, and especially to my family for always being r eady to jump on a plane when needed. I am forever grateful to Chris for supporting me in following my dreams, even when that meant moving across continents. I will always be thankful for your love and friendship.
NOTE TO THE READER The original of this document contains colo r that is necessary fo r understanding the data. The original dissertation is on file with the USF library in Tampa, Florida.
i TABLE OF CONTENTS LIST OF TABLES iv LIST OF FIGURES v ABSTRACT .. ix CHAPTER 1: THE WESTERN ANTARC TIC PENINSULA: PHYSICAL AND BIOLOGICAL SETTING 1 CHAPTER 2: VALIDATION OF SEAWIFS CHLOROPHYLL A CONCENTRATIONS IN THE SOUTHERN OCEAN: A REVISIT Introduction .. 10 Methods 14 Results .. 19 Discussion 26 CHAPTER 3: SPATIAL AND TEMPOR AL VARIABILITY OF SEAWIFS CHLOROPHYLL A DISTRIBUTIONS WEST OF THE ANTARCTIC PENINSULA: IMPLICATIONS FOR KRILL PRODUCTION Introduction 30 Methods .. 34 Results . 43 Discussion 55 CHAPTER 4: PHYSICAL AND BIOLOGICAL CONTROLS ON INTERANNUAL VARIABILITY OF ZOOPLANKTON IN MARGUERITE BAY, WESTERN ANTARCTIC PE NINSULA, AUSTRAL FALL 2001 AND 2002
ii Introduction . 64 Methods 68 Results Abundance and Percent Contribution 72 Vertical Distribution .. 89 Horizontal Distribution .. 96 Fall Environmental Parameters .. 98 Summer Chlorophyll Concentratio ns and Krill Recruitment 101 Discussion Composition and Abundance of Z ooplankton in Marguerite Bay 106 Variability in Euphausiid Life History Strategies 111 Summer Chlorophyll and Zoopla nkton Population Response . 114 Controls on Zooplankton Spatial Pa tterns in Marguerite Bay .. 116 Relationship between Zooplankton and Fall Environmental Parameters . 120 Summary .. 121 CHAPTER 5: SUMMARY AND CONCLUDING REMARKS .. 123 REFERENCES CITED . 126 APPENDICES Appendix 1 146 Appendix 2 . 150 Appendix 3 . 154 Appendix 4 . 160
iii ABOUT THE AUTHOR ... End Page
iv LIST OF TABLES Table 2.1. Statistics for the comparisons between Ca SWF and in situ Ca (Ca FLUOR, Ca HPLC) . 20 Table 2.2. Statistics for the comparisons between Ca FLUOR and Ca HPLC (mg m-3) .. 24 Table 3.1. Years of elevated kril l recruitment between 1975 and 2002 . 54 Table 4.1. Copepod abundance (ind m-2) in the vicinity of Marguerite Bay during austral fall 2001 and 2002 .. 74 Table 4.2. Zooplankton abundance (ind m-2) in the vicinity of Marguerite Bay during austral fall 2001 and 2002 .. 75 Table 4.3. Mean depth of maximum abunda nce (Z, m) of copepods in 2001 and 2002 .. 90 Table 4.4. Mean depth of maximum abunda nce (m) of euphausiids, amphipods, and mysids during fall 2001 and 2002 .. 92 Table 4.5. Spearman rank order correlati ons between integrated abundance of zooplankton (ind m-2) and vertically integrated pigment concentrations (chloroph yll + phaeopigment; mg m-2), salinity at 10 m (S10), and bottom depth (bottom Z, m) 100 Table 4.6 Percentage of net hauls in which macrozooplankton were located primarily shallower, deeper, or at the same depth as the thermocline/pycnocline 101
v LIST OF FIGURES Figure 1.1. Map of the western Anta rctic Peninsula (WAP) region and geographic references 3 Figure 1.2. General circulation in Margue rite Bay and schematic paths of the Antarctic Circumpolar Current (ACC) (blue) and Antarctic Peninsula Coastal Current (APCC) (red) . 6 Figure 1.3. Bathymetric map of Marguerite Bay and adjacent waters of the western Antarctic Peninsula .. 7 Figure 2.1. Sampling stations overla id on SeaWiFS images of mean Ca for January (a) 1998, (b) 1999, (c ) 2000, (d) 2001 and (e) 2002 ... 18 Figure 2.2. Distribution of in situ depth-weighted (a) Ca Fluor and (b) Ca HPLC during January-February 1999 . 19 Figure 2.3. Comparison between Ca SWF (mg m-3, SeaDAS4.8, OC4v4 algorithm) and in situ Ca (mg m-3) .. 22 Figure 2.4. Comparison between Ca predicted by the OC4v4 algorithm (using SeaWiFS-derived Rrs as input) and measured in situ Ca (mg m-3) 23 Figure 2.5. Comparison between Ca HPLC and Ca Fluor (mg m-3) between January and February 1998 2001 .. 25 Figure 2.6. Normalized histogram of Ca SWF distributions (mg m-3) in the Southern Ocean during austral summer .26 Figure 2.7. Relationship between HPLC Cb/ Ca and Ca Fluor/ Ca HPLC and between HPLC Cc/ Ca and Ca Fluor/ Ca HPLC . 29
vi Figure 3.1. Map of the Antarctic contin ent showing the location of the study area and other geographic references ... 37 Figure 3.2. Mean abundance (ind m-3) of development stages of Euphausia superba in offshelf waters (Sta. 1) we st of the Antarctic Peninsula and coastal waters (Sta. 5) of Marguerite Bay during fall 2001 (light grey) and 2002 (dark grey) 41 Figure 3.3. Biweekly climatology ( 1997 2004) of SeaWiFS chlorophyll concentrations (mg m-3) between October and March . 44 Figure 3.4a. Location of the 14 subregions along the western Antarctic Peninsula superimposed over the climatology (1998 2004) of SeaWiFS chlorophyll concentrations (mg m-3) for January 1 14 .. 47 Figure 3.4b. Time series of geometric m ean chlorophyll concen trations in each subregion for each biweekly period during 1997 2004 . 48 Figure 3.5a. Biweekly SeaWiFS chlor ophyll concentrations (Chl, mg m-3) in October (Oct), November (Nov) and December (Dec) of 2000 .. 49 Figure 3.5b. Biweekly SeaWiFS chlor ophyll concentrations (Chl, mg m-3) in October (Oct), November (Nov) and December (Dec) of 2001 .. 50 Figure 3.6. Biweekly time series of ch lorophyll accumulation after the ice had receded at the September, October, and November 2000 and 2001 locations of the ice edge shown in figure 3.5 .. 52 Figure 3.7. Daily ice-free area (km2) in (a) northern and (b) southern Marguerite Bay during 1997 2004 . 53 Figure 3.8. Monthly mean SeaWiFS ch lorophyll concentrat ions in the Bellingshausen Sea during February (a) 2001 and (b) 2002 ... 63
vii Figure 4.1. (a) Location of the study area (red rectangle) and MOCNESS net hauls (circles) during fa ll of (b) 2001 and (c) 2002. .. 70 Figure 4.2a. Water column integr ated abundance of copepods (ind m-2) from net hauls in the vicinity of Marguerite Bay during austral fall 2001 (top) and 2002 (bottom) 76 Figure 4.2b. Percent composition of copepods at coastal stations in Marguerite Bay during austral fall 2001 (top) and 2002 (bottom) . 77 Figure 4.3a. Water column integrat ed abundance of euphausiids (ind m-2) from net hauls in the vicinity of Marguerite Bay during austral fall 2001 (top) and 2002 (bottom) 79 Figure 4.3b. Percent composition of euphausiid s from net hauls in the vicinity of Marguerite Bay during austral fall 2001 (top) and 2002 (bottom) .. 80 Figure 4.4. Length frequency of E. superba juveniles and adults during fall 2001 (top) and 2002 (bottom) .. 84 Figure 4.5 Length frequency of E. superba juveniles and adults during fall 2001 (top) and 2002 (bottom) . 85 Figure 4.6 Length frequency of E. crystallorophias juveniles and adults during fall 2001 (top) and 2002 (bottom) 86 Figure 4.7a. Water column integrated abundances of zooplankton other than copepods and euphausiids (ind m-2) from net hauls in the vicinity of Marguerite Bay dur ing austral fall 2001 (top) and 2002 (bottom) ... 87 Figure 4.7b. Percent composition of z ooplankton from net hauls in the vicinity of Marguerite Bay dur ing austral fall 2001 (top) and 2002 (bottom) .. 88
viii Figure 4.8. Mean depth of maximum abundance (m) of copepods during 2001 (black) and 2002 (grey) 91 Figure 4.9a. Vertical distribution of euphausiids, amphipods and mysids (ind m-3) in Crystal Sound during au stral fall 2001 (top) and 2002 (bottom) 93 Figure 4.9b. Vertical distri bution of euphausiids, amphipods and mysids (ind m-3) in Laubeuf Fjord during au stral fall 2001 (top) and 2002 (bottom) 94 Figure 4.9c. Vertical distribution of euphausiids, amphipods and mysids (ind m-3) in the vicinity of Alexande r Island during au stral fall 2001 (top) and south of Adelaide Island in fall 2002 (bottom) . 95 Figure 4.10a. Linear correlation betw een integrated abundances (ind m-2) of total macrozooplankton (euphaus ids, amphipods, and mysids) and copepods at different stations during fall 2001 and 2002 (n = 12; r = -0.39; p = 0.208) 96 Figure 4.10b. Mean water-column integrated abundance (ind m-2) of macrozooplankton (euphausiids, amphipods and mysids) (ind m-2) (grey bars, left axis) and copepods (black circles, right axis) at different stations within Margue rite Bay during fa ll 2001 (top) and 2002 (bottom) .. 97 Figure 4.11. (a) Location of the subregi ons analyzed for median SeaWiFS chlorophyll concentrations (mg m-3) in (b, c) oceanic and (d) coastal waters of the Bellings hausen Sea, (e) northern, and (f) southern Marguerite Bay in spring/summer 2000/2001 (grey) and 2001/2002 (black) 103 Figure 4.12. (b) Spearman R correlation be tween geometric m ean chlorophyll concentrations in oceanic waters of the Bellingshausen Sea during November 1997 2004 and recruitment of E. superba (R1) in waters west of the Antarc tic Peninsula; n = 6; R = 0.81; p < 0.05) .. 105
ix Characterization of the West ern Antarctic Peninsula Ecosystem: Environmental Controls on the Zooplankton Community Marina Marrari ABSTRACT The zooplankton community of Marguerite Bay, western Antarctic Peninsula, was investigated in relation to va riability in chlorophyll concen trations and sea ice dynamics, using a combination of satellite remote sensing techniques and plankton net data. SeaWiFS chlorophyll data were validated with concurrent in situ data measured by HPLC and fluoromentric methods, and results indicat e that SeaWiFS chlor ophyll is an accurate measure of in situ values when HPLC data are used as ground truth. Climatology data of SeaWiFS chlorophyll we st of the Antarctic Peninsula showed that the Bellingshausen Sea and Marguerite Bay usually had higher and more persistent chlorophyll concentrations co mpared with northern regi ons. These predictable phytoplankton blooms could provide the Antarctic krill, Euphausia superba with the food required for successful reproduction and la rval survival. U nusually high krill reproduction in 2000/2001 was coincident with above-average chlor ophyll concentrations throughout the study area and was followed by the largest juvenile recruitment since 1981. High larval densities at the shelf break along the Antarctic Peninsula may have
x resulted, in part, from krill spawning in the Bellingshausen Sea. Interannual differences in sea ice also probably contributed to the variability in larval krill abundances. Interannual differences were observe d in the species composition of the zooplankton of Marguerite Bay during fall, a nd these were linked to variability in the environmental conditions. Thysanoessa macrura was the most abundant euphausiid in 2001, while Euphausia crystallorophias dominated in 2002, and E. superba had intermediate densities duri ng both years. Copepods were more abundant in 2001 by a factor of 2.6. Copepods and T. macrura showed a rapid population response to unusually high chlorophyll concentrations in the Bellingshausen Sea and Marguerite Bay during spring-summer 2000/2001, whereas E. superba and E. crystallorophias had a longer term response and showed increased recruitment in fall 2002. There were no clear associations between the distribution of zooplankton and environmental conditions in fall; however there was a significant relations hip between chlorophyll concentrations in the Bellingshausen Sea during the preceding sp ring and zooplankton patterns during fall.
1 CHAPTER ONE THE WESTERN ANTARCTIC PENINSUL A: PHYSICAL AND BIOLOGICAL SETTING The Southern Ocean covers approximately 10% of the worlds ocean, supporting large concentrations of zooplankt on and higher trophic level organisms. The region plays an important role in the world s ocean biogeochemical cycle (Sarmiento et al., 1998) and has a profound influence on globa l circulation by connecting the Atlantic, Pacific, and Indian Ocean basins, and through deep water formation. Based on biogeography, three main zones have been de scribed for the Southe rn Ocean: (1) an icefree zone rich in nutrients, but relatively poor in primary produc tion, with a zooplankton community dominated by copepods, salps a nd small euphausiids, (2) a productive seasonal pack-ice zone that includes the area covered by sea ice during winter, but mostly ice-free during summer and fall, where zoop lankton may be dominated by the Antarctic krill, Euphausia superba and (3) a permanent pack i ce zone, with generally low zooplankton biomass, where E. superba is often replaced by the neritic euphausiid, Euphausia crystallorophias (Hempel, 1985). Within the Atlantic sector of the Sout hern Ocean, the contin ental shelf along the western Antarctic Peninsula (WAP) includes wa ters of the seasonal and permanent pack ice zones (Fig. 1.1). This region supports high concentrations of zooplankton and
2 predators, and is considered one of the most productive areas in Anta rctic waters (Deibel and Daly, 2007). The Antarctic Peninsula region also is of interest as it is warming more rapidly than almost any other place on the pl anet. A considerable increase in atmospheric and upper-ocean temperatures has occurred in the last 50 years, while ice shelves have retreated (King, 1994; Smith et al., 1996b; Vaughan and Doake, 1996; Vaughan et al., 2003). In addition, there have been marked ch anges in winter sea ice extent and duration (Parkinson 2002). These changes have importa nt implications for Antarctic organisms that rely directly on sea ice for reproduction, such as Ad lie penguins (Trivelpiece and Fraser, 1996), as well as for others that wi ll suffer indirect effects through changes in prey populations (Costa and Crocker, 1996). The Antarctic krill, E. superba is a relatively large pe lagic crustacean (up to 65 mm in length) and a keystone species in th e Antarctic ecosystem, acting as a primary grazer on phytoplankton and prey for a variety of higher trophic leve l predators. Most Antarctic predators, including species of fish, seals, whales, penguins, and many seabirds, rely on zooplankton for survival, and in par ticular Antarctic kri ll (Lowry et al., 1988; Ainley and DeMaster, 1990; Costa and Crocker, 1996; Murase et al., 2002). E. superba has a circumpolar distribution, w ith highest concentrations in the Atlantic sector of the Southern Ocean (Marr, 1962). It is believed that krill originated in the WAP region are the source of large populations observed dow nstream in the Scotia Sea and at South Georgia Island (Fach and Klinck, 2006; Thorpe et al., 2007). Since the Discovery expeditions in the 1920-30s, which surveyed almost the entire Southern Ocean and resulted in a large amount of information regarding the distribution, biology and ecology of Antarctic krill (Marr, 1962) research in the vicinity
3 of the WAP has been mostly restricted to nor thern waters along the continental shelf and downstream in the Scotia Sea. Figure 1.1. Map of the western Antarctic Peninsula (WAP) region and geographic references. The dotted line indicates the 1000 m isobath. Some interdisciplinary programs that i nvestigated the structure and dynamics of the marine ecosystem from the midto nor thern sectors of the WAP include BIOMASS Program (Biological Investigations of Mari ne Antarctic Systems and Stocks) in 1985/86
4 (El-Sayed, 1994), RACER (Research on Antarctic Coastal Ecosystem Rates) (Huntley et al., 1991), Palmer LTER (Long Term Ecological Research), which si nce the early 1990s has been dedicated to studying how physical fo rcing affects the structure and function of the ecosystem (Ross et al., 1996), and AMLR (Antarctic Marine Living Resources Program), established in 1990 to assess reso urces in the area around Elephant Island (e.g., Loeb et al., 1997; Siegel et al., 2002; Watkin s et al., 2004). The information resulting from these efforts, as well as from othe r field expeditions from the United Kingdom, Germany, Argentina, Poland, Russia, and Ch ile, among others, has greatly improved our understanding of the processes controlling populations in the ar ea, and has provided insight into the dominant physical components, such as circulation, hydrography, nutrient distributions, and sea ice dynamics. Howeve r, in spite of the significant amount of information available for the northern sectors, there have been few studies in southern regions of the WAP (Marguerite Bay and th e eastern Bellingshaus en Sea) (Atkinson, 1995; Siegel and Harm, 1996; Lascara et al., 1999; Me yer et al., 2003). The Southern Ocean Global Ocean Ecosystem Dynamics Program (SO GLOBEC) field efforts were focused in th e vicinity of Marguerite Bay, as limited information suggested that it was an importa nt overwintering site for upper trophic level predators and, therefore, probably of their f ood source, Antarctic krill (Hoffman et al., 2004). The primary goal of the U.S. SOGL OBEC Program was to investigate the physical and biological factors that influence the growth, recruitment, and overwintering survival of E. superba in the vicinity of Marguerite Bay as well as the associated predators and prey of kri ll (Hoffman et al., 2004). Initial results of the US SOGLOBEC Program have described the general
5 circulation in Marguerite Bay. There are tw o major currents: the wind-driven westerly Antarctic Circumpolar Current (ACC) that flows northward along the continental shelf break, and the buoyancy-driven Antarctic Pe ninsula Coastal Current (APCC), which flows southward along the coast (Moffat et al., 2008) (Fig. 1.2). The APCC enters Marguerite Bay at the south end of Adelaide Island, flows clockwise along the coast and exits around the north end of Alexander Island. Mesoscale circ ulation features, such as gyres on the outer shelf and eddies within inner Marguerite Bay, contribute to making Marguerite Bay a favorable re tention area for phytoand z ooplankton (Beardsley et al., 2004; Klinck et al., 2004; Dorland and Zhou, 2008) Marguerite Bay encompasses a relatively wide and deep continental shelf, with mean depths of ~ 400 m (Bolmer et al., 2004). The bathymetry is influenced by Marguerite Trough, a deep canyon up to 1600 m deep, which intersec ts the continental shelf off Marguerite Bay, crosses the shelf, and extends into George VI Sound (Fig. 1.3). In addition, the area is charac terized by numerous seamounts and depressions that result in a variable and complicated topography, partic ularly in the norther n sectors, such as Laubeuf Fjord and Crystal Sound north of Marguerite Bay. Intrusions of Upper Circumpolar Deep Water (UCDW), found at dept hs greater than 500 m in oceanic areas, have been observed on the continental shel f, supplying Marguerite Bay with warm and nutrient rich waters at depth. These intr usions occur through Ma rguerite Trough, as well as in other shelf break areas such as those off Crystal Sound and west of Alexander Island (Fig. 1.2) (Dinniman and Klinck, 2004). The hydrographic structure of c ontinental shelf waters along the Antarctic Peninsula has been previously described (Smith et al., 1999; Klinck et al., 2004). The primary water
6 masses include Antarctic Surface Water (ASSW ), characterized by temperatures and Figure 1.2. General circulation in Marguerite Bay, and schematic paths of the Antarctic Circumpolar Current (ACC) (blue) and Anta rctic Peninsula Coastal Current (APCC) (red). The dotted blue line represents the a pproximate location of intrusions of Upper Circumpolar Deep Water (UCDW) onto the continental shelf (Dinniman and Klinck, 2004). Sections of broken red line represent suggested paths of the APCC, not derived from observations (adapted from Moffat et al., 2008). The approximate location of a mesoscale clockwise gyre is indicated by th e broken black line (Klinck et al., 2004) salinity ranges between -1.8 and 1 C and 33.0 and 33.7, respectively. Beneath a pycnocline, generally established at ~ 100 120 m, is warm (1.5 C) and salty (34.6 34.73) water derived from Circumpolar Deep Wa ter (CDW), which is typically present in waters over the outer shelf of the WAP at de pths of 200 600 m. During late fall, surface
h ( < r e m F A a n a l eat loss an d < -1 C) Wi n e mnant laye r m odified CD W igure 1.3. B A ntarctic Pe n A m n d retreat o f l lowing den s wind forcin n ter Water ( W r of cold W W W layers. athymetric m n insula. Bat h m ajor physic a f sea ice. D u s e phytopla n g result in a W W). This W is normal l m ap of Mar g h ymetry in m a l character i u ring summ e n kton bloo m 7 deep winte r surface lay e l y observed g uerite Bay a m eters (m). R i stic of the S e r most of t h m s to develo p Al e Ma r T r r mixed laye e r is partiall y at ~ 80 10 0 a nd adjacen t R eproduced f S outhern Oc e h e WAP con t p During w i e xander Is. rg uerite r ou g h Cr y So u Adelaid e Is. r occupied b y restratifie d 0 m betwee n t waters of t h f rom Bolme e an is the an n t inental shel i nter, sea ic e Antarctic Peninsula y stal u nd e b y nea r -free z d in spring a n n the AAS W h e western r et al. (200 4 n ual advanc f is ice-free, e cover can z ing n d a W and 4 ). e
8 completely cover the continental shelf. In recen t years, however, a dram atic decline in sea ice extent has been observed in the WAP, resulting in areas of the northern WAP which may remain ice-free throughout the year. Ho wever, in southern sectors, such as Marguerite Bay and the eastern Bellingshausen Sea, winter se a ice generally extends over most of the continental shelf and, thus, can be considered an importa nt factor influencing marine organisms. Extensive sea ice reduces the penetration of ra diant heat into the water column during spring, summer, and fall, limiting phytoplankton growth and, thus, food available to zooplankton. To summarize, the Marguerite Bay region is characterized by a complex seafloor topography, extreme variability in seasonal sea ice conditions, and circulation features, which include opposite flowing currents, intrusions of oceanic waters onto the shelf, and mesoscale gyres and eddies. Gi ven the rapidly changing conditions of this region and the potential effects on Anta rctic organisms, understanding zooplankton population dynamics and their response to en vironmental variability is critical to predicting the effects of climate change on the Antarctic ecosystem as a whole. This dissertation aims to investigate th e effects of environm ental variability on the population dynamics of phytoplankton and zoopl ankton in the vicinity of Marguerite Bay, with special emphasis on E. superba For this purpose, I use a combination of ocean color remote sensing techniques, zooplankton net data, and environmental information. The main objective of Chapter 2 is to evalua te the performance of the SeaWiFS satellite sensor (Sea-Viewing Wide Field-of-Vie w Sensor) in estimating chlorophyll concentrations in the Southern Ocean. Ch apter 3 focuses on using SeaWiFS data to describe the temporal and spatial dynamics of chlorophyll distributions along the WAP
9 during spring and summer, and relate the inte rannual variability observed in chlorophyll dynamics to changes in the reprodu ction and recruitment success of E. superba Chapter 4 investigates the patterns of abundance and distribution of the dominant zooplankton groups in Marguerite Bay, in relation to va riability in environmental conditions, and a summary of the major findings and future direction is presented in Chapter 5.
10 CHAPTER TWO VALIDATION OF SEAWIFS CHLOROPH YLL A CONCENTRATIONS IN THE SOUTHERN OCEAN: A REVISIT INTRODUCTION Since the launch of the Sea-viewing Wide Field-of-view Sensor (SeaWiFS, McClain et al., 1998) onboard the Orbview-II satellite in August 1997, ocean color data products, in particular con centrations of chlorophyll a ( Ca, mg m-3) in the surface ocean, have been used to investigate a wide vari ety of fundamental topics including ocean primary productivity, biogeochemistry, coastal upwelling, eutrophication, and harmful algal blooms (e.g., Hu et al., 2005; Muller-Kar ger et al., 2004). Other ocean color missions, such as the ongoing MODerate-r esolution Imaging Spectroradiometer (MODIS, Esaias et al., 1998; Terra satell ite for morning pass since 1999 and Aqua satellite for afternoon pass sin ce 2002) or the future Nationa l Polar-Orbiting Operational Environmental Satellite System (NPOESS), a ssure the continuity of remotely sensed ocean color in assessing the long-term gl obal change in several key environmental parameters, including Ca. Quantitative use of ocean colo r data products requires a high level of accuracy. During algorithm development, the errors in the Ca data products after logarithmic transformation were about 0.2 or less (OReilly et al., 2000), which corresponds to roughly 50% root mean square (R MS) relative error. Global validation efforts show that in most ocean basins Ca errors are about 0.3 (Gregg and Casey, 2004)
11 although in regions such as the Southern O cean, reported errors are significantly larger. The Southern Ocean was defined by the International Hydrographic Organization in 2000 to encompass waters betw een the northern coas t of Antarctica and 60 S. Oceanographers, however, traditionally have defined the northern limit of the SO as the Subtropical Front (at approximately 40 S) (Orsi et al., 1995). Typical chlorophyll concentrations in the Southern Ocean range between 0.05 and 1.5 mg m-3 (Arrigo et al., 1998; El-Sayed, 2005). It is beli eved that the interaction of light and deep mixing, iron, and grazing limit phytoplankton growth through out the Southern Ocean, in addition to low silicate concentrations which can limit diatom productio n north of the Polar Front (Moline and Przelin, 1996; Daly et al., 2001; Boyd, 2002). However, elevated chlorophyll concentrations (1 to > 30 mg m-3) are characteristic of many regions, including continental shelf and ice edge areas (Holm-Ha nsen et al., 1989; Moore and Abbott, 2000; El-Sayed, 2005), and even values of up to 190 mg m-3 have been reported (El-Sayed, 1971). The Antarctic Peninsul a region, in particular, supports large concentrations of phytoplankton, zooplankt on, seabirds, seals and whales, and is considered one of the most productive areas of the Southern Ocean, for reasons that are not fully understood (D eibel and Daly, 2007). Several studies have relied on ocean co lor data to investigate phytoplankton spatial patterns (Moore a nd Abbott, 2000; Holm-Hansen et al., 2004a), interannual variability during summer (Smith et al ., 1998a; Korb et al., 2004) and primary productivity (Dierssen et al., 2000; Smith et al., 2001) west of the Antarctic Peninsula and in the adjoining Scotia Sea. These studies used in situ Ca determined from water samples using fluorometric methods ( Ca Fluor) to validate monthly/weekly averages of
12 SeaWiFS Ca ( Ca SWF) data product at ~ 9 x 9 km2 or ~ 4 x 4 km2 resolution and concluded that in the Southern Ocean, Ca SWF values are significantly lower than those estimated from in situ water samples. For example, Dierssen and Smith (2000) applied in situ biooptical data measured between 1991 and 1998 to the OC2v2 algorithm to test its applicability west of the Anta rctic Peninsula in the Southern Ocean. They concluded that Ca derived from the OC2v2 algorithm using in situ reflectance was 60% lower than in situ Ca ( Ca between 0.7 and 43 mg m-3, median ~ 1 mg m-3). Korb et al. (2004) reported that Ca SWF values were only 87% of Ca Fluor for concentrations lower than 1 mg m-3 and only 30% for concentrations above 5 mg m-3 in the South Georgia area (54.5 S, 37 W). In addition, Moore et al. (1999) found a strong linear relationship between Ca SWF and Ca Fluor (R2 = 0.72, n = 84) in the Ross Sea, although they noted that SeaWiFS tended to underestimate Ca values between 0.1 and 1.5 mg m-3. The previous validation methods may present several limitations. First, in situ samples are point measurements while satellit e pixels cover a larger area (up to 9 x 9 km2). Patchiness within a pixel will affect the comparison of results between areas and over time (e.g., Hu et al., 2004). Second, the in situ and satellite measurements are not strictly concurrent and the tim e differences can be large ( up to a month). Finally, and most importantly, previous validation studies used in situ Ca from fluorometric measurements, while it is now widely recognized that High Performance Liquid Chromatography (HPLC) may yield more accurate results in determining Ca from water samples. Fluorometric methods may result in biased results, particularly in the presence of certain accessory pigments (L orenzen, 1981; Welschmeyer, 1994). In a study that included three different ar eas of the worlds oceans, Trees et al.
13 (1985) reported that errors in the Ca Fluor ranged between -68 and 53% with a mean of 39%. In addition, Bianchi et al. (1995) found that Ca Fluor in the northern Gulf of Mexico was approximately 30% lower than Ca HPLC, except in near coastal areas. It is believed that the presence of significan t amounts of chlorophyll b ( Cb), characteristic of chlorophytes, prochlorophytes and prasinophy tes, causes fluorometric techniques to underestimate Ca. On the other hand, high conc entrations of chlorophyll c ( Cc), typically found in diatoms, dinoflagellates, cryp tophytes, and haptophytes, lead to an overestimation of Ca with respect to fluorometric measurements. The fluorescence emission spectra of degradation products (phaeopigments) of Ca and Cb overlap considerably, causing an overestimation of Ca phaeopigments and, thus, an underestimation of Ca. On the other hand, Ca and Cc have partially overlapping fluorescence spectra, causing an overestimation of Ca and subsequent underestimation of phaeopigments a (Gibbs, 1979; Jeffrey et al., 1997). The filters used in the standard fluorometric method (Lorenzen, 1981) cannot effectively disc riminate between Ca, Cb, Cc and their degradation products; thus, dependi ng on the type of phyt oplankton present and their associated pigments, Ca may be overestimated or underestimated by fluorometric methods. Herein, I use concurrent HPLC and fluor ometric data collected between 1998 and 2002 in waters west of the Antarctic Peninsul a, as well as high-re solution SeaWiFS data, to re-examine whether SeaWiFS Ca is underestimated in the Southern Ocean as reported in previous studies. I also discuss possibl e explanations for the observed results and investigate the effects of different accessory pigments on Ca estimations.
14 METHODS SeaWiFS daily Level 2 data between December 1997 and December 2004 were obtained from NASA Goddard Space Flight Ce nter. These data were derived from the high-resolution (~ 1 km/pixel near nadir) Le vel 1 data collected by ground stations, as well as occasional satellite onboard recordi ng over the area using the most current algorithms and software package (SeaDAS4.8). A total of 6606 data files were obtained and mapped to a rectangular proj ection with approximately 1 km2/pixel for the area between 45 75 S and 50 80 W west of the Antarctic Pe ninsula (Fig. 1.1). The data product used in this study is the surface Ca estimated with the OC4v4 empirical algorithm (OReilly, 2000): 4 3 2532 1 649 0 93 1 067 3 366 010R R R R aC (1) where R = log10[(max(Rrs443, Rrs490, Rrs510))/Rrs555)] and Rrs is the remote sensing reflectance, a data product af ter atmospheric correction. Chlorophyll fluorescence and HPLC pigmen t data were collected and analyzed by Drs Raymond Smith (Univers ity of California Santa Ba rbara) and Maria Vernet (University of California San Diego) as part of the Palmer Long Term Ecological Research (LTER) program during cruises west of the Antarctic Peninsula (see http://pal.lternet.edu/data/ for detailed methods). The lo cation of the LTER chlorophyll sampling stations between 1998 and 2002 are sh own in figure 2.1. Most of the samples were collected within the 2000 m isobath, al though two transects we re conducted across
15 Drake Passage in January-February 1999 and 2000 to measure Ca Fluor. At each station, water column samples were collected at disc rete depths for both fluorometric and HPLC measurements. Ca, Cb and Cc were obtained by HPLC from samples collected at fixed stations during January-February 1998 and 1999 following the methods of Wright et al. (1991), and during January-Fe bruary 2000 and 2001 following the methods of Zapata et al. (2000). Ca and phaeopigment concentrations also were obtained by fluorometric methods by measuring total fluorescence and subtracting phaeopigments after acidification from samples collected during January February 1998, 1999, 2000, 2001 and 2002 following Smith et al. (1981, 1996a, 1998). Welschme yers (1994) method, which effectively measures fluorescence from Ca only and reduces interference from Cb or its phaeoderivatives, was not applied (M. Vernet, pers. comm .). Because the signal detected by the satell ite sensor is an optically-weighted function of signals at all dept hs (up to 50 60 m for clear waters), the method of Gordon (1992) was used to calculate a depth-wei ghted chlorophyll concentration,
16 and 775 fluorometric Ca values were used in these an alyses. Because the weighting function, g(z) decreases exponentially with increasi ng depth,
17 2 1 11 ()100 1 ()100n i i n i iRMSx n biasxx n SI x I (3) where S is satellite data, I is in situ data, and n is the number of data pairs. For a normally distributed x RMS should equal the standard de viation. Further, because the natural distribution of Ca is lognormal (Campbell, 1995), error estimates were also made on the logarithmically transformed (base 10) data: 2[(log()log()] log_ [log()log()] log_ SI RMS n SI bias n (4) These error estimates have been used in rece nt publications to de scribe the performance of the ocean color algorithms (OReilly et al., 2000) and to validate SeaWiFS global and regional estimates of Ca (Darecki and Stramski, 2004; Gregg and Casey, 2004; Zhang et al., 2006). Note that these la tter error estimates cannot be expressed as percentages because they are logarithmically transformed (Campbell, submitted).
F 1 p igure 2.1. S a 998, (b) 19 9 ink triangle s a mpling sta t 9 9, (c) 2000, s : HPLC sa m t ions overlai d (d) 2001 a n m ples, white 18 d on SeaWi F n d (e) 2002. W line: 2000 m F S images o W hite circl e m isobath. f mean Ca f o e s: fluorome t o r January ( a t ric samples a )
R J a w m w d o b F J a w c o o C R ESULTS Typi c a nuary-Febr u w ith a media n m g m-3 with a w ere consist e etected thro u b served in M igure 2.2. D a nuary-Febr u A tot a w ere obtaine d o mparisons. f 0.55 0.6 3 C a Fluor/ Ca SWF c al Ca Fluor an d u ary 1999 ( F n of 0.86 m g a median of e ntly found o u ghout the c M arguerite B D istribution u ary 1999. W a l of 96 Ca S W d using the m In general, 3 between t h is 2.73 2. d Ca HPLC dis t F ig. 2.2). I n g m -3. Ca HP L 1.04 mg m 3 o ffshelf in D c ontinental s h B ay. of in situ W hite line: 2 W FCa HPLC m a m ethod desc r Ca SWF is si g h e two (Tab l 19, consiste n 19 t ributions d u n all years, C L C was typic a 3 In genera l D rake Passa g h elf, with t h depth-wei g 2 000 m isob a a tching pair s r ibed above g nificantly l o l e 2.1). The n t with pre v u ring austra l C a Fluor range d a lly lower a n l the lowest g e. Elevated h e highest v a g hted (a) C a a th. s and 307 C a Table 2.1 l o wer than C a inverse rati o v ious observ a l summer ar e d from 0.052 n d ranged fr o Ca values ( < Ca values ( > a lues (>10 m a Fluor and ( b a SWFCa Fluor m l ists the stat i a Fluor (Fig. 2 o i.e., the r a a tions in the e presented f to 27.6 mg o m 0.017 to < 0.1 mg m-3 > 1 mg m-3) w m g m -3) alwa y b ) Ca HPLC d m atching p a i stics of the s 3), with a r a a tio of Southern O f or m-3, 14.6 ) w ere y s d uring a irs s e a tio cean
20 where Ca Fluor was used to validate Ca SWF and the same pattern of underestimation was observed (Moore et al., 1999; Di erssen and Smith, 2000; Korb et al., 2004). In contrast, Ca HPLC showed a more satisfactory agreement with Ca SWF over a wide dynamic range (0.1 4 mg m-3) (Fig. 2.3). The mean ratio of Ca SWF/ Ca HPLC is close to 1 (i.e. 1.12), in contrast to the lower ratio of 0.55 for Ca SWF/ Ca Fluor. Table 2.1. Statistics for th e comparisons between Ca SWF and in situ Ca ( Ca Fluor, Ca HPLC). n is the number of matching pairs, RMS is root mean square error, SD is standard deviation Parameter Ca SWF vs. Ca Fluor Ca SWF vs. Ca HPLC n 307 96 Ratio SD 0.55 0.63 1.12 0.91 RMS 77.2% 91.4% Bias -45.2% 12% log_RMS 0.44 0.34 logbias -0.36 -0.07 Although the RMS errors for the two comp arisons are comparable (Table 2.1), Ca HPLC is nearly equally scattered around the 1:1 line (Fig. 2.3), suggesting that the bias errors in Ca SWF/ Ca HPLC are significantly smaller than those in Ca SWF/ Ca Fluor. Clearly, the agreement between Ca SWF and Ca HPLC is much improved over that between Ca SWF and Ca Fluor. Similar results were also obtained from th e algorithm perspective. By using the spectral remote sensing reflectance data ( Rrs) derived from satellite measurements (Fig. 2.4), the OC4v4 algorithm yielded comparable results to those obtained from HPLC measurements. In contrast, Ca Fluor values are significantly high er than those predicted by the OC4v4 algorithm for the entire range considered.
21 Are these results representative of the entire Southern Ocean? Due to cloud cover, satellite data were not available for a ll pixels every day. This reduced the number of Ca SWF data points, which resulted in a limite d number of matching pairs for comparing satellite and in situ data (307 for fluorometric and 96 for HPLC). However, the in situ data itself comprised a much larger datase t that included 832 concurrent fluorometric and HPLC measurements. When this in situ dataset was used to compare Ca Fluor and Ca HPLC, similar results were obtaine d, i.e., the mean ratio of Ca Fluor/ Ca HPLC is 2.43 3.37 (Fig. 2.5). The ratio of Ca Fluor/ Ca HPLC appears to decrease with increasing concentrations (Table 2.2), although for Ca HPLC <0.05 mg m-3 and Ca HPLC >3.0 mg m-3 the statistical results may not be reliable because of the fe w matching pairs availabl e and the scatter of the data (Fig. 2.5). For Ca HPLC between 1.5 and 3.0 mg m-3, the bias is small (15%) and the mean ratio of Ca Fluor/ Ca HPLC is close to unity (1.15 0.73). Between 0.05 and 1.5 mg m-3, however, Ca Fluor is much higher than CaHPLC (mean Ca Fluor/ Ca HPLC = 2.48 2.23, n = 647). This difference is believed to be due to errors in the CaFluor measurements as described above. Because most (> 90%) of the waters in the Southern Ocean have surface Ca SWF values between 0.05 and 1.5 mg m-3 (Fig. 2.6), this assessment can be generalized and applied to most regions.
22 C aSWF 0.010.1110100 in situ C a 0.01 0.1 1 10 100 y = 2.56x1.11, n = 307, R2 = 0.68 y = 1.11x0.93, n = 96, R2 = 0.45 Figure 2.3. Comparison between Ca SWF (mg m-3, SeaDAS4.8, OC4v4 algorithm) and in situ Ca (mg m-3). Grey circles and line: Ca Fluor, blue diamonds and black solid line: Ca HPLC. The dashed line shows the 1:1 relationship. The statistics of the comparisons are listed in Table 1.
23 Max(Rrs443, Rrs490,Rrs510)/Rrs555 0.1110 C a 0.01 0.1 1 10 100 C aFluor n = 307, y = 4.83x-2.1, R2 = 0.71 C aHPLC n = 96, y = 2.05x-1.89, R2 = 0.45 CaSWF Figure 2.4. Comparison between Ca predicted by the OC4v4 algorithm (using SeaWiFSderived Rrs as input) and measured in situ Ca (mg m-3). Black broken line: OC4v4 prediction ( Ca SWF); grey circles and solid line: Ca Fluor, blue diamonds and thick line: Ca HPLC.
24 Table 2.2. Statistics for th e comparisons between Ca Fluor and Ca HPLC (mg m-3) for data shown in Figure 2.5. a0 and a1 are the power fitting coefficients in the form of Ca Fluor = a0 ( Ca HPLC)a1, R2 is the corresponding coefficient of determination, n is the number of matching pairs, RMS is root mean square error, SD is standard deviation. Ca HPLC range 0.01 15 < 0.05 0.05 1.5 1.5 3.0 > 3.0 n 832 21 647 96 68 a0, a1 1.40, 0.66 0.28, 0.14 1.34, 0.63 1.01, 0.96 2.15, 0.55 R2 0.67 0.01 0.49 0.11 0.14 Ca Fluo r / Ca HPLC SD 2.43 3.37 10.06 15.21 2.48 2.23 1.15 0.73 1.37 1.04 RMS 366% 1739% 268% 74% 110% bias 143% 905% 148% 15% 37% log_RMS 0.40 0.87 0.40 0.23 0.34 logbias 0.25 0.79 0.29 -0.01 0.02
25 C aHPLC 0.0010.010.1110100 C aFluor 0.001 0.01 0.1 1 10 100 C aFluor = 1.34( C aHPLC)0.63, R2 = 0.49, n = 647 C aFluor = 1.01( C aHPLC)0.96, R2 = 0.11, n = 96 Figure 2.5. Comparison between Ca HPLC and Ca Fluor (mg m-3) between January and February 1998 2001 (n = 832). Grey squares: Ca < 0.05 mg m-3, cyan circles: Ca between 0.05 1.5 mg m-3, green triangles: Ca between 1.5 3 mg m-3, blue diamonds: Ca > 3 mg m-3. The dashed line shows the 1:1 relati onship. Statistics for the comparison are listed in Table 2.2.
26 Figure 2.6. Normalized histogram of Ca SWF distributions (mg m-3) in the Southern Ocean during austral summer. (a) For the study region (Fig. 1.1) bound by 75 60o S and 75 60o W; (b) for the entire S outhern Ocean (south of 60oS). The y-axis shows the percentage surface area. 91% and 96% of the su rface waters for (a) and (b), respectively, fall within the range of 0.05 to 1.5 mg m-3. DISCUSSION Although HPLC has been recommended as the most reliable method to determine Ca (e.g. Trees et al., 1985), most cruise survey s still use the fluorometric method because it is faster, requires less technical expertis e and is less expensive than HPLC. The Ca data originally used in the development of the OC4v4 algorithm (OReilly et al., 2000) included 2,853 in situ measurements from a variety of o ceanic environments (but not the Southern Ocean), of which 72% were fluorom etric and 28% were HPLC measurements. Therefore, the predicted Ca satellite measurements should naturally lean toward the fluorometric values. However, this was not observed in the present study, suggesting that the species composition and their associated pi gment absorption characteristics in waters west of the Antarctic Peninsula region ma y be different from the mean composition and absorption on which the or iginal algorithm was based. (a) (b)Jan Feb, 1998-2004 75oS to 60oS 75oW to 60oW Jan Feb, 1998-2002 90oS to 60oS 180oW to 180oE 0 10 20 30Percen t 0.010.1110Ca SWF (mg m-3) 0.010.1110Ca SWF (mg m-3)
27 The large difference observed between Ca Fluor and Ca HPLC from the same water samples was likely due, in part, to interferen ce of the fluorescence signal by chlorophyll accessory pigments ( Cb, Cc and their degradation products). In this study, Cb only occurred in low concentrations compared to Ca (mean ratio Cb/Ca = 0.023, n = 486); however, Cc was relatively high (mean ratio Cc/Ca = 0.25, n = 486) (Fig. 2.7). The presence of significant amounts of Cc is known to cause an overestimation of Ca by the fluorometric method (Gibbs, 1979; Lorenzen, 1981). Cb is an accessory pigment in prochl orophytes, chlorophytes and prasinophytes, while Cc is generally present in diatoms, di noflagellates, crypt ophytes and haptophytes (Parsons et al., 1984). Diatoms are the dom inant phytoplankton in waters west of the Antarctic Peninsula, with dinoflagellates bei ng very abundant at times (Przelin et al, 2000, 2004). Prochlorophytes, a type of cyanob acteria first identifie d in the late 1980s (Chisholm et al., 1988), have not yet been observed in the Southern Ocean, while chlorophytes can be abundant (Przelin et al, 2000, 2004). Similarly, cryptophytes are usually scarce in the water column, but can be very abundant in coastal surface melt water during spring and summer (Moline and Przelin, 1996). Alloxanthin, the biomarker pigment for cryptophytes (Przelin et al, 2000), occurred in 91% (n = 516) of the pigment samples. Hence, chlorophytes were probably the dominant source of Cb during our study period, while the dominant sources of Cc appear to be diatoms, dinoflagellates and cryptophytes identified by the presence of fucoxanthin, peridinin, and alloxanthin in 99.5%, 53% and 91% of the samples, respectively. Cb and Cc vary widely throughout the world s ocean (Jeffrey, 1976; Lorenzen, 1981; Trees et al., 1985; Bidigare et al. 1986; Goericke and Repeta, 1993; Bianchi et al.,
28 1995). Overall, these studies found that Cb can cause an underestimation of Ca by the fluorometric method with ratios of Cb/Ca ranging from 0.15 to 0.51, while the presence of significant amount of Cc can lead to an overestimation of Ca. Typical ratios of Cc/ Ca for assemblages dominated by phytopl ankton containing chlorophyll c range from 0.15 to 0.44 (Bidigare et al., 1986; Bianchi et al., 1995; Lohrenz et a., 2003). Results reported here are consistent with these previous findings. Can the presence of significant amount of Cc lead to overestimation of Ca when the latter is derived from remote sensing re flectance data? The inversion of remote sensing reflectance to Ca is an implicit (e.g., OC4v4) or explicit (e.g., Maritorena et al., 2002) function of phytoplankton pigment absorpti on. Lohrenz et al. (2003) reported that even if the amount of accessory pi gments (sum of carotenoids and Cb + Cc) is equal to Ca, the perturbation to the pigment absorption is < 30%, suggesting a rela tively small error in the satellite-retrieved Ca. Hence, the large differences between Ca SWF and Ca Fluor observed here cannot be explained by the a dditional absorption of accessory pigment, but can be explained by the interf erence of these accessory pigm ents to the fluorescence peak when Ca is determined using the fluorometric method.
29 HPLC C b /C a and C c /C a 0.00010.0010.010.11 C aFluor/ C aHPLC 0.1 1 10 100 HPLC C b /C a HPLC C c /C a Figure 2.7. Relationship between HPLC Cb/ Ca and Ca Fluor/ Ca HPLC (y = 4.36 x0.26, R2 = 0.11, n = 482), and between HPLC Cc/ Ca and Ca Fluor/ Ca HPLC (y = 3.09 x0.39, R2 = 0.19, n = 482). Note that the slope for the latter (0.39) is significantly larger than for the former (0.26). Here Cb/Ca = 0.023 0.034 (n = 482) and Cc/ Ca = 0.25 0.59 (n = 482). In summary, contrary to previ ous reports that estimates of Ca SWF in the Southern Ocean were significantly lower than those measured in situ satellite estimates reported here agree with those determined from water samples for Ca between 0.05 and 1.5 mg m-3 for January-February between 1998 and 2001. This is primarily because the in situ Ca data were determined by HPLC ( Ca HPLC) rather than by fluorometric methods ( Ca Fluor), which are known to introdu ce significant errors in Ca estimates in the presence of certain accessory pigments.
30 CHAPTER THREE SPATIAL AND TEMPORAL VARIABILIT Y OF SEAWIFS CHLOROPHYLL A DISTRIBUTIONS WEST OF THE ANTA RCTIC PENINSULA: IMPLICATIONS FOR KRILL PRODUCTION INTRODUCTION Chlorophyll distributions in the Southern Ocean show high spatial and temporal variability. Most Antarctic open waters ha ve relatively low chlorophyll concentrations despite the availability of nutrients and, thus the Southern Ocean is generally considered to be a High-Nutrient Low-Chlorophyll region (Holm-Hansen et al., 1977). Nevertheless, large phytoplankton blooms do occur during spri ng and summer, particularly in waters associated with ice edges, polynyas, isla nds, and continental shelves (e.g., Smith and Nelson, 1985; Sullivan et al., 1993; Moore and Abbott, 2000; Arri go and van Dijken, 2003). Some of the largest blooms develop in the marginal ice zone (MIZ), in waters associated with the seasonal advance and retrea t of sea ice. As the ice edge recedes, low salinity meltwater produces a low density surf ace lens that reduces vertical mixing and shallows the mixed layer; consequently, phytoplankton are able to grow in a high irradiance stable environment. This phenomenon is common in the MIZ of the Weddell and Ross Seas, as well as in other regions around Antarctica (El-Sayed, 1971; El-Sayed and Taguchi, 1981; Smith and Nelson, 1985; Sedwick and DiTullio, 1997), with reported chlorophyll concentratio ns up to 190 mg m-3 (El-Sayed, 1971).
31 Polynyas are areas of open water surrounde d by sea ice, usually characterized by elevated surface chlorophyll concentrations. Arrigo and van Dijken (2003) examined surface chlorophyll concentra tions in 37 Antarctic polynyas and report values ranging from 0.24 to 7 mg chl m-3 during summer. Maximum chlorophyll concentrations occurred in the Amundsen and Ross Seas, in agreemen t with previous reports of up to 40 mg m-3 in the Ross Sea polynya (Arrigo and McClain, 1994). Antarctic islands can also be surrounded by chlorophyll-rich waters. For example, th e region around South Georgia has been characterized as highly productiv e, with high densities of phytoplankton and zooplankton, as well as large colonies of seals and seabirds (Atkinson et al., 2001). Phytoplankton blooms in this area have been observed from Novemb er to April with chlorophyll concentrations reaching 30 mg m-3. The high productivity at South Georgia has been attributed to a combination of factors including enhanced supply of iron and rapid recycling of nitrogen, favorable temper atures, and a shallow stable inshore water column (reviewed in Atkinson et al., 2001). The continental shelf waters west of th e Antarctic Peninsula (55 75 S, 50 80 W) (Fig. 3.1) in particular, are considered to be one of the most productive regions of the Southern Ocean, supporting high densities of phytoplankton, zooplankton, and upper trophic level predators (e.g., Fraser and Trivel piece, 1996; Arrigo et al. 1998; Deibel and Daly, 2007). Chlorophyll concentrations in the northern sectors along the Antarctic Peninsula shelf have been reported to reach values up to 38 mg chl m-3 during December and January (Holm-Hansen and Mitchell, 1991; reviewed in Smith et al., 1996a; Rodriguez et al., 2002). More recently, A rrigo and van Dijken (2003) reported that SeaWiFS chlorophyll in a phytoplankton bloom related to the Mar guerite Bay polynya
32 averaged 2.30 mg chl m-3, while Garibotti et al. (2003) a nd Meyer et al. (2003) observed summer chlorophyll concentrations up to 17.86 mg m-3 in 1997 and 25 mg m-3 in 2000, respectively, within Ma rguerite Bay. These findings sugge st that the southern sector of the Antarctic Peninsula also may support large phytoplankton blooms. Studies of the spatial and temporal dyna mics of phytoplankton along the northern part of the Peninsula reveal a pattern of chlorophy ll accumulation in coastal areas over the summer, with lower concentrations offshore (S mith et al., 1998a; Garibotti et al., 2003). These studies also identify an alongshore gr adient with higher bi omass in the northern sectors of the Antarctic Penins ula earlier in the productive season, which later progresses to the southeast as the sea i ce retreats in the same direc tion. Most studies to date, however, have only dealt with chlorophyll distributions nor th of Marguerite Bay and, thus, information on phytoplankton dynamics in the vicinity of Marguerite Bay and to the south in the Bellingshausen Sea is scarce. The Southern Ocean Global Ocean Ecosystems Dynamics (SO GLOBEC) program investigated the physical and biol ogical factors that in fluence the growth, recruitment, and overwintering survival of Antarctic krill, Euphausia superba, in the vicinity of Marguerite Bay (F ig. 3.1), west of the Antarctic Peninsula, during austral fall and winter of 2001 and 2002. Krill play a key role in the Antarctic ecosystem as one of the primary pelagic herbivores and prey for many predators. The large krill population west of the Antarctic Peninsula appears to be maintained by occasional strong year classes, with often poor recr uitment in the intervening years (Siegel and Loeb, 1995; Hewitt et al., 2003; Quetin and Ross, 2003). The suite of physical and biological factors that govern krill reproduc tion and recruitment, howev er, remain poorly known.
33 During the GLOBEC study, large differences in abundances of larval and juvenile krill were observed between the two years (Daly, 2004). During fall 2001, larvae were very abundant (< 0.01 132 ind m-3), with younger stages dominant offshelf and older stages dominant onshelf (Fig. 3.2, see Fig. 3.1 for station locations). Offshore larval densities (up to 132 ind m-3) are amongst the highest reported for the area and are comparable to those reported by RakusaSuszczewski (1984) for 1981. Few juveniles were observed anywhere in 2001. During fa ll 2002, relatively high concentrations of young larvae were again detected in oceanic waters (< 0.01 211 ind m-3), although average abundances were significantly lowe r than in 2001 and all larval stages were scarce in coastal areas (Fig. 3.2). In cont rast, juveniles were re latively abundant on the middle and inner shelf in the vicinity of Marguerite Bay (< 0.01 2.37 ind. m-3), indicating a successful recruitm ent from the 2001 larval populat ion. These results led me to investigate the environmenta l conditions that contributed to the large krill reproduction during austral spring and summer 2000/2001, and subsequent high larval densities. Herein, I investigate chlor ophyll dynamics west of the Antarctic Peninsula using SeaWiFS ocean color data between 1997 and 2004, with special emphasis on the Marguerite Bay region, to bett er understand the conditions that make it a suitable habitat for krill. I also examine the effects of the retreat of the ice edge on the timing and location of phytoplankton blooms west of the An tarctic Peninsula. Finally, I discuss the environmental mechanisms that potentially support successful kr ill reproduction and recruitment in this area.
34 METHODS The study area consisted of th e coastal waters west of the Antarctic Peninsula, adjacent deep waters in the Drake Passage and coastal and oceanic waters of the Bellingshausen Sea (55 75 S and 50 80 W), as chlorophyll concentrations in these areas are most likely to influence regional krill populations (Fig. 3.1). Krill were collected aboard the R.V. Lawrence M. Gould during austral autumn between 23 April and 6 June 2001 and between 7 April and 20 May 2002 as part of the US SO GLOBEC program. A total of 18 and 16 net tows were done at several stations during 2001 and 2002 respectively. Individuals were collected at eight discrete depth intervals using a 1-m2 MOCNESS (Multiple Opening-Closing Net and Environmental Sensing System) net system, with a 333 m mesh. Krill were identified for stages of larvae (calyptopis I III, furc ilia I VI), juveniles, or adul ts (males or females) after Makarov (1980) and measured for total length (f rom the base of the eye to the tip of the telson, excluding setae). Herein, the distri bution of krill stage abundances are compared at two representative stations; one offshelf (S ta. 1) and one onshelf (Sta. 5) (Fig. 3.2). Abundances are the mean of four onshelf a nd two offshelf net tows in 2001, and three onshelf and two offshelf net tows in 2002. Stage abundances ( A ind m-3) were calculated as a weighted mean for the sampling depth using: where i represents each of the eight nets (collections from diffe rent depth strata) in each
35 cast, x is the abundance of each kr ill stage in each net (ind m-3), and z is the depth interval of each stratum (m). A recruitment index (R1) for E. superba was calculated as the proportion of one year-old krill compared to age-class one plus all older age classes from all net samples for fall 2001 and 2002. Age-class one was defined as juvenile krill with total length ranging from 20 30 mm following Siegel et al (1998). The total abundance (ind m-3) of juveniles 20 30 mm and of juveniles 20 30 mm plus all older stages was estimated for each net and summed over the entire cruise to estimate a recruitment index for each cruise. Offshelf net tows (Sta. 1), which did not contain any juvenile or adult krill, were excluded from the recruitment calculation. A ll except one of the onshelf stations were located in inner coastal waters; therefore, recr uitment estimates were not biased as a result of migration of large krill to the inner shelf in late summe r and fall (Siegel et al., 2003; Hewitt et al., 2003). The chlorophyll dataset includes 6606 SeaW iFS daily Level 2 files (~ 1 km/pixel near nadir) between September 1997 and D ecember 2004 obtained from NASA Goddard Space Flight Center. These data were collected by ground stations, as well as occasional satellite onboard recording ove r the area using the most recent algorithms and software package (SeaDAS4.8). The level 2 data were mapped to a rectangular projection with approximately 1 km2/pixel for the western Antarctic Peninsula region (Fig. 3.1). The parameter used in this study is the surface chlorophyll concentration derived from the OC4v4 empirical band-ratio (blue versus gr een) algorithm (OReilly et al., 2000). SeaWiFS chlorophyll estimates are biased toward surface values as the sensor only sees the first few meters of the water column. Nevertheless, SeaWiFS chlorophyll provides a
36 good assessment of water column concentra tions in the Southern Ocean, as maximum chlorophyll concentrations are near surface a nd surface chlorophyll is well correlated with depth integrated chlorophy ll (Holm-Hansen and Mitchell, 1991; Holm-Hansen et al., 2004a; Korb et al., 2004). The accuracy of the SeaWiFS algorithm in the Southern Ocean is being debated. Several studies using in situ chlorophyll fluorescence determined that the SeaWiFS algorithm underestimates chlorophyll a concentrations in the Southern Ocean (e.g., Dierssen and Smith, 2000; Korb et al., 2004). However, recent results from more accurate in situ HPLC (High Performance Liquid Chroma tography) values demonstrated that SeaWiFS chlorophyll a estimates are accurate (bias = 12 %) for chlorophyll concentrations between 0.1 and ~ 4 mg m-3, which include > 90% of the wa ters in the Southern Ocean (Chapter 2; Marrari et al ., 2006). The mean ratio between SeaWiFS chlorophyll and in situ HPLC chlorophyll was close to unity (i.e ., 1.12) (n = 96), indicating good agreement between the two sets of data Although these previous re sults could not verify the SeaWiFS algorithm for chlorophyll concen trations greater than ~ 4 mg m-3, here I use the algorithm to estimate a larger range of chlor ophyll concentrations in order to investigate relative changes in chlorophyll spatial and temporal dynamics. In addition, because colored dissolved organic matter (CDOM) may introduce significant errors to the estimation of surface chlorophyll a from ocean color remote sensing data (e.g., Carder et al., 1989), I measured CDOM absorption from 10 surface water samples collected in the study area between April 4 and May 10, 2002. Ab sorbance spectra were obtained at 1 nm intervals from 200 to 750 nm with a Hitach i U-3300 double-beam spectrophotometer with 10-cm quartz cells. MilliQ water was used in the reference cell. Three scans were run for
37 Figure 3.1. (a) Map of the Antarctic contin ent showing the location of the study area and other geographic references. The solid line represents the 1000 m isobath and the dashed line indicates the Antarctic Ci rcle (66.3S). (b) Details of the Antar ctic Peninsula region and location of the US Sout hern Ocean GLOBEC offshelf (Sta. 1) a nd onshelf (Sta. 5) net sampling stations () represented in Fig. 2. (a) (b)
38 each sample and the resulting spectra were averaged to reduce noise. Data were corrected for scattering by subtracting absorbance at 700 nm from all measurements. Absorbance values were converted to absorption coeffici ents (Kirk, 1983). The absorption coefficient at 375 nm, a(375), was used as an index of CDOM concentration. Because a(375) was low in all samples (0.02 0.148 m-1), I conclude that the accura cy of the SeaWiFS chlorophyll estimates is not affected by the presen ce of significant levels of CDOM. The Southern Ocean daily satellite images normally have missing data due to a relatively high percentage of cl oud cover; therefore, at leas t weekly composites of data typically are needed to obtain good spatial c overage (e.g., Holm-Hansen et al., 2004a). In this study, biweekly composites of SeaWiFS chlorophyll data are used. A seven-year climatology of biweekly chlorophyll a concentrations (mean chlorophyll concentration at each pixel) was generated from the mapped Level 2 data from September 1997 through December 2004 and biweekly composite images were produced. When calculating the mean value in either the climatology or the bi weekly data, only valid data were used. Suspicious data identified by various quality fl ags associated with each pixel (for example cloud contamination, large solar/view angle, etc.) were excluded from the calculations. The study area was divided into fourteen subregions, each representing different oceanographic conditions in order to (1) i nvestigate the initiati on and progression of phytoplankton blooms, and (2) analyze the relative differences in chlorophyll concentrations between regions along the coas tal Antarctic Peninsula and offshore areas. Regions 1 6 represent offshelf oceanic regi mes with depths greater than 2000 m, while regions 7 and 8 are located ove r the continental shelf slope, defined as the area between 500 and 2000 m. Regions 9, 10 and 11 repres ent coastal waters along the Antarctic
39 Peninsula shelf and regions 13 and 14 are locat ed in Marguerite Bay. Region 12 includes both coastal and oceanic waters in the Sco tia-Weddell confluence area. The biweekly geometric mean chlorophy ll concentrations between September March 1997 to 2004 were calculated for each region and the resu lts plotted over time in relation to the climatology. Although values up to 55 mg m-3 were estimated from SeaWiFS for Marguerite Bay during su mmer, concentrations lower than 0.01 mg m-3 and greater than 20 mg m-3 were excluded from the geometric mean calculations, since the accuracy of these values could not be verified with concurrent in situ data. However, as most of the SeaWiFS chlorophyll concentrations during th e study period were within the 0.01 20 mg m-3 range, the geometric mean calculations were not significantly affected by excluding the few extreme values. Gaps in the time seri es occurred when the available data points were < 10% of the total number of pixels within each region duri ng the biweekly period of maximum spatial coverage. The mean location of the ice edge w ithin the study area during October, November, and December of 2000 and 2001 was determined using a two-dimensional linear interpolation of monthly ice concentration on a 25 km resolution grid. Monthly averaged gridded ice concentrations ge nerated using the NASA Team algorithm and Nimbus-7 SMMR and DMSP SSM/I passive mi crowave data were obtained from the National Snow and Ice Data Cent er (Cavalieri et al., 2005). The ice edge was considered to be the location where sea ice concentration was 15% (Gloersen et al., 1992). The mean location of the ice edge during these m onths was superimposed over the concurrent biweekly SeaWiFS chlorophyll images. In a ddition, the mean locati on of the ice edge during the preceding month also was plotted in order to evaluate changes in chlorophyll
40 concentrations within the regi on of ice edge retreat. To examine whether there was a relationship between the retr eat of the ice edge and the formation of phytoplankton blooms, chlorophyll concentrations along the location of the ice edge in September, October, and November 2000 and 2001 were ex tracted from the sa tellite data. The chlorophyll values at the same locations were extracted for the subsequent 6-biweekly periods (three months). Thus, at the Se ptember location of the ice edge, chlorophyll concentrations were analyzed during Septem ber November and at the November ice edge location, chlorophyll was analyzed during December January. The median chlorophyll concentration for each biweekly pe riod was estimated and plotted over time to generate a time series of chlorophyll con centrations at the ice edge location. The daily area free of sea ice (km2) in the northern and southern sections of Marguerite Bay was estimated from satellite data following the methods described in Arrigo and van Dijken (2003). A daily climatology from 1997 through 2004 was calculated as the mean daily ice-free area within each subregion. A running average was applied to reduce the daily vari ability (bandwidth = 9.1 days). The time series of daily ice-free area from 1997 to 2004 were plotted in relation to the 8-y ear climatology with special emphasis on 2001 and 2002.
41 Figure 3.2. Mean abundance (ind m-3) of development stages of Euphausia superba in offshelf waters (Sta. 1) west of the Antarc tic Peninsula and coasta l waters (Sta. 5) of Marguerite Bay during fall 2001 ( light grey) and 2002 (dark grey). CI CIII = calyptopis 1 3; FI FVI = furcilia 1 6; juvs = j uveniles ; males = adult males; females = adult females. Offshelf Sta. 10 2 4 6 8 10 12 14 16 C1C2C3F1F2F3F4F5F6JuvsFemMaleMean krill m-3 2001 2002 Onshelf Sta. 50 0.1 0.2 0.3 0.4 0.5 C1C2C3F1F2F3F4F5F6JuvsFemMaleMean krill m-3
42 RESULTS Biweekly climatological patterns of chlorophyll a indicate that oceanic and coastal areas in the Bellingshausen Sea and coasta l Marguerite Bay had persistently high chlorophyll concentratio ns (0.1 > 7 mg m-3) during spring and summer in comparison with any other area west of the Antarctic Peninsula between September and March 1997 2004 (Fig. 3.3). Oceanic waters offshore of th e northern Antarctic Peninsula typically had relatively low concentra tions (0.1 0.2 mg m-3). Intermediate values (0.1 ~ 2 mg m-3) were generally observed over the more norther n continental shelf regions west of the Antarctic Peninsula and downstream in the Scot ia Sea, although small, short-lived blooms with chlorophyll values greater than 2 mg m-3 occurred nearshore in this region. The spatial and temporal changes in ch lorophyll patterns suggest that biomass accumulations initially occurred during October and November in offshelf waters, mainly in the Bellingshausen Sea and to a lesser extent near the shelf break in the vicinity of the Shetland Islands at the northern end of the Antarctic Peninsul a (Fig. 3.3). As the season progressed (mid-December), phytoplankton blooms developed onshore, especially in the vicinity of Marguerite Bay a nd the coastal Bellingshausen Sea, where they remained well established until early April. In contrast, ther e didnt appear to be a significant seasonal increase in chlorophyll in northern Peninsula wa ters, except near some of the islands such as the South Shetlands and a few short-lived blooms nearshore. A cross-shelf gradient in chlorophyll concen trations (higher valu es in coastal areas to lower values at the shelf break) was obs erved for Marguerite Bay and to the south during summer (January March). Between Anvers and Adelaide islands, there was a similar cross-shelf gradient in January, but by late February maximum chlorophyll
43 concentrations had shifted to mid-shelf and out er shelf areas near A nvers Island. At the northern end of the Peninsula, there was no apparent cross-shelf gradient. Instead, maximum chlorophyll values occurred in the vi cinity of the Shetland Islands and Elephant Island near the outer shelf. The geometric mean chlorophyll concentra tions for the 14 subregions defined in the study area (Fig. 3.4a) illustrate the intera nnual (1997 2004) variability in patterns of chlorophyll distribution west of the Antarctic Peninsula. Variations from the climatology were relatively small (maximum 0.4 mg m-3) in offshelf regions of the Antarctic Circumpolar Current (ACC), Drake Passage, and the Scotia Sea (regions 1 6), with the Bellingshausen and Scotia Sea regions having higher variability than the Drake Passage regions (Fig. 3.4b). Regions 1 2 and 4 5 had a small chlorophyll peak in early November (year day 315), whereas the Scotia Sea generally sustained higher levels of chlorophyll for longer periods (unt il early April). Variability in waters at the tip of the Peninsula (region 12) was influenced by the larg e chlorophyll concentrat ions to the east of the Antarctic Peninsula in the Weddell Sea. Shelf slope waters (regions 7 and 8) had similar interannual ranges in chlorophyll, although the years with chlorophyll concentrations greater than climatology were not necessarily the same. For waters over the continental shelf (regions 9 11), the s outhern Bellingshausen and Marguerite Bay shelf region had the highest ch lorophyll concentrations and variability, with declining values to the northeast.
44 Figure 3.3. Biweekly climatology (1997 2004) of SeaWiFS chlorophy ll concentrations (mg m-3) between October and March. The date ra nge included in each image is indicated on the lower right hand corner. White areas indi cate no data due to the presence of clouds and/or sea ice. The white thin line represents the 1000 m isobath.
45 Figure 3.3. (Continued) The highest chlorophyll concen trations at the nor thern tip of the Peninsula (regions 3, 6, and 12) occurred during 1999/2000. Else where, 2000/2001 had substantially higher chlorophyll concentrations during summer (Decem ber February) compared with most of the other years analyzed, partic ularly in the Bellingshausen Sea (regions 1, 4, 7 and 9) and Marguerite Bay (regions 13 and 14). In wa ters over the continental shelf of the Bellingshausen Sea (regions 7 and 9), chlo rophyll was generally elevated during the spring and summer seasons of both 2000/2001 and 2001/2002, with high variability observed both between and within years. During 2001/2002, however, these chlorophyll peaks did not occur for as extended a period of time as in 2000/2001. Shelf break and
46 coastal regions along the Antarctic Peninsul a (regions 8, 10 and 11) had elevated mean chlorophyll concentrations rela tive to offshore areas, although the variations (up to 0.41 mg m-3) with respect to the climatology were less evident than in the Bellingshausen Sea (region 9, up to 0.97 mg m-3) or Marguerite Bay (up to 3.72 mg m-3), and mean values never exceeded 0.72 mg m-3 (note the different y-axis scales between plots). Marguerite Bay had the highest chlorophy ll concentrations in compar ison with any other region analyzed. In northern Margue rite Bay, average values duri ng January for all years were 1.24 1.31 mg m-3. However, during January 2001, mean chlorophyll reached 4.95 mg m3, whereas 2002 values were consistently near or below the 7-year average. In southern Marguerite Bay, 2000/2001 showed high mean values of up to 2.76 mg chl m-3 from late December through February, a factor of 2.5 higher than average (1.09 mg chl m-3) conditions. The presence of sea ice prevented satellite ch lorophyll data co llection during most of the 2001/2002 summer. The location, timing, and extent of sea ice during 2000/2001 and 2001/2002 were examined in relation to chlo rophyll concentrations to bett er understand the relationship between sea ice and phytoplankton blooms (Fi g. 3.5). Chlorophyll concentrations were highly variable in relation to the receding ice edge in our study area. During September 2000 and 2001, the ice edge was located in oceani c waters of the ACC. By October, the ice margin had retreated considerably and occurr ed closer to the coast at its eastern extent, but chlorophyll had not increased significantly at the Se ptember ice edge locations (Fig. 3.5a and 3.5b, top 2 panels). This suggests th at October was too ear ly in the productive season for any significant chlor ophyll accumulations to occur within the i ce edge zone. In November, the ice edge had receded onshelf in the mid-Antarctic Peninsula, but in the
47 Bellingshausen Sea the ice edge remained o ffshore in 2000 and approximately at the shelf break in 2001. Chlorophyll concentrations incr eased significantly in this region, reaching ~ 5 mg m-3 (Fig. 3.5a and 3.5b, center panels). A lthough ice edge blooms appear to have occurred in some parts of the Bellingshaus en Sea during November and December, most of this region of enhanced chlorophyll was pr esumably too far from the ice edge to have been influenced by ice melt processes. Figure 3.4a. Location of the 14 subregions along the western Antarctic Peninsula superimposed over the climatology (1998 2004) of SeaWiFS chlorophyll concentrations (mg m-3) for January 1 14.
48 Figure 3.4b. Time series of geometric mean ch lorophyll concentrations in each subregion for each biweekly period during 1997 2004. The 7-year climatology is also included for each region (thick black line with circles). No te the difference in scale of the y-axis for the different regions. The 2000/2001 (red squares), 2001/2002 (blu e circles), 2002/2003 (cyan broken line), and 2003/2004 (green dot ted line) spring-summer seasons are highlighted in color.
49 Figure 3.5a. Biweekly SeaWiFS chlo rophyll concentrations (Chl, mg m-3) in October (Oct), November (Nov) and December (Dec) of 2000. The mean monthly location of the ice edge is also shown: the red line indi cates the location of th e ice edge during the preceding month, the yellow line represents the current month. The 1000 m isobath is indicated by the white line.
50 Figure 3.5b. Biweekly SeaWiFS chlor ophyll concentrations (Chl, mg m-3) in October (Oct), November (Nov) and December (Dec) of 2001. The mean monthly location of the ice edge is also shown: the red line indi cates the location of th e ice edge during the preceding month, the yellow line represents the current month. The 1000 m isobath is indicated by the white line.
51 The analysis of chlorophyll buildup during the weeks following the retreat of the ice edge reveals an increase in chlorophyll c oncentrations in the vicinity of the October and November ice edge locations, both in 2000 and 2001; however, values usually reached a maximum 4 6 weeks after the ice had receded (F ig. 3.6). In addition, other areas along the Antarctic Penins ula that were never influen ced by sea ice also showed a similar increase during our study period. Fo r example in November and December of 2001 (Fig. 3.5b), the ice edge occupied coasta l areas of the Bellingshausen Sea and along the Antarctic Peninsula to Anvers Island (see Fi g. 3.1 for site locations). Even though ice never occupied the northern end of the Peni nsula, elevated chlorophyll concentrations were observed along the shelf break and in coas tal areas. Thus, pro cesses other than the retreat of the ice edge likely influenced phytoplankton dynamics in this more northern area. Weeks after ice has receded 0246810 Median chlorophyll 0.0 0.2 0.4 0.6 0.8 1.0 Ice in September 2000 Ice in October 2000 Ice in November 2000 Ice in September 2001 Ice in October 2001 Ice in November 2001 Figure 3.6. Biweekly time series of chlor ophyll accumulation after th e ice had receded at the September, October, and November 2000 an d 2001 locations of the ice edge shown in figure 3.5.
52 Sea ice coverage in Marguerite Bay also showed strong variability between the years analyzed and these differences are particularly marked between 2000/2001 and 2001/2002. During summer and early fall (January April), typical values of ice-free areas range from approximately 9,000 to 11,000 km2 in northern Marguerite Bay and from ~ 4,500 to 7,500 km2 in southern Marguerite Bay (F ig. 3.7c). A comparison of the climatology and 2001 and 2002 daily ice-free areas (km2) indicates that 2002 had above average sea ice in both the northern and s outhern sectors throughout the spring, summer, and fall (Fig. 3.7a and 3.7b). In addition, ice formed earlier in 2002 than in 2001. In contrast, 2001 had sea ice values significantl y below the 8-year mean, particularly from January through July, as indica ted by the unusually large icefree areas observed both in the northern and southern sectors. In 2001 these values reached approximately 12,000 and 11,500 km2 in the northern and southern regions respectively. On the other hand, the areas free of ice only reached 6,000 9,000 km2 in the northern and 0 2,000 km2 in southern sectors during the same months in 2002, suggesting an especially extensive sea ice cover. During winter (start ing in mid-July or year day ~ 200), sea ice conditions were similar for both years, although, as mentione d above, ice occupied both the northern and southern sectors considerably earlier in 2002. Other years with above-average sea ice cover were 1997/1998 and 1999/2000, while during 1998/1999 and 2003/2004 sea ice was lower than normal. Recruitment indices (R1) for E. superba collected during fall in the GLOBEC study area were 0 for 2001 and 0.4 for 2002, repres enting no juvenile recruitment from the 1999/2000 larval year class and significant re cruitment from the 2000/2001 larval year class, respectively. Publishe d krill recruitment index valu es for years of elevated
53 recruitment west of the Antarctic Peni nsula between 1975 and 2002 are provided for comparison (Table 3.1). Figure 3.7. Daily ice-free area (km2) in (a) northern and (b) southern Marguerite Bay during 1997 2004. The 8-year climatology also is shown (black thick line). (c) Location of the northern (dark grey) and southern (light gr ey) Marguerite Bay regions. (a) (b) (c)
54 Table 3.1. Years of elevated krill recruitment between 1975 a nd 2002. Source publications are designated by symbols in the table. Recruitment indices (R1) were estimated from the proportion of one year-old krill compared to age-cl ass one plus all older ag e classes. Year class is the year larvae were produced; whereas recruitm ent to the juvenile st age occurs the following year. All krill were collected during austral summer (January Ma rch) in waters west of the Antarctic Peninsula, except th is study when krill were coll ected during aust ral fall (April June). Only years of high recruitment (R1> 0.5) are shown Year class R1 Location Study period 1979/1980 0.559 Elephant Island area 1975 1996 1980/1981 0.757 Elephant Island area 1975 1996 1987/1988 0.651 Elephant Island area 1977 1997 1994/1995 0.622 Elephant Island area 1975 1996 0.639 Western Antarctic Peninsula # 1985 2002 1999/2000 0.573 0 0 Elephant Island area Western Antarctic Peninsula # Marguerite Bay region § 1975 2000 1985 2002 2001 2002 2000/2001 0.748 Western Antarctic Peninsula # 1985 2002 0.400 Marguerite Bay region § 2001 2002 Siegel et al. (2002); # Siegel et al. (2003); § this study
55 DISCUSSION Chlorophyll concentrations showed consider able temporal and spatial variability in waters west of the Antarctic Peninsula during austral spring and summer between 1997 and 2004, with the largest and most persis tent phytoplankton blooms consistently occurring in Marguerite Bay and the Bellingshausen Sea areas. The climatology showed a cross-shelf chlorophyll gradient in early su mmer in the middle and southern regions of the Peninsula. Investigators from the Pa lmer Long Term Ecological Research (LTER) program have reported a similar gradient in chlorophyll concentrations for January 1991 1995 (Smith et al., 1998a) and January 1997 (Gar ibotti et al., 2003) between Anvers and Adelaide islands, suggesting th at this pattern of chlorophyl l distribution has remained relatively constant in this area for abou t two decades. In 1997, average chlorophyll concentrations in coastal areas were an or der of magnitude greater than those in the vicinity of the shelf break (4.38 mg m-3 vs. 0.22 mg m-3) and corresponded to a shallower mixed layer and greater vertical stability of the water column in coastal waters (Garibotti et al., 2003). Climatology results presente d here indicate that these environmental conditions may usually change by midFebrua ry in the middle region of the Peninsula (Anvers to Adelaide islands) as this crossshelf pattern was no longe r present, except in Marguerite Bay and to the south. The above mentioned studies also descri be an alongshore gradient in chlorophyll concentrations with higher va lues in the northern sectors of the Peninsula earlier in the season possibly associated to the seasonal al ongshore retreat of the sea ice and/or to latitudinal differences between areas. Pres ent results, however, i ndicate that chlorophyll concentrations in the southern sectors are c onsistently higher than in any other area
56 analyzed west of the Antarctic Peninsula. In addition, chlorophyll ac cumulations in these areas occur earlier in the spri ng and persist longer throughout the summer than most areas in the northern regions. T hus, the southern areas are vita lly important to the Antarctic Peninsula ecosystem in terms of overall ch lorophyll standing stoc k and the phytoplankton blooms are likely to play an important role in supporting higher troph ic level dynamics. Few studies provide inform ation on phytoplankton dynamics in the Bellingshausen Sea (Savidge et al., 1995; Barlow et al., 1998). SeaWiFS images of the circumpolar distribution of mean annual chlo rophyll concentrations show that the Bellingshausen and Amundsen Seas support large phytoplankton bloo ms (Fig. 1 in El-Sayed, 2005). Satellite derived estimates of primary productivity also indicate that the Bellingshausen/Amundsen Sea area is one of the most productive in Antarctic waters, only exceeded by the Ross and Weddell Seas (Arrigo et al., 1998) Typically, ice edge bloo ms are tightly coupled to spring ice edge retreat in Antarctic waters (Sullivan et al., 1993; Arrigo and McClain, 1994; Garibotti et al., 2005), where blooms develop within about two weeks after the ice recedes from a particular location. During austral spring 1992, investigators from the UK STERNA Program observed elevated chlorophyll concentrations in oceanic waters of the Bellingshausen Sea in late November (up to > 7 mg m-3) and early December (up to 2.4 mg m-3), but reported that they were not relate d to sea ice retreat (Savidge et al., 1995; Barlow et al., 1998). During the present study, some blooms appeared to occur along the ice edge in the Bellingshausen Sea; however, most blooms occurred with approximately a 4 6 week lag and, therefore, probably were not related to the ice re treat, in agreement with the STERNA Program observations. Clearly, there is a need for future phytoplankton studies in the Bellingshausen S ea in order to elucidate the factors that
57 control the formation and persistenc e of spring blooms in the region. Ice edge blooms have been suggested to be an important feature in the northern Antarctic Peninsula region (e.g., Siegel and Lo eb, 1995; Smith et al., 1998b). In contrast, findings reported herein indicate that the formation of spring blooms was not necessarily coupled to the retreat of the ice edge in the vicinity of the northern Peninsula between 1997 2004. Instead, blooms first appeared near the shelf break and gradually progressed to more coastal areas, suggesting that shelf break processes were likely an important factor influencing phytoplankton growth in th e northern Peninsula region. The strong currents of the eastward flowing ACC interact with the bathymetry when it encounters the shelf break, generating meanders that usually can be detected at the surface. Antarctic surface waters are rich in m acronutrients; however, iron deficiency has been proposed as a factor limiting phytoplankton growth (De Baar et al., 1995; Holm-Hansen et al., 2004b; 2005). The importance of upwelled iron-ri ch deep ACC waters to chlorophyll aggregations has been described for several regions of the Southern Ocean, including the Scotia Sea, the Polar Front region downstream of South Georgia, the Ross Sea, and the Antarctic Peninsula shelf break (De Baar et al., 1995; Meas ures and Vink, 2001; Przelin et al., 2000; 2004; Holm-Hansen et al., 2005). Hence, upwelling of iron-rich deep water, rather than the retreat of the ice edge, may be a major factor controlling phytoplankton bloom development during spring and summer in the vicinity of the shelf break, and in coastal waters along the north ern Antarctic Peninsula. Although sea ice extent and duration in the Bellingshausen Sea and along the Antarctic Peninsula has decreased over the past 25 years (Parkinson, 2002; Ducklow et al., 2006), high interannual variab ility in sea ice is still obs erved. Ducklow et al. (2006)
58 analyzed 14 years (1991 2004) of sea ice extent data near Palm er Station in the vicinity of Anvers Island, and found that 2001 had the lowest (69,932 km2) winter sea ice extent of all years analyzed, while 2002 had the highest (109,936 km2) (mean = 91,112 km2). The early retreat of sea ice in Marguerite Bay in spring of 2001 resulted in large phytoplankton blooms in ice-free waters duri ng summer. In contrast, the persistent presence of sea ice in Marguerite Bay dur ing summer fall 2002 resulted in overall lower chlorophyll concentrations in coastal surface waters. The positive relationship between sea ice cover and chlorophyll c oncentrations is further s upported by observations during other years of this study. For example in 2003/2004, peaks of above-average chlorophyll concentration in northern and southern Marguerite Bay durin g January (Fig. 4b; regions 13 and 14) coincided with an early retreat of the sea ice in la te 2003 and lower than normal sea ice extent during late 2003 early 2004, particularly in the northern region (Fig. 3.7). In addition, the ice-free area in northern Marguerite Bay during late 1998 early 1999 was larger than average, concurrent with high chlorophyll co ncentrations in the region during December January 1998/1999. Several investigations have suggested th at sea ice extent and duration are the primary environmental factors influencing krill recruitment in the northern regions of the Antarctic Peninsula, as spring summer ice edge blooms were believed to support krill reproduction and winter sea ice biota to provide food for overwintering larvae (Kawaguchi and Satake, 1994; Siegel and Lo eb, 1995; Quetin and Ross, 2003). Results from this study, however, indica te that ice edge blooms are not prevalent in this region and, thus, may not be a primary source of f ood for reproducing krill. In addition during both winters of the GLOBEC study, sea ice bi ota concentrations were very low (0.05
59 0.07 mg chl m-3) at the ice-water interface where larval krill feed (Daly, 2004). Indeed, a large percentage of larvae were not even a ssociated with the unders urface of sea ice and instead remained in the water column, particularly in 2002. Al so in both winters, larvae showed evidence of food limitati on, as indicated by delayed development, decrease in growth rates, increased intermolt period, and decrease in dry we ight, body carbon, and nitrogen (Daly, 2004). Hence, the presence of extensive sea ice during winter is not necessarily a good predictor of food avai lability for overwintering larvae and, environmental factors other than winter sea ic e conditions must play an important role in recruitment. Krill typically reproduce during late spri ng and summer (November March) west of the Antarctic Peninsula (Siegel, 1988). Results from net samples suggest that the majority of the females migrate near the vici nity of the shelf break where they spawn in oceanic waters (e.g., Siegel, 1992; Hoffma n et al., 1992), possibly owing to the predictable shelf break blooms that occur earl y in the productive seas on as observed in the climatology (Fig. 3.3). Successful krill repr oduction and larval survival require an adequate food supply (Ross and Quetin, 1983; 1989). Adult females may require above average phytoplankton concentrations (1 5 mg chl m-3) to initiate repr oduction (Ross and Quetin, 1986) and relatively high chlor ophyll concentrations (> 0.5 mg chl m-3) to sustain multiple spawning throughout the summer (Nicol et al., 1995). It also is critical for the first-feeding larvae (calyptopi s I) to encounter an adequa te food supply in the euphotic zone within 10 14 days, otherwise they will not survive (Ross and Quetin, 1986). Hence, knowledge about differences in the timing, extent, and evol ution of phytoplankton blooms is critical for understanding the in terannual variability observed in krill
60 recruitment success. The recruitment indices in Table 1 indi cate that only six out of 27 years had successful recruitment. Our fall (A pril May) estimated recruitment (R1 = 0.4) for 2000/2001 is lower than the 0.748 reported for summer by Siegel et al. (2003) (Table 3.1). A decrease in krill recruitm ent indices is commonly observed between summer and fall (Siegel and Loeb, 1995), owing to the seas onal decline in kril l population abundance (Lascara et al. 1999). The lowe r fall indices correla te with summer values and, therefore, can still serve as a relative indicator of recr uitment. The two highest recruitment years resulted from high reproduction in 1980/1981 and 2000/2001. Elevated recruitment from 2000/2001 larvae also was observed downstream at South Georgia (Siegel et al., 2003). In contrast, conditions during 2001/2002 were not as favorable for a successful reproduction, as evidenced by the lower numbe r of larvae recorded during fall 2002 (this study; Daly, 2004) and the subsequent low numbe rs of krill observed during January 2003 along the Peninsula (Ducklow et al., 2006). Results reported herein suggest that seasonal pers istence of elevated food concentrations from phytoplankton blooms dur ing spring and summer likely were a strong influence on krill reproducti on and recruitment, particular ly in 2001. Recruitment also may vary along the Peninsula depending on th e location of phytoplankton blooms. For example during 1999/2000, higher chlorophyll con centrations occurred in the Elephant Island Scotia Sea areas rela tive to any other year analy zed (Fig. 3.4b: regions 3, 6, and 12). Published krill recruitment index values indicate that elevated recruitment occurred in 2001 from the 1999/2000 larvae (R1 = 0.573) in this area (Siegel et al., 2002). On the other hand, chlorophyll concentrat ions in the southern sectors of the western Antarctic
61 Peninsula were average or below-average in coastal waters incl uding Marguerite Bay early in the season (regions 7, 8, 9, 10, 11, 13), although by February an increase could be observed in most regions along the continen tal shelf. Consistent with the lower availability of food during the critical early reproductive peri od, recruitment indices in the southern part of the study area for the 1999/2000 larvae were low [R1 = 0.076 in summer; (Siegel et al., 2003) and R1 = 0 in fall (this study)]. Chlorophyll concentrations in the Be llingshausen may have influenced downstream densities of larval krill in the vici nity of Marguerite Bay, especially at the offshelf station in the ACC. Recent modeli ng studies suggest that krill spawned in the Bellingshausen Sea are transported downstream to the Western Antarctic Peninsula area and into the Scotia Sea (e.g., Thorpe et al., 2007). The wide range of larval stages observed in offshelf waters during fall 2001 indicated that krill reproduction started relatively early in the season and continued for an extended period. The dominant larval modes included CIII, FI, and FII, in additi on to considerable numbers of late stage furcilia. Based on experimentally determin ed growth rates (Ikeda, 1984), these larvae represent a range of spawning episodes betw een mid-December (FVIs: ~ 127 days old) and early to late-March (CIIs: ~ 44 days ol d). Assuming that most reproducing adult females released eggs in the vicinity of the shelf break (Siegel, 1988), the approximate location of the spawning populat ion upstream in the ACC may be estimated from the age of larvae and the transport rate of the curren t. For example, the dominant FI mode in 2001 is estimated to be about 63 days old and, therefore, likely originated from a late February early March reproductive event. Surface velocities in the ACC reach 0.25 0.4 m sec-1, but decrease monotonically with dept h (Klinck and Nowlin, 2001). Mesoscale
62 meanders and eddies also may act to redu ce the transport rate. Assuming an average eastward current velocity of the ACC of ~ 0.1 0.2 m sec-1, the spawning location possibly occurred in offshelf waters of the Bellingshausen Sea between 83 94 W (Fig. 3.8; white box). Although this spawning locatio n is west of our study area, larger scale ocean color imagery (Fig. 3.8a) shows large bl ooms in this region in 2001, which would provide adequate food for an early and ex tended reproduction by adults and provide food for larvae transported to the east. In contrast during 2002 chlorophyll concentrations during summer were relatively low (Fig. 3.8b) and, thus, there was less food available for reproduction or larval growth. The results obtained during the present study indicate that spring and summer phytoplankton blooms appear to be a significant factor influenc ing krill recruitment in the vicinity of the Antarctic Peni nsula. Clearly, further stud ies of the factors controlling phytoplankton blooms in waters adjacent to th e southern Antarctic Peninsula and in the Bellingshausen Sea are warranted, especially sin ce this area may play a major role in krill reproduction and influence othe r components of the Antarc tic food web. In addition, further physiological-based krill studies are needed to better understand the relative impact of summer phytoplankton blooms vers us winter sea ice cover in governing recruitment, particularly in light of the regional decline in sea ice.
63 Figure 3.8. Monthly mean SeaWiFS chlorophyll concentrations in the Bellingshausen Sea during February (a) 2001 and (b) 2002. The white box represents the area of potential krill spawning. The red star indicates the locat ion of the US SO GLOBEC sampling Station 1 where krill were collected. Images correspond to the level-3 standard mapped images (resolution of 9 km/pixel) and were obtained from the global ocean color imagery ( http://oceancolor.gsfc.nasa.gov/cgi/level3.pl ). Marguerite Bay Bellingshausen Sea Bellingshausen Sea Marguerite Bay Antarctic Psula (b) February 2002 (a) February 2001
64 CHAPTER 4 PHYSICAL AND BIOLOGICAL CONTROLS ON INTERANNUAL VARIABILITY OF ZOOPLANKTON IN MARGUERITE BAY, WESTERN ANTARCTIC PENINSULA, AUSTRAL FALL 2001 AND 2002 INTRODUCTION Zooplankton are a key com ponent of marine ecosystems, acting as the link between primary producers and higher trophic levels. In the Southern Ocean, the Antarctic krill, Euphausia superba, often dominates zooplankton biomass and is considered a keystone species It is a relatively large euphausiid, up to 65 mm in length, and an important phytoplankton grazer, partic ularly during spring and summer, and prey for many upper trophic level predators, incl uding fish, penguins, seals, and whales. E. superba has a circumpolar distribution and in de ep water can often be observed in dense aggregations over the continental shelf. Due to its vital role in the Antarctic ecosystem, the zooplankton literature is dom inated by studies of this spec ies. Extensive research has focused on examining the distribution and abundance of E. superba (e.g., Marr, 1962; Siegel, 1992), its reproduction and growth (R oss and Quetin, 1983; Brinton et al., 1986; Daly, 1990; Quetin et al., 2003), behavior (e.g., Hamner et al., 1983; Marschall, 1988) and role in the Antarctic foodweb (Hempe l, 1985; Laws, 1985). In addition to this species, other zooplankton, including othe r euphausiid species and copepods, are
65 important components of the Antarctic eco system. Nonetheless, the information available on their role in the ecosystem population dynamics, and ecology is limited compared to that of E. superba Thysanoessa macrura an omnivorous euphausiid up to ~ 29 mm in length (Nordhausen, 1992), is broadly di stributed in coastal and ocean ic waters of the Southern Ocean and at all temperature and salinity ranges (Hempel and Marschoff, 1980). Its distribution overlaps that of E. superba and both species often co -occur. The larvae of T. macrura are herbivorous and graze on phytoplan kton during spring and summer, whereas the adults prey mainly on copepods (Hopkins, 1985). E. crystallorophias is another relatively small euphausiid (up to ~ 36 mm) fo und exclusively in neritic waters, where it either co-exists or replaces E. superba (Daly and Zimmerman, 2004). E. crystallorophias is mainly herbivorous thr oughout its life cycle. Both E. cyrstallorophias and T. macrura can be a significant food item for higher troph ic level predators, in cluding penguins, fish and whales and, thus, are also important me mbers of the Antarctic foodweb (Croxall, 1984; Nemoto and Nasu, 1958; Ai nley and DeMaster, 1990). Copepods dominate zooplankton abundan ce in many regions of the Southern Ocean and their biomass may exceed that of krill at times (Schnack-Schiel and Mujica, 1994). The dominant species include the herbivore Calanoides acutus and several omnivorous species, including Metridia gerlachei Calanus propinquus and Ctenocalanus spp. Several studies have analyzed the patterns of distribution and abundance of the dominant copepods in different areas of the Southern Ocean and have emphasized their critical role in the zooplankton community, bot h as prey for carnivorous zooplankton and fish, and also as predat ors of phytoplankton a nd microzooplankton (e.g.
66 Hopkins, 1985; Conover and Huntley, 1991; Sc hnack-Schiel et al., 1991; Lopez and Huntley, 1995; Atkinson, 1998; Voronina, 1998; Schnack-Schiel, 2001). In addition, other zooplankton, such as E. superba have been observed to rely on copepods during fall and winter when phytoplankton is scarce in the water column (Hopkins, 1985; Daly, pers. comm .). The western Antarctic Peninsula (WAP) is one of the most productive areas of the Southern Ocean, supporting high concentr ations of phytoplankton, zooplankton and upper trophic level predators (Dei bel and Daly, 2007). This area also is of interest due to its rapid warming relative to any other area in the worlds ocean (Vaughan and Doake, 1996) and the dramatic decline in sea ice c over observed in recent decades (Parkinson, 2002), which in turn has important implicat ions for Antarctic organisms. Although elevated chlorophyll concentrations have been observed along the continental shelf of the northern WAP during austral spring and su mmer (Moline et al., 1997; Smith et al., 1998a; Garibotti et al., 2003), recent results have shown that phytoplankton blooms are particularly large and persiste nt in the southeastern sector s, including Marguerite Bay and the western Bellingshausen Sea (Chapter 3; Marrari et al ., 2008). The factors that contribute to the relatively hi gh productivity of this region include its northern location relative to other Antarctic shelf areas, a relatively wide continental shelf, and the intrusion onto the shelf of nutrient-rich U pper Circumpolar Deep Water (UCDW) (Deibel and Daly, 2007). Numerous st udies also have examined zooplankton in the WAP region, which have improved our understanding of th e patterns of abundance, distribution and reproduction of the dominant taxa (e.g. H opkins, 1985; Siegel, 1988; 1992; Mujica, 1989; Schnack-Schiel and Mujica, 1994; Ross et al., 19 96; Siegel and Harm, 1996; Loeb et al.,
67 1997). However, these studies almost exclusively took place in the northernmost sectors of the Peninsula, from Adelaide Island toward the northeast, and only a few extended further south to the outer shelf off Marguer ite Bay (Lascara et al., 1999, Meyer et al., 2003). Despite the critical role of the southern sector in the ecosystem of the WAP, studies of zooplankton from Marguerite Bay and the Bellingshausen Sea have been scarce (Atkinson, 1995; Siegel and Harm, 1996; Meyer et al., 2003). The Southern Ocean Global Ocean Ecosystem Dynamics Program (SO GLOBEC) focused its study in the vicinity of Marguerite Bay, as this region was believed to be an important overwintering habitat for E. superba based on observations of predators during winter (Hofma nn et al., 2004). The main ob jective of SO GLOBEC was to investigate the physical and biological fa ctors that influence the growth, recruitment and overwintering survival of E. superba (Hoffman et al., 2004). The program supported two ships operating simultaneously during two fa ll and winter cruises in the vicinity of Marguerite Bay in 2001 and 2002. In additi on to krill data, an extensive dataset involving other important zooplankton groups was generated from net samples. Marguerite Bay should be a favorable environment for zooplankton owing to persistent elevated concentrations of phytopl ankton, the availability of protected areas such as fjords and bays that serve as ref uge from advection out of the area, and the intrusion of UCDW onto the shelf through Marguerite Trough, a deep canyon (> 500 m) that intercepts the continental shelf break off Ma rguerite Bay. In addi tion, the flow of the Antarctic Peninsula Coastal Current (APPC) (B eardsley et al., 2004; Klinck et al., 2004; Moffat et al., 2008) in this area involves a gyre-like feature that contributes to making Marguerite Bay a favorable retention area for both phytoand zooplankton, and thus a
68 potentially favorable feed ing ground for predators. Understanding the dynamics of zooplankton populations and how they respond to environmental change is critical to assessi ng the impact that these changes will have on the Antarctic ecosystem as a whole, and in pa rticular on upper trophic level predators that rely on zooplankton as a food source. Herei n, I investigate the pa tterns of abundance and distribution of dominant z ooplankton in Marguerite Ba y during austral fall 2001 and 2002 in relation to the interannual variability of environmental conditions, and examine possible ecological re lationships within and between groups. METHODS The study area consisted of coastal waters in the vicinity of Marguerite Bay along the western Antarctic Peninsula region betw een 66 70 S and 67 73 W. Zooplankton and environmental data were collected betw een 23 April and 6 June 2001 and 7 April and 20 May 2002 during process cruises onboard the R.V. Lawrence M. Gould as part of the SO GLOBEC Program. The process cruises o ccupied six coastal stations and samples from a total of 12 zooplankton net hauls were obtained during each cruise (Fig. 4.1). In addition, Video Plankton Record er data of zooplankton were obtained aboard the R.V. Nathaniel B. Palmer during survey cruises along 13 tr ansects spaced 40 km apart on the mid to outer shelf off Marguerite Bay (Ashjia n et al., 2008). Results reported here are derived from net samples obtai ned suring survey cruises. Zooplankton samples were collected at ei ght discrete depth intervals using a 1m2 Multiple Opening-Closing Net System and Environmental Sensing System (MOCNESS), having 333 m mesh. Maximum sampling depths ranged between 200 and 800 m,
69 depending on bathymetry. Samples were pr eserved in 10% formalin and stored for analysis in the laboratory. Samples were initially split to include approximately 100 individuals of the dominant e uphausiid species. For copepod counts and identification, samples were split further to include appr oximately 100 individuals of the dominant copepod. All zooplankton taxa present in the subsamples were identified and counted. Euphausiids were identified to species and deve lopmental stage (larva l stages, juveniles, adult females and adult males) after Ma karov (1981) and Mauchline (1981), and measured for total length to the nearest half mm (from the base of the eye to the tip of the telson, excluding setae). For plotting purposes, length data were grouped in 1 mm length bins. Only data for juveniles > 20 mm total length and adults are in cluded in this study. Copepods were identified for species and e numerated. For Euchaetidae, the designation of Park (1994) was followed, who ascribed the Antarctic species to the genus Paraeuchaeta All other zooplankton groups were identified, counted, measured, and classified into > 15 mm or < 15 mm total length; however, anal yses in this study combined the abundances of these size categories.
F ( c r e a l w igure 4.1. ( a c ircles) duri n e ferences ar e Wate r l l net hauls u w here n is ab sampled 2001 (b) a ) Location o n g fall of (b ) e also indic a r column int e u sing: undance (in d by net i ca l 3 0 Sta3 A o f the study a ) 2001 and ( c a ted. MB is M e grated abu n d m -3) in ne t l culated as t h 500m 0 00 m St a Sta5 Sta2 Sta6 Sta4a Sta4b A vian Is. (a) 70 a rea (red re c c ) 2002. Ge n M arguerite B n dance ( A, i n i i nA8 1 t i, and z is t h h e differenc e a 7 ( c tangle) and n eral station B ay. n d m -2) was i i z h e depth int e e between t h 2002 Laza r Ba y ( c) MOCNES S locations a n calculated fo e rval (m) of h e depth at w r ev y Ge o V So u Adelai d Alexander Is. Marguerite Trou g h Laubeu Fjord S net hauls n d geograph fo r each taxa the stratum w hich the ne t Bourgeoi Fjord Crystal Sound Inner MB o rge V I u nd d e Is. Neny Fjord f ic in t was s
71 closed and the depth at whic h it was opened (e.g., if stratum sampled was 100 150 m, z is 50 m). A mean water column integrated abundance was estimated for each taxa during each year (n = 12). In addition, a mean inte grated abundance was calculated for all taxa at each station, when multiple net hauls were done at a station. The depth of maximum abundance ( Z ) was obtained for each taxa in each net haul by calculating the center-depth of the stratum with the highest abundance at a given location. A mean depth of maximum abundance was calculated for each taxa during each year. Abundances and depth distributions for all taxa were tested for normality (Shapiro-Wilk W Test) (Shapiro et al., 1968). Abundances ( A ) and depths of maximum abundance ( Z ) for all dominant copepod species followed a normal distribution; therefore, statistical analyses for copepod parameters included arithmetic means, Students t-Test, a nd one-way ANOVA at = 0.05 (Zar, 1984). Abundances and vertical distributions of all other groups, including euphausiids, we re not normally distributed; thus, statistical analyses re ported here are nonparametric and included geometric means, Mann-Whitney U Test, and Kruskall-Wallis ANOVA at = 0.05 (Zar, 1984). Vertical profiles of conductivity, temperat ure, and density were obtained with a CTD mounted on the MOCNESS during all ne t hauls. In addition, water column chlorophyll samples were collected at seve ral CTD stations in proximity to the MOCNESS locations. A total of 15 and 6 casts are available for fall 2001 and 2002, respectively. At each CTD station, chlorophyl l samples for 4 9 depths were collected using 10-L Niskin bottles mounted on a rosette. Water was filtered on to GF/F filters and pigments were extracted in 90% acetone at -20 C in the dark for at least 24 hours.
72 Chlorophyll fluorescence was measured on a Turner Design Digital 10-AU-05 fluorometer calibrated prior to each cruise. MOCNESS tows were matched to the geographically closest CTD cast for comparisons between zooplankton and environmental data (Fig. 4.1). Spearman rank or der correlations were calculated between the vertically integrated abundance of all zooplankton taxa and se veral environmental variables: vertically integrated pigment c oncentrations (chloro phyll + phaeopigment; mg m-2), bottom depth (m), and salinity at 10 m. In addition to the concurrent enviro nmental variables measured during fall, monthly mean chlorophyll concentrations between 45 75 S and 50 80 W were obtained from SeaWiFS for the preceding summer season. More detail on the processing of SeaWiFS data are presented in Chapters 2 a nd 3. Analyses in this chapter include five of the subregions defined in Chapter 3 (Fi g. 3.4a), which represent oceanic and coastal waters of the Bellingshausen Sea and Marguerite Bay. The geometric mean chlorophyll for the five subregions was calculated for each biweekly period from October April during spring-summer 2000/2001 and 2001/2002. Da ta are presented in relation to the seven-year climatology calculated as th e geometric mean for 1997 2004 at each subregion. RESULTS Abundance and Percent Contribution The zooplankton of Marguerite Bay co mprised 13 major taxonomic groups,
73 including 12 species of cope pods, three species of euphausiids, and 10 other taxa (Tables 4.1 and 4.2). The dominant copepods included Calanoides acutus Metridia gerlachei Ctenocalanus spp., and Paraeuchaeta spp. Oithona spp., Oncaea spp., Calanus propinquus Rhincalanus gigas Gaidius tenuispinus and Scolecithricella minor were present in generally lower numbers. Aetidopsis antarctica and Metridia curticauda were present only occasionally in low abundances and, therefore, were excluded from the remaining analyses. Overall, copepods we re significantly more abundant in fall 2001 (24,362 11,948 ind m-2) than 2002 (9,496 6,450 ind m-2) (Students t-Test; p < 0.001) (Fig. 4.2a). During 2001, abundances were highe st in Laubeuf Fjord (Sta. 5), at the northern end of Alexander Island (Sta. 4a), George VI Sound (Sta. 4b), Lazarev Bay (Sta. 3), and Neny Fjord (Sta. 6). In 2002, copepods were most abundant in the vicinity of Alexander Island and southern Laubeuf Fjord. Densities were lowest in Crystal Sound (Sta. 7), Bourgeois Fjord (Sta. 5), and at the southern end of Adelaide Island (Avian Island). Abundances also were low in nor thern Laubeuf Fjord during 2002. In addition, copepod species percent composition showed interannual differences (Fig. 4.2b). In 2001, C. acutus and M. gerlachei dominated the community (mean = 52 and 38% of the total copepod abundance, respectively), followed by Paraeuchaeta spp. (5%). In contrast, during 2002 the number of relative ly abundant species (> 5%) was higher and included M. gerlachei (32%) Ctenocalanus spp. (27%), C. acutus (18%), Oithona spp. (10%), and Paraeuchaeta spp. (8% ) C. propinquus and R. gigas two relatively large species frequently found in high abundances in the Southern Ocean, only accounted for < 3% of total copepod abunda nce during both years. Oncaea spp., a frequently dominant small copepod (< 1 mm) was generally presen t in low abundances throughout the study
74 area, and only accounted for 1.4% and 2.6% of total copepods during 2001 and 2002, respectively. Table 4.1. Copepod abundance (ind m-2) in the vicinity of Marguerite Bay during austral fall 2001 and 2002. Mean = arithmetic mean, SD = standard deviation, n = number of net hauls in which a species was present Taxonomic group Fall 2001 Fall 2002 MeanSDnMean SD n Copepods Aetidopsis antarctica 10.937.916.29 21.8 1 Calanoides acutus 133587979121909 1891 12 Calanus propinquus 14512110173 174 11 Ctenocalanus spp. 263174122497 1856 11 Gaidius tenuispinus 79.875.9932.9 36 7 Metridia curticauda 0007.87 27.3 1 Metridia gerlachei 88835487123153 2390 12 Oithona spp. 55.255.912908 994 12 Oncaea spp. 27428612239 264 10 Paraeuchaeta spp. 116569612560 440 12 Rhincalanus gigas 58.667.01012.4 30.0 7 Scolecithricella minor 67.7121413.2 39.6 3
75 Table 4.2. Zooplankton abundance (ind m-2) in the vicinity of Marguerite Bay during austral fall 2001 and 2002. Geo mean = geometric mean, Range = minimum maximum, n = number of net hauls in which a taxon was present Fall 2001 Fall 2002 Taxonomic group Geo meanRangenGeo meanRangen Ostracods 77532.4 47281278883.6 251712 Euphausiids Euphausia crystallorophias 2.780.24 13.21018.51.93 13712 Euphausia superba 6.220.24 72.61114.81.63 11711 Thysanoessa macrura 45.97.19 2661216.24.27 51.912 Amphipods 15.115.63 91.51212.76.77 86.612 Mysids 1.730.23 34.9114.260.75 29.810 Medusae 13.30.27 52090.390.12 1.476 Siphonophores present12present11 Polychaetes 1360.65 11101011.90.48 27911 Pteropods 2542.33 2119122737.63 155611 Chaetognaths 57.03.05 6871231.34.23 16112 Appendicularians 4.460.78 32.331447147 56934 Salps 3.220.61 43.930.220.15 0.273
76 Figure 4.2a. Water column in tegrated abundance (ind m-2) of copepods from net hauls in the vicinity of Marguer ite Bay during austral fall 2001 (top) and 2002 (bottom). 0 10000 20000 30000 40000 50000C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d B o u r g e o i s F j o r d N e n y Fj o r d N A l e x a n d e r I s N A l e x a n d e r I s .N A l e x a n d e r I s N G e o r g e V I S o u n d S G e o r g e V I S o u n d S G e o r g e V I So u n d L a z a r e v B a yInd m-2 0 10000 20000 30000 40000 50000C r y s t a l S o u n d C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d S La u b e u f F j o r d S A d e l a i d e I s i n n e r M a r g u e ri t e B a y M a r g u e r i t e T r o u g h M a r g u e r i t e T r o u g h N A l e x a n d e r I s N A l e x a n d e r I s N A l e x a n d e r I s .Ind m-2 Calanoides acutus Calanus propinquus Ctenocalanus spp. Paraeuchaeta spp. Metridia gerlachei Oithona spp. Oncaea spp. Rhincalanus gigas Scolecithricella minor Gaidius tenuispinus Aetidopsis antarctica Metridia curticauda
77 Figure 4.2b. Percent composition of copepods at coastal stations in Marguerite Bay during austral fall 2001 (top) and 2002 (bo ttom). Color legend as in Figure 4.2a. 0% 20% 40% 60% 80% 100%C r y s t a l S o u n d N L a u b e u f F j o r d S La u b e u f F j o r d B o u r g e o i s Fj o r d N e n y Fj o r d N A l e x a n d e r I s N A l e x a n d e r I s N A l e x a n d e r I s N G e o r g e V I S o u n d S G e o r g e V I So u n d S G e o r g e V I So u n d La z a r e v B a y 0% 20% 40% 60% 80% 100%C r y s t a l S o u n d C r y s t a l S o u n d N L a u b e u f F j o r d S La u b e u f F j o r d S La u b e u f F j o r d S A d e l a i d e I s i n n e r M a r g u e r it e B a y M a r g u e r i t e T r o u g h M a r g u e r i t e T r o u g h N A l e x a n d e r I s N A l e x a n d e r I s N A l e x a n d e r I s .
78 The overall geometric mean abundance of total euphausiids was very similar during both years: 66.3 and 66.8 ind m-2 in 2001 and 2002, respectively (Mann Whitney U; p = 1.00) (Fig. 4.3a). Although abundances were variable within sampling locations and over the study area, distribut ion patterns were consistent between years. Northern areas, such as Laubeuf Fjord and Crystal S ound, consistently had th e highest integrated total euphausiid abundances. The southern sectors, including Neny Fjord, inner Marguerite Bay and near Al exander Island, had relatively lower euphausiid abundances during both years. George VI Sound and Lazarev Bay were only sampled during 2001 and had the lowest abundances of all areas surveyed. Despite the similarity in total eupha usiid abundance and distribution between years, there were significant interannual differences in sp ecies percent composition (Fig. 4.3b). T. macrura was the most abundant species during 2001 with a geometric mean abundance of 45.9 ind m-2, whereas E. crystallorophias had overall low abundances (2.78 ind m-2), and E. superba had intermediate values (6.22 ind m-2). During fall 2002, abundances were similar among species, with E. crystallorophias having the highest mean value (18.5 ind m-2), followed by T. macrura (16.2 ind m-2), and finally E. superba with the lowest abundance (14.8 ind m-2). Interannual differences in abundances of T. macrura (Mann Whitney U; p = 0.020) and E. crystallorophias (Mann Whitney U; p = 0.004) were significant, whereas values for E. superba were not significantly different between years (Mann Whitney U; p = 0.200).
79 Figure 4.3a. Water column integrated abundances of euphausiids (ind m-2) from net hauls in the vicinity of Marguerite Bay dur ing austral fall 2001 (top) and 2002 (bottom). 0 50 100 150 200 250 300C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d B o u r g e o i s F j o r d N e n y F j o r d N A l e x a n d e r I s N A l e x a n d e r I s. N A l e x a n d e r I s N G e o r g e V I S o u n d S G e o r g e V I So u n d S G e o r g e V I So u n d L a z a r e v B a yInd m-2 ES EC TM 0 50 100 150 200 250 300C r y s t a l S o u n d C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d S L a u b e u f F j o r d S A d e l a i d e I s i n n e r M a r g u er i t e B a y M a r g u e r i t e T r o u g h M a r g u e r i t e T r o u g h N A l e x a n d e r I s N A l e x a n d e r I s N A l e x a n d e r I s .Ind m-2
80 Figure 4.3b. Percent composition of euphausiids from net hauls in the vicinity of Marguerite Bay during austral fall 2001 (t op) and 2002 (bottom). Color legend as in Figure 4.3a. 0% 20% 40% 60% 80% 100%C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d B o u r g e o i s F j o r d N e n y Fj o r d N A l e x a n d e r I s N A l e x a n d e r I s .N A l e x a n d e r I s N G e o r g e V I S o u n d S G e o r g e V I So u n d S G e o r g e V I S o u n d L a z a r e v B a y 0% 20% 40% 60% 80% 100%C r y s t a l S o u n d C r y s t a l S o u n d N L a u b e u f F j o r d S La u b e u f F j o r d S La u b e u f F j o r d S A d e l a i d e I s i n n e r M a r g u e r it e B a y M a r g u e r i t e T r o u g h M a r g u e r i t e T r o u g h N A l e x a n d e r I s N A l e x a n d e r I s N A l e x a n d e r I s .
81 Even though abundances of E. superba were comparable between years, there were key interannual differences in developmental stage composition (Fig. 4.4). During 2001, non-larval E. superba were almost exclusively adult males and females and juveniles were essentially absent. Only a few year-two juveniles, 34 mm total length (TL), were observed at one sta tion in Laubeuf Fjord (0.08 ind m-3). Overall, juveniles accounted for only ~ 2% of the postlarval E. superba densities in 2001. In contrast, yearone juveniles were abundant and constituted ~ 41% of the populations in 2002, suggesting a successful recruitmen t from 2001 larvae. Postlarval T. macrura were mostly juveniles 8 10 mm TL during 2001 and 2002 (F ig. 4.5). Juveniles comprised 91% of individuals in fall 2001, and ~ 69% in 2002, when the remaining 31% were adults ranging primarily between 16 20 mm TL. During 2001, E. crystallorophias juveniles (10 18 mm TL) comprised 11% of the populat ion, while the remaining fraction (89%) included adults 20 36 mm TL (mode: 29 mm TL) (Fig. 4.6). In 2002, juveniles were a larger fraction of the population (28%), while the adults had a bimodal distribution, with a larger mode at 20 mm TL, and lower propor tions of larger indi viduals between 29 33 mm TL. Even though the percent composition of postlarval E. crystallorophias was comparable between years, larval abundan ces showed marked in terannual differences, with higher values in 2001 throughout the study area. Maximum abundances in 2001 were 829 ind m-2 in Neny Fjord, whereas the highest values recorded in 2002 were ~ 18 ind m-2 in northern Laubeuf Fjord (data not shown). In addition to copepods and euphausiids, other dominant zooplankton groups included ostracods, pteropods, polychaetes, ch aetognaths, appendicularians, amphipods, and mysids (Fig. 4.7a and b). Ostracods and pteropods were nu merically dominant
82 during both years and showed no significant interannual differences (Mann Whitney U test, p = 1.00 for ostracods, p = 0.805 for pt eropods). Appendicularians were more abundant during 2002 (geome tric mean = 1447 ind m-2) although their distribution was very patchy and they were pres ent in only four net hauls. During 2001, they were present at three locations, but in very low abundances. Polychaetes and chaetognaths were relatively abundant during 2001, but had lowe r densities in 2002. Abundances for amphipods and mysids were not significantly different between years (Mann Whitney U test, p = 0.32 for amphipods, p = 0.204 for my sids). Gelatinous zooplankton were generally rare. Siphonophores could not be numer ically quantified due to the presence of only fragments of colonies in the samples; however, fragments were classified as few or numerous and were frequent throughout th e water column at most locations during both years, particularly in the upper 300 m. Laubeuf Fjord had the highest abundan ces of these groups of zooplankton, followed by the areas in the vicinity of Alexander Island and Neny Fjord/inner Marguerite Bay (Sta. 6) duri ng both years (Fig. 4.7a). In 2001, abundances were lowest in Bourgeois Fjord, one station in southe rn George VI Sound, and Lazarev Bay, while during the following fall, densities were lowe st in Crystal Sound and south of Adelaide Island. In terms of percent compos ition, during 2001 ostracods, pteropods and polychaetes dominated at all stations except in Lazarev Ba y. At this location, these groups were scarce or absent, and the comm unity was composed primarily of medusae, chaetognaths, and salps (Fi g. 4.7b). Although ostracods dominated the zooplankton numerically at most locations in 2001, their percent contri bution varied among stations. Their contribution was maximum in Crysta l Sound, Laubeuf Fjord, and George VI
83 Sound, and minimum in the vicinity of Al exander Island and Lazarev Bay. In 2002, ostracods again dominated numerically at mo st locations, but were only a minor fraction of zooplankton in some net hauls in Laube uf Fjord, Adelaide Is land, and Marguerite Trough, where appendicularians were most important.
84 Figure 4.4. Length-frequency of E. superba juveniles and adults during fall 2001 (top) and 2002 (bottom). Data represent all coasta l net hauls for each y ear. Only juveniles 20 mm are included. Fall 2001 0 5 10 15 202224262830323436384042444648505254565860 Total length (mm)Percent (%) Adults Juveniles Fall 2002 0 5 10 202224262830323436384042444648505254565860 Total length (mm) Percent (%)
85 Figure 4.5. Length-frequency of T. macrura juveniles and adults during fall 2001 (top) and 2002 (bottom). Data represent all coastal net hauls for each year. Fall 2001 0 10 20 30 405678910111213141516171819202122232425262728293031323334Total length (mm)Percent (%) Adults Juveniles Fall 2002 0 10 20 30 4056789101112131415161718192021222324252627282930313233Total length (mm)Percent (%)
86 Figure 4.6. Length-frequency of E. crystallorophias juveniles and adu lts during fall 2001 (top) and 2002 (bottom). Data represent all coastal net hauls for each year. Fall 2001 0 5 10 15 20 25101112131415161718192021222324252627282930313233343536Total length (mm)Percent (%) Adults Juveniles Fall 2002 0 5 10 15 20 251011121314151617181920212223242526272829303132333435363738Total length (mm)Percent (%)
87 Figure 4.7a. Water column integrated abunda nces of zooplankton other than copepods and euphausiids (ind m-2) from net hauls in the vicinity of Marguerite Bay during austral fall 2001 (top) and 2002 (bottom). 0 2000 4000 6000 8000C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d B o u r g e o i s Fj o r d N e n y Fj o r d N A l e x a n d e r I s N A l e x a n d e r I s N. A l e x a n d e r I s N G e o r g e V I S o u n d S G e o r g e V I So u n d S G e o r g e V I S o u n d L a z a r e v B a yInd m-2 Amphipods Appendicularia Mysids Chaetognaths Medusae Polychaetes Thecate pteropods Salps Ostracods 0 2000 4000 6000 8000C r y s t a l S o u n d C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d S L a u b e u f F j o r d i n n e r M a r g u e r i t e B a y S A d e l a i d e I s M a r g u e r i t e T r o u g h M a r g u e r i t e T r o u g h N A l e x a n d e r I s N A l e x a n d e r I s N A l e x a n d e r I s .Ind m-2
88 Figure 4.7b. Percent composition of zooplankt on other than copepods and euphausiids from net hauls in the vicinity of Margue rite Bay during austral fall 2001 (top) and 2002 (bottom). Color legend as in Figure 4.5a. 0% 20% 40% 60% 80% 100%C r y s t a l S o u n d N L a u b e u f F j o r d S L a u b e u f F j o r d B o u r g e o i s F j o r d N e n y F j o r d N A l e x a n d e r I s N A l e x a n d e r I s. N A l e x a n d e r I s N G e o r g e V I S o u n d S G e o r g e V I S o u n d S G e o r g e V I So u n d La z a r e v B a y 0% 20% 40% 60% 80% 100%C r y s t a l S o u n d C r y s t a l S o u n d N L a u b e u f F j o r d S La u b e u f F j o r d S La u b e u f F j o r d i n n e r M a r g u e r i t e B a y S A d e l a i d e I s M a r g u e r i t e T r o u g h M a r g u e r i t e T r o u g h N A l e x a n d e r I s N A l e x a n d e r I s N A l e x a n d e r I s .
89 Vertical Distribution Overall, the mean depth of maximum abundance for all copepods was significantly deeper in the water column during May June 2001, than in April May 2002 (2001: 252 m; 2002: 75 m; Students tTest; p < 0.001) (Fig. 4.8). Looking at individual species, only depths of Paraeuchaeta spp., Oithona spp., and R. gigas were significantly different between years (ANOVA, p < 0.05), alt hough other species such as M. gerlachei Oncaea spp., and S. minor also had somewhat sha llower distributions in 2002. Copepod species occurred within th ree depth ranges: shallow (0 100 m), intermediate (100 150 m), and deep (> 150 m) water groups (Table 4.3). In 2001, shallow species included Ctenocalanus spp. and C. propinquus while Oithona spp., S. minor, Paraeuchaeta spp. M. gerlachei, C. acutus, G. tenuispinus and Oncaea spp. were identified as deep species. Only R. gigas was intermediate during the first year. In contrast, Oithona spp., R. gigas and Paraeuchaeta spp. were grouped as shallow species during 2002, while M. gerlachei and S. minor were intermediate. Oncaea spp., C. acutus and G. tenuispinus remained deep species during fall 2002. The vertical distributions of euphausiid s were similar within and between years (Table 4.4) and geometric mean depths of maximum abundance for all species were in the upper 100 m (65 84 m). Mysids were distributed the deepest during 2001 (249 m) and 2002 (224 m), while amphipods were located at intermediate depths in 2001 (125 m) and represented the shallowest group duri ng 2002 (55 m). The depths of maximum abundance for E. superba, E. crystallorophias and T. macrura were not significantly different within or between years (Kru skall-Wallis ANOVA; p = 0.882 in 2001, and p = 0.873 in 2002). However, even though their general depth ranges overlapped throughout
90 Table 4.3. Mean depth of maximum abundance (Z, m) of cope pods in 2001 and 2002. Species in bold indicate those that were classified in the same depth cate gory during both years. The range represents the shallowest and deepest depth (m) at which each species was present. Depth categories were based on the general hydrography of the area. At most net hauls, the upper mixed layer ranged between 0 ~100 m, with a thermoclin e/pycnocline generally between ~100 150 m, and warmer saltier water below ~150 m Fall 2001 Shallow Z Range Intermediate Z Range Deep Z Range Ctenocalanus spp. 51 0 500 R. gigas 147 0 800 Oithona spp. 162 0 525 C. propinquus 80 0 525 S. minor 184 0 350 Paraeuchaeta spp.201 0 800 M. gerlachei 213 0 800 C. acutus 213 0 800 G. tenuispinus 249 0 800 Oncaea spp. 348 0 800 Fall 2002 Shallow Z Range Intermediate Z Range Deep Z Range Ctenocalanus spp. 39 0 500 M. gerlachei 142 0 500 C. acutus 226 0 500 Oithona spp. 63 0 500 S. minor 149 100 200 G. tenuispinus 230 0 500 C. propinquus 71 0 460 Oncaea spp. 243 0 500 R. gigas 77 0 200 Paraeuchaeta spp. 100 0 500
91 Figure 4.8. Mean depth of maximum abundan ce (m) of copepods during 2001 (black) and 2002 (grey). Values represent the arithmetic mean of the depth of maximum abundance of a species for all net haul s. Vertical bars represent 1 standard deviation. Dotted horizontal lines repr esent the limits of the 0 100, 100 150, and > 150 m depth ranges. the study area, the depths of maximum abundan ce of species usually did not overlap at any given location (Fig. 4.8). For exampl e in Crystal Sound during 2001, the maximum abundances of E. superba occurred at 50 m, while T. macrura primarily occurred at 100 m (Fig. 4.9a). In 2002, the depth of maximum abundance for E. superba was 150 m, E. crystallorophias was primarily found at 50 m, and T. macrura was at 75 m. In Laubeuf Fjord, E. superba and T. macrura had peaks at 100 m in 2001, while E. crystallorophias was observed at 50 m, and a second mode for T. macrura occurred at 400 500 m (Fig. 4.9b). In 2002, E. superba had peaks of maximum abundance at 75 m and 150 m, while C a c u t u s C p r o p i n q u u s C t e n o c a l a n u s s p p P a r a e u c h a e t a s p p G t e n u i s p i n u s M g e r l a c h e i O i t h o n a s p p O n c a e a s p p R g i g a s S m i n o r Depth (m) 0 100 200 300 400 500 600 2001 2002
92 E. crystallorophias occurred primarily at 100 m. In the vicinity of Alexander Island during fall 2001, E. superba could be found mostly at 50 m, while T. macrura was mainly at 100 m (Fig. 4.9c). Finally, south of Adelaide Island, elevated abundances of E. superba were located at shallow depths in 2002 (0 100 m, maximum at 50 m), while T. macrura showed a smaller peak at 150 m, and E. crystallorophias was not present in significant numbers. The geometric mean depths of maximu m abundances of all euphausiid species ranged between 65 73 m in 2001, while the majo rity of copepods were located at 252 m during the same year. Looking at indi vidual net hauls for 2001, the maximum abundances of both groups did not overlap at almost any location, with the bulk of the copepod community located consistently de eper than the mode of the euphausiids (Appendix 1). On the other hand, during 2002, euphausiids were observed at similar depths as the previous fall, with maximu m abundances between 70 84 m (Table 4.4), while copepods were significantly shallowe r than in 2001, mostly at 75 m (Appendix 2). Table 4.4. Mean depth of maximum abundan ce (m) of euphausiids, amphipods, and mysids during fall 2001 and 2002. Data include all coastal stations for each cruise. Values represent the geometric mean dept h of maximum abundance of each group for all net hauls during each year (n = 12). The range (m) represents the range of depths in the water column that were occupied by each taxon Fall 2001 Fall 2002 Depth (m) Range Depth (m) Range E. crystallorophias 65 0 500 70 0 500 E. superba 73 0 500 78 0 500 T. macrura 68 0 800 84 0 500 Amphipods 125 0 800 55 0 500 Mysids 249 0 500 224 0 500
93 Abundance (ind m -3 ) 0.00.51.01.52.0Depth (m) 0 100 200 300 400 500 E. crystallorophias E. superba T. macrura Amphipods Mysids Abundance (ind m-3) 0.00.51.01.52.02.53.0Depth (m) 0 100 200 300 400 500 Figure 4.9a. Vertical distri bution of euphausiids, amphipods, and mysids (ind m-3) in Crystal Sound during austral fa ll 2001 (top) and 2002 (bottom).
94 Abundance (ind m -3 ) 0.00.51.01.52.0Depth (m) 0 100 200 300 400 500 E. crystallorophias E. superba T. macrura Amphipods Mysids Abundance (ind m -3 ) 0.00.10.20.30.40.5Depth (m) 0 100 200 300 400 500 Figure 4.9b. Vertical distri bution of euphausiids, amphipods, and mysids (ind m-3) in Laubeuf Fjord during austral fa ll 2001 (top) and 2002 (bottom).
95 Abundance (ind m -3 ) 0.00.20.40.60.81.01.21.4Depth (m) 0 100 200 300 400 500 E. crystallorophias E. superba T. macrura Amphipods Mysids Abundance (ind m-3) 0.00.20.40.60.81.0Depth (m) 0 100 200 300 400 500 Figure 4.9c. Vertical distri bution of euphausiids, amphipods, and mysids (ind m-3) in the vicinity of Alexander Island dur ing austral fall 2001 (top) and s outh of Adelaide Island in fall 2002 (bottom).
96 Horizontal Distribution The horizontal distribution of zooplankt on varied within the study area. In general, in areas where euphausiids, am phipods, and mysids (i.e. macrozooplankton) were abundant, copepod densities were low, and vice versa (Fig. 4.10a). Macrozooplankton were most abundant in nort hern sectors, whereas copepod abundances were highest in southern sectors, such as i nner Marguerite Bay, the vicinity of Alexander Island, George VI Sound and Lazarev Bay (Fi g. 4.10b). Exceptions to this pattern were observed in fall 2001 in northern Laubeuf Fjord, where copepods and macrozooplankton were highly abundant, in Bourge ois Fjord, where all zooplank ton were scarce, and south of Adelaide Island in 2002, where copepod a nd macrozooplankton abundances were low. Figure. 4.10a. Linear correlation be tween integrated abundances (ind m-2) of total macrozooplankton (euphausids, amphipods, an d mysids) and copepods at different stations during fall 2001 and 2002 (n = 12; r = -0.39; p = 0.208). When multiple net hauls were sampled, values represent the mean. Plot includes data for stati ons in Crystal Sound (Sta. 7), Laubeuf and Bourgeois fjords (Sta 5), Neny Fjord/inner Marguerite Bay (Sta. 6), Marguerite Trough (Sta. 2), S. Adelaide Isla nd (Avian Is.), Alexander Island (Sta. 4a), George VI Sound (Sta. 4b), and Lazarev Ba y (Sta. 3). Data for Laubeuf Fjord in 2001 were excluded (outlier). Macrozooplankton 101102103 Copepods 103104105
97 Figure 4.10b. Mean water column integrated abundance (ind m-2) of macrozooplankton (euphausiids, amphipods, and mysids) (grey bars, left axis) and copepods (black circles, right axis) at different st ations within Marguerite Bay during fall 2001 (top) and 2002 (bottom). 0 10000 20000 30000 40000 50000 0 50 100 150 200 250 300 350 Macrozooplankton Copepods 0 10000 20000 30000 0 50 100 150 200 250 300 350
98 Fall Environmental Parameters The environmental parameters investigat ed here were chosen based on their potential influence on zoopl ankton distributions: verti cally integrated pigment concentrations (chlorophy ll + phaeopigment, mg m-2), surface salinity, and bottom depth. Integrated pigment represents food availabilit y, while surface salinity is an indicator of sea ice formation and melting, and the presen ce of glacial meltwater nearshore, which affects the type of phytoplankton present in the water column. Relationships between zooplankton and bottom depth potentially indi cate concentration of organisms nearshore or in association with Marguerite Trough, or deep shelf depressions. Fall integrated pigment concentrations in Margueri te Bay ranged between 9.6 33 mg m-2 in 2001, and 72.4 236 mg m-2 in 2002. Maximum pigment concentrations were located in Laubeuf Fjord during both years, while lowest values were observed in the in George VI Sound during fall 2001, in Crystal Sound during fall 2002, as well as in the vicinity of Alexander Island during both years. Even thoug h pigment concentrations were lowest in Crystal Sound during 2002, values were more than double those observed in the same area during the previous year. Surface salin ity values were similar between years and ranged between 33.04 and 33.55. The shallowest stations had bottom depths of ~ 300 400 m and were located near the coast of Alexander Island, the northern edge of Marguerite Trough in central Marguerite Bay, southern Adelaide Island, and inner Marguerite Bay, while the deepest areas, with bottom depths > 800 m, were in northern George VI Sound, Lazarev Bay, and a location in the vicinity of Alexander Island. Overall, there were no clear associations between the horizont al distribution of zooplankton (ind m-2) and concurrent environmental va riables, including vertically
99 integrated pigment concentrations (chlorophyll + phaeopigment, mg m-2), surface salinity, and bottom depth (Table 4.5). In general, only a few copepods showed a significant correlation with enviro nmental variables. During 2001, Ctenocalanus spp. showed a positive correlation with pigment concentrations, Oithona spp. showed a negative relationship, and G. tenuispinus was positively correlated with bottom depth. During fall 2002, Paraeuchaeta spp. showed a positive correlation with pigment concentrations, while C. acutus and Oncaea spp. were positively correlated with bottom depth. Only two other zooplankton taxa show ed significant relationships with pigment concentrations: T. macrura in 2001 and mysids in 2002. Only two groups, E. superba and the copepod R. gigas showed significant relationships with surface salinity, and these were only observed during 2002. Of th e 138 relationships examined between zooplankton abundance and environmental variab les, 11 were significant, which is only slightly higher than the nu mber that would be expected with 95% confidence if all relationships were statistically insignificant (i.e., seven). At many locations there was a negative tre nd between the vertical distribution of macrozooplankton (i.e., euphausiids, amphi pods, and mysids) and gradients in temperature, salinity, and density (examples in Appendix 3). E. superba was located either above or below the thermocline/pycnoc line in 92% of the net hauls in which the species was present, while corresponding values for E. crystallorophias and T. macrura were 77% and 60% respectively (Table 4.6). However, euphasu iids coincided with physical gradients at 8 23% of the locations sampled (examples in Appendix 4). Most amphipods and mysids were located primarily deeper than the thermocline/pycnocline, with only smaller fractions present at the dept h of the temperature/density gradient. In
100 general, no patterns were observed in the ver tical distribution of zooplankton in relation to fall pigment concentrations. Table 4.5. Spearman rank order correlati ons between integrated abundance of zooplankton (ind m-2) and vertically integrated pi gment concentrations (chlorophyll + phaeopigment; mg m-2), salinity at 10 m (S10), and bottom depth (bottom Z). Significant correlations (p < 0.05) are in bold Fall 2001 Fall 2002 Taxonomic group PigmentS10 Z PigmentS10 Z Copepods C. acutus 0.415-0.021-0.1470.262-0.161 0.630 C. propinquus 0.0850.385-0.4030.5820.301 -0.144 Ctenocalanus spp. 0.799 -0.0630.0280.0000.210 0.256 G. tenuispinus -0.2570.211 0.641 0.077-0.500 0.004 M. gerlachei 0.427-0.126-0.0770.2110.049 0.441 Oithona spp. -0.701 0.2660.1820.1430.189 0.014 Oncaea spp. -0.463-0.1610.2800.152-0.378 0.635 Paraeuchaeta spp. -0.0180.154-0.336 0.743 -0.223 0.217 R. gigas -0.3000.413-0.0390.536 0.609 0.004 S. minor -0.1730.295-0.033-0.0810.037 0.221 TOTAL copepods 0.543-0.049-0.1470.3970.203 0.497 Ostracods 0.341-0.126-0.0490.379-0.132 0.382 Euphausiids E. crystallorophias -0.355-0.4870.2490.110 -0.657 -0.060 E. superba 0.220-0.273-0.4270.110-0.582 -0.109 T. macrura 0.707 -0.133-0.287-0.6330.217 0.193 Amphipods 0.628-0.035-0.5170.008-0.356 -0.199 Mysids 0.567-0.028-0.028 0.802 -0.347 -0.091 Medusae -0.3000.0990.2180.206-0.250 0.411 Polychaetes 0.122-0.5530.0390.649-0.189 0.466 Pteropods -0.3110.434-0.3010.253 0.580 0.311 Chaetognaths -0.2800.203-0.028-0.0840.308 -0.045 Appendicularians -0.1650.340-0.1190.5560.104 -0.025 Salps 0.0090.431-0.211-0.055-0.257 0.354
101 Table 4.6. Percentage of net hauls in whic h macrozooplankton were located primarily shallower, deeper, or at the same depth as the thermocline/pycnoclin e. Percentage of net hauls with widespread vertical di stributions are also indicated. n is the total number of net hauls in which a taxon was present during 2001 and 2002 combined. A total of seven net hauls were excluded from the cal culations due to a vertically uniform water column (four net hauls in 2001) or lack of concurre nt CTD data (three net hauls in 2002) Shallower (%) Deeper (%) Same depth (%) Widespread (%) n Euphausiids E. crystallorophias 54 23 23 0 13 E. superba 58 33 8 0 12 T. macrura 33 27 13 27 15 Amphipods 19 56 6 19 16 Mysids 0 89 11 0 9 Summer Chlorophyll Concentrat ions and Krill Recruitment Surface chlorophyll concentrati ons in the Bellingshausen Sea and Marguerite Bay during the spring and summer influenced the abundance and composition of the zooplankton during fall (Fig. 4.11a). In ocean ic waters of the Bellingshausen Sea (Fig. 4.11b and c), chlorophyll concentrations we re above average during austral springsummer 2000/2001 (November January). As the summer progressed, phytoplankton blooms moved onshore and above average chlo rophyll concentrations were observed in the coastal Bellingshausen Sea during Janua ry March 2001 (Fig. 4.11d). During the 2001/2002 season, conditions offshore were av erage between November and January (Fig. 4.11b and c), while in coastal waters, a bove normal chlorophyll concentrations were primarily observed by late February (Fig. 4.11d) Within Marguerite Bay, the interannual differences were even more striking, with extremely high chlor ophyll concentrations during austral summer fall 2001 and below av erage conditions in 2002 (Fig. 4.11e f).
102 There was a strong correla tion (Spearman R = 0.81, p < 0.05) between geometric mean chlorophyll concentrations in the Bellingshausen Sea during November and summer recruitment indices for E. superba previously reported for the western Antarctic Peninsula region (Fig. 4.12a). Geometric m ean chlorophyll concentrations for November 1997 2004 in oceanic waters of the Bellingshausen Sea were highest during the spring summer 2000/2001 season (n = 8), coincident with the highest recruitment index (R1) observed for E. superba during all years for which da ta are available (1997/1998 2002/2003) (Fig. 4.12b). Relatively low chlorop hyll concentrations were registered during November 1997, 1998, and 2002, when lo w recruitment values also were recorded.
103 Figure 4.11. (a) Location of the subregions analyzed for median SeaWiFS chlorophyll concentrations (chl, mg m-3) in (b, c) oceanic and (d) coas tal waters of the Bellingshausen Sea, (e) northern, and (f) southern Mar guerite Bay in spring /summer 2000/2001 (grey) and 2001/2002 (black). A climatology fo r spring/summer 1997/1998 2003/2004 is represented by the red line and corresponds to the median chlorophy ll in each subregion for the seven seasons analyzed. Region 1 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35S e p t e m b e r O c t o b e r N o v e m b e r D e c e m b e r J a n u a r y F e b r u a r y M a r c h A p r i lmedian chl (mg m-3) 2000/2001 2001/2002 climatology Region 2 0.00 0.20 0.40 0.60 0.80S e p t e m b e r O c t o b e r N o v e m b e r D e c e m b e r J a n u a r y F e b r u a r y M a r c h A p r i lmedian chl (mg m-3) 2000/2001 2001/2002 climatology(a) (b) (c)
104 Figure 4.11. (Continued) Region 3 0.00 0.30 0.60 0.90 1.20 1.50 1.80S e p t e m b e r O c t o b e r N o v e m b e r D e c e m b e r J a n u a r y F e b r u a r y M a r c h A p r i lmedian chl (mg m-3) 2000/2001 2001/2002 climatology Region 4 0 1 2 3 4 5 6S e p t e m b e r O c t o b e r N o v e m b e r D e c e m b e r J a n u a r y F e b r u a r y M a r c h A p r i lmedian chl (mg m-3) 2000/2001 2001/2002 climatology Region 5 0 1 2 3 4S e p t e m b e r O c t o b e r N o v e m b e r D e c e m b e r J a n u a r y F e b r u a r y M a r c h A p r i lmedian chl (mg m-3) 2000/2001 2001/2002 climatology(e) (f) (d)
105 Figure 4.12. (a) Spearman R correlation be tween geometric mean chlorophyll concentrations (Chl, mg m-3) in oceanic waters of the Bellingshausen Sea during November 1997 2004 and recruitment of E. superba (R1) in waters west of the the Antarctic Peninsula, for the period 1997/ 1998 2002/2003; n = 6; R = 0.81; p < 0.05). (b) Time series of geometric mean chlorophyll concentration in ocean ic waters of the Bellingshausen Sea during November 1997 2004 (grey bars) and recruitment of E. superba (R1) (black circles) in waters west of Antarctic Peninsula. Chlorophyll estimated as geometric mean SeaWiFS chlorophyll concentration during late November in subregion 2 of Fig. 4.11a. Recruitment indices (R1) for 1997/1998 1999/2000 from Siegel et al. (2002); for 2000/2001 from Siegel et al. (2003); and for 2001/2002 and 2002/2003 from the Palmer LTER DataZoo databa se (data provided by L. Quetin and R. Ross). Chlorophyll 0.00.20.40.60.81.0 R1 0.0 0.2 0.4 0.6 0.8 1.0 0.0 0.4 0.81 9 9 7 / 1 9 9 8 1 9 9 8 / 1 9 9 9 1 9 9 9 / 2 0 0 0 2 0 0 0 / 2 0 0 1 2 0 0 1 / 2 0 0 2 2 0 0 2 / 2 0 0 3 2 0 0 3 / 2 0 0 4 2 0 0 4 / 2 0 0 5Ch l 0.0 0.2 0.4 0.6 0.8 1.0R1
106 DISCUSSION Composition and Abundance of Zooplankton in Marguerite Bay Total zooplankton abundances in the WAP, and Marguerite Bay in particular, are generally higher than those re ported for other areas of the Southern Ocean. Deibel and Daly (2007) summarized zoopl ankton abundance data for continental shelf regions around Antarctica and report that values in Marguerite Bay and Crocker Passage are one or two orders of magnitude higher than in any other area considered, including the Weddell and Ross seas. Copepods numerically dominated the zooplankton community of Marguerite Bay during fall, with a mean abundance of 24,362 ind m-2 in 2001, and values up to 44,135 ind m-2 in Laubeuf Fjord. Also during SO GLOBEC, Ashjian et al. (2008) investigated zooplankton on the outer contin ental shelf of Margue rite Bay using Video Plankton Recorder (VPR) data, and estimated copepod mean integrated abundances of 2,832 2,983 ind m-2 for fall 2001, only ~ 10% of the estimates reported here for coastal waters in Marguerite Bay. In comparison, ne ar South Georgia in the Scotia Sea, total copepod abundances during summer 1994 1996 were up to an order of magnitude higher than those calculated here for fall, with median values ranging between 66,684 and 235,793 ind m-2 for the upper 200 m of the water column (Atkinson et al., 1999). Copepod species composition in Margueri te Bay was comparable to that described for other Antarctic areas, including wa ters west of the Antarctic Peninsula (e.g., Hopkins, 1985; Schnack-Schiel and Mujica, 1994), the Bell ingshausen Sea (Atkinson, 1995; Siegel and Harm, 1996), and western Weddell Sea (Hopkins and Torres, 1988). However, during SO GLOBEC, C. acutus M. gerlachei Ctenocalanus spp., and C.
107 propinquus dominated total copepod abundance and, the smaller Oncaea spp. and Oithona spp. did not represent a major fraction of the community, in c ontrast to previous results. A study in Crocker Passage, on the nor thern Antarctic Peninsula shelf, reported a similar fall copepod species co mposition, but observed that Oncaea curvata comprised over half of the total abu ndance (Hopkins, 1985). Other results from the WAP also indicate that these smaller copepods generally outnumber la rger species (Schnack-Schiel and Mujica, 1994; Cabal et al., 2002 ). In contrast to the rela tively finer mesh used during these previous studies (162 200 m), zooplankton were sampled with 333 m mesh during SO GLOBEC; thus, it is possible that abundances of smalle r cyclopods, including Oncaea spp. (0.6 1.1 mm in length) and Oithona spp. (0.7 1.2 mm), were underestimated. T. macrura was the most abundant euphausiid in 2001, with integrated abundances up to 266 ind m-2. These values are higher than maximum integrated abundances of 78 ind m-2 reported during spring 1989 for Gerl ache Strait, at the northern end of the WAP (Nordhausen, 1994). Mean abundances of T. macrura during fall in Marguerite Bay (2001: 45.9 ind m-2; 2002: 16.2 ind m-2) also were higher than average spring values in the Weddell Sea, wh ich ranged between 0.75 and 2.98 ind m-2 in the upper 200 m (Donnelly et al., 2006). E. superba is often a major fraction of the macrozooplankton community, comprising up to 95% of the larger zooplankton in waters along the northern WAP (Lancraft et al., 2004) This species was observed at almost every location surveyed during the present study; however, abundances did not exceed those of other euphausiids and macrozoopla nkton groups, particularly during 2001. The distribution of E. superba is patchy, with post larval indi viduals usually aggregated in
108 compact and dense swarms that can often be undersampled by plankton nets (Wiebe et al., 2004); thus, the relatively low abundances reported here should be interpreted with caution. Densities of E. superba estimated from acoustic surveys during the fall survey cruises in Laubeuf Fjord and Crystal Sound we re approximately an order of magnitude higher than those from net samples in the same area, and the difference was attributed, in part, to krill net avoidanc e (Lawson et al., 2008). T. macrura forms more diffuse aggregations and is more evenly distributed throughout the water column, which allows for more accurate abundance estimates fr om net data. Despite being likely undersampled, E. superba still comprised a major fraction of the fall zooplankton biomass in Marguerite Bay (see estimates be low). Mean integrat ed abundances of juvenile and adult E. superba in Marguerite Bay were 6.22 and 14.8 ind m-2 in 2001 and 2002, respectively, with maximum values up to 117 ind m-2 in Crystal Sound during 2002. These abundances are in the same order of magnitude as values reported for other areas of the WAP. In a review of da ta for Elephant Island between 1977 and 2004, Siegel (2005) reports abundances ranging between 1.4 and 336 ind m-2 (median for all years = 9.1 ind m-2), while Lancraft et al. (2004) es timated fall abundances of 788 ind m-2 for the upper 200 m of the water column in Crocker Passage. Although T. macrura may at times outnumber E. superba the latter is generally more important in terms of biomass. For the present study, biomass data are not available; however, some simple calcula tions can provide useful estimates of interspecific biomass differences. Duri ng 2001, the overall geometric mean abundance of T. macrura in Marguerite Bay was 45.9 ind m-2, with most individuals at 9 10 mm TL (Fig. 4.5). In the case of E. superba the average abundance was 6.22 ind m-2 during
109 the same year, with a mode at 51 mm TL (Fig. 4.4). The mean biomass for T. macrura in 2001 was 0.68 g WW m-2, while the corresponding value for E. superba was 9.11 g WW m-2, based on the length frequency distribu tions for both species during fall 2001 and a length-wet weight relationshi p from Ashjian et al. ( 2004) (wet weight = 0.0054 x Length3.214). These biomass estimates are with in the range of values reported for euphausiids in other Antarctic areas. In the western Antarctic Peninsula shelf region, mean fall biomass of krill estimated from acoustic data was 12 g m-2 (Lascara et al., 1999), whereas summer estimates for krill ar ound Elephant Island in 1978 2004 ranged between 0.76 and 75.2 g m-2 (Siegel, 2005). In addition, Voronina (1998) summarized published biomass estimates for E. superba from plankton nets in different areas of the northern WAP and reported values gene rally ranging between 0.2 and 54 g WW m-2, although biomass at Crocker Passage during fall 1983 reached 229 g WW m-2. In the Weddell Sea, average spring biomass for E. superba ranged from 0.54 g WW m-2 in open waters, to 1.2 g WW m-2 in the vicinity of the ice e dge, while corresponding values for T. macrura were 0.57 and 1.4 g WW m-2, respectively (Donnelly et al., 2006). Despite the rapid population response to el evated food concentrations and numerical dominance of T. macrura during fall 2001, E. superba was still the dominant euphausiid in terms of biomass. Other numerically important taxa obser ved in Marguerite Bay during fall 2001 and 2002, such as ostracods, pteropods, pol ychaetes, and chaetognaths, are common members of the zooplankton of the WAP (Schnack-Schiel a nd Mujica, 1994; Siegel and Harm, 1996). Pteropods and ostracods c onstituted the most abundant non-copepod zooplankton in the study area during both years analyzed. Pteropod abundances recorded
110 during this study did not vary interannually, and are similar to other SO GLOBEC values reported for the outer shelf off Marg uerite Bay, which reached ~ 1200 ind m-2 in the vicinity of Marguerite Tr ough (Ashjian et al., 2008). Small pteropods (< 15 mm), which dominated numerically in Marguerite Bay with values up to 2,119 ind m-2, are mostly herbivores (Hopkins, 1985). Ostrac ods abundances reached 4,728 ind m-2 in southern Laubeuf Fjord during 2001, which comprised ~ 76% of the non-copepod zooplankton at this location. Ostracods are generally om nivores and feed primarily on debris of E. superba phytoplankton, and copepods (Hopkins 1985). To the authors knowledge, other ostracod integrated abundances have not been reported for the Marguerite Bay area, and VPR results indicated that they were not a dominant component of the zooplankton of the mid and outer shelf of Marguerite Bay (C. Ashjian, pers. comm. ). A study in the Weddell Sea observed that ostracods comprise d 5.4% of total zooplankton biomass under the ice, and 2.6% in open wate rs (Hopkins and Torres, 1988). Chaetognaths were more abundant in Mar guerite Bay than in other Antarctic areas, with mean integrated abundances of 57 and 31.3 ind m-2 in 2001 and 2002 respectively, although va lues reached 687 ind m-2 in the vicinity of Alexander Island during 2001. Donnelly et al. (2006) report mean values of 1.75 ind m-2 for the Weddell Sea, while Lancraft et al. (2004) estimate d mean integrated abundances of 6.37 ind m-2 for Crocker Passage during fall. Duri ng SO GLOBEC, chaetognaths comprised, on average, 8% of the non-copepod zooplankt on during 2001, with values up to 42% in Lazarev Bay, and 23% around Alexander Island. The mean percent composition reported here agrees well with values from th e Weddell Sea ranging between 3.8 and 29.3% (Hopkins and Torres, 1988; Donnelly et al., 200 6). Chaetognaths are predators and feed
111 mostly on copepods, although the larger specie s also have been observed to prey on mysids and amphipods (Hopkins, 1985). Thus, high abundances of chaetognaths in copepod-rich waters near Alexa nder Island are not surprising. Polychaetes were more abundant in fall 2001 relative to 2002. This group includes mostly herbivorous species in An tarctic waters (Hopkins, 1985) and, thus, higher densities in fall 2001, wh en chlorophyll concentrations were higher, are expected. Vertically integrated abundances reported fo r the outer shelf off Marguerite Bay reached ~ 2,300 ind m-2 west of Alexander Island in 2001 (A shjian et al., 2008), more than double the maximum values of 1,110 ind m-2 estimated here in coastal Marguerite Bay during the same year. Variability in Euphausiid Life History Strategies Variability in life history strategies between euphasuiid species resulted in interannual differences in species percent composition. During the chlorophyll-rich 2001 season, the ubiquitous T. macrura dominated the euphausiid community, followed by E. superba and E. crystallorophias The numerical dominance of T. macrura has been previously reported as well for other areas of the Southern Ocean (e.g., Atkinson and Peck, 1988; Mujica, 1989). Ther e exist marked differences in the timing of the onset of reproduction between species of Antarctic euphausiids. T. macrura is the first species to start reproducing, with spawning as early as September (Makarov, 1979). Later in spring, E. crystallorophias reproduction begins, followed by E. superba which spawns mainly between November and March (Marr, 1962). Most individuals of T. macrura observed during fall 2001 were juveniles between
112 8 11 mm TL (87% total) (Fig. 4.5). T. macrura has a more rapid development than the Euphausia species. Although information on larv al growth rates is available for T. macrura (Nordhasuen, 1992, Siegel, 1987), juvenile development has not been described. Nordhausen (1992) reports that it takes 90 days for calyptopis II (C2) to develop into the last larval stage, furcilia VI (F6). In addition, Makarov (1979 ) estimates that it takes 15 20 days for eggs to change into C2, indicat ing a total larval development time of 105 110 days. Siegel (1987) estim ated age and growth of T. macrura in the Weddell Sea from length-frequency data and suggested that larvae can first develop into juveniles at 8 mm in length, during the second half of their first year of life. If reproduction of T. macrura along the western Antarctic Peninsula star ts in September October, individuals observed during fall could be up to seven or eight months old. Development estimates suggest that the T. macrura juveniles observed during fa ll 2001 originated from a reproductive event(s) during spring-summer 2000/2001 (age-class 0+), when chlorophyll concentrations were above climatology values. Development of E. superba and E. crystallorophias is slower and, therefore, populations take longer to res pond to environmental changes. In contrast to the 105 110 days estimated for larval development of T. macrura laboratory experiments indicate that E. superba can develop from an egg to F6 in approximately 127 days (Ikeda, 1984), while the larval development time for E. crystallorophias is even longer (Ikeda, 1986; Brinton and Townsend, 1991). Th is longer development coupled with a later onset of reproduction in November Marc h suggest that juveniles of E. superba and E. crystallorophias will not be present in the water column of the western Antarctic Peninsula region until the spri ng of their second year (age class 1+), which is supported
113 by field observations (Daly 2004; Daly and Zimmerman, 2004). During SO GLOBEC, elevated chlorophyll concentrations in 2001 supported a successf ul reproduction of E. superba and E. crystallorophias during spring-summer 2000/2001, as evidenced by the presence of numerous larvae in fall 2001 and juveniles during 2002. Hence, T. macrura had a rapid population response to elevated chlorophyll concentrati ons, demonstrated by the large numbers of juveniles pres ent in fall 2001 (Fig. 4.5), whereas E. superba and E. crystallorophias showed a slower popul ation response, supporte d by the scarcity of juveniles of either species during fall 2001 and the high proportion of juveniles present during fall 2002 (Figs 4.4 and 4.6). E. superba females are believed to spawn offshore in the vicinity of the shelfbreak (Marr, 1962; Siegel, 1988; 1992). Because eggs are denser than seawater, they sink and hatch at depths of 800 1000 m. The young larvae then swim to the surface before turning into the first feeding stage, calyptopis I (Mar r, 1962), at which point they need to find food within approximately 10 da ys or otherwise will not survive (Ross and Quetin, 1986). This life strategy is hypothe sized to have several ecological advantages, including preventing the eggs from reaching th e seafloor and becoming unviable or eaten by benthic organisms. In addition, the larv ae produced at these depths will develop in warmer (> 1 C) Circumpolar Deep Water (CDW) found below 500 m, thus reducing the risk of predation by epiand mesopelagic fa una. The developmental ascent described by Marr (1962) for E. superba also has been reported for larvae of T. macrura (Makarov, 1979). These two species share several ecol ogical characteristics, such as a common widespread distribution, a sim ilar reproductive strategy involv ing deep hatching of eggs and larval ascent, and a primarily herbivorous larval phase. Their similar life strategies
114 would indicate that these euphausiids ar e strong competitors; however, it has been suggested that the two month separation between the onset of reproduction is an adaptation to avoid competition for food between their larvae (Makarov, 1979). In addition, during the postlarval stag es, when both species coexist, E. superba are mainly herbivores during spring and summer, while adult T. macrura have been described as omnivores, feeding mostly on large copepods, such as C. acutus and M. gerlachei (Hopkins, 1985). Additionally, their depths of maximum abundances in the water column generally do not ove rlap. In summary, although E. superba and T. macrura share a common distribution and similar life history strategies, they have developed individual adaptations to minimize competition between them, allowing them to coexist as widespread, successful, Antarctic species. Summer Chlorophyll and Zooplan kton Population Response Interannual variability in chlorophyll concentrations strongly influences zooplankton populations. Population responses for T. macrura E. crystallorophias and E. superba were described above. In addi tion, the influence of spring/summer chlorophyll concentrations is demonstrated by the strong correlat ion observed between November chlorophyll concentrations in th e Bellingshausen Sea and recruitment of E. superba along the WAP during the following year. Previous studies re ported correlations between sea ice extent du ring winter and successful juvenile recruitment of E. superba during the following spring (Kawaguchi and Sa take, 1994; Siegel and Loeb, 1995; Hewitt et al., 2003). Sea ice biota on the undersurf ace of sea ice constitutes an alternative food source for overwintering larval krill, while s ea ice provides refuge from predators, which
115 further reduces winter larval mortality (D aly 1990; Daly and Macaulay, 1991). During SO GLOBEC, there were interannual differences in the extent and timing of the advance and retreat of sea ice (Chapter 3; Parkins on, 2002; Marrari et al., 2008); however, winter sea ice conditions were similar in 2001 and 2002, and sea ice biota concentrations were low at the ice-water interf ace during both years (0.05 0.07 g l-1) (Daly, 2004). Despite similar winter sea ice, krill recruitmen t showed marked interannual differences, suggesting that other processe s influence recruitment of E. superba in this area. It is here hypothesized that the early and persistent availability of phytoplankton in offshore waters of the Bellingshausen Sea supports early and repeated reproductive events during spring and summer and results in a high reproductive ou tput. In addition, high concentrations of phytoplankton in offshore and coastal areas dur ing summer and fall will lead to faster growth and development of larvae, which will be in better condition to survive overwinter. Low food conditions or the late onset of blooms will lead to late and/or poor reproduction and smaller/weaker larvae which may not be able survive overwinter. Although phytoplankton blooms in the Bellingsha usen Sea and Marguerite Bay were spatially variable, spring chlorophyll concentr ations exceeded minimum values (1 5 mg m-3) required to initia te reproduction of E. superba (Ross and Quetin, 1986) during both SO GLOBEC years. It is th e variability in the timing and persistence of these blooms, rather than the absolute chlorophyll concentr ations, what will infl uence the reproduction and recruitment success of euphausiids. The composition of the copepod community also showed marked interannual differences, which appear to be related to variability in food availability. During the chlorophyll-rich 2001 season, C. acutus dominated, comprising 52% of the total
116 copepods. C. acutus has been described as the only tr ue herbivore in Antarctic waters (Conover and Huntley, 1991; Atkinson, 1998), feeding mainly on phytoplankton during spring and summer. This species spawns ove r a short time and produces one distinct cohort. Maximum densities can be observe d in the surface layer during summer, when the population reaches stage copepodid V (C 5). In areas of deep bathymetry, C. acutus undergoes a seasonal ontogenetic migration, mo ving to deep waters during the fall and winter, where individuals do not feed (Sc hnack-Schiel and Mujica, 1994), whereas in shallower coastal areas, C. acutus moves to intermediate depths. M. gerlachei the second numerically dominant species during 2001 and most abundant copepod in 2002, is primarily omnivorous and its success doe s not rely heavily on high phytoplankton concentrations. This species starts to spawn in December and continues during summer. M. gerlachei does not undergo a seasonal ontogenetic migration and is more widely distributed throughout the water column (A tkinson and Peck, 1988; Schnack-Schiel and Hagen, 1995). The elevated phytoplankton concen trations observed in Marguerite Bay throughout the summer of 2001 likel y favored the reproduction of C. acutus which reached relatively high abundances and domi nated the copepod community during fall. Lower chlorophyll concentrations during 2002 re sulted in lower abundances of copepods. The copepod community was dominated by omnivorous species, such as M. gerlachei and Ctenocalanus spp., which together accounted for ~ 60% of total copepods, whereas C. acutus had an 85% population re duction relative to 2001. Controls on Zooplankton Spatia l Patterns in Ma rguerite Bay Copepods and macrozooplankton showed opposite horizontal distributions in
117 Marguerite Bay, with higher abundances of m acrozooplankton in the northern sectors of the study area, and greatest de nsities of copepods in southe rn areas. These distinct distributions are likely the re sult of a combination of circ ulation features and seasonal predation. The general circulat ion in Marguerite Bay (Fig. 1.2) involves a coastal current (APCC) that enters Marguerite Bay at the northern end, flows along the coast in a clockwise direction (Moffat et al., 2008), a nd exits Marguerite Bay along the outer coast of Alexander Island. In addition, drifter data revealed a clockwise gyr e-like circulation in the central areas of Marguerite Bay duri ng fall 2001 and 2002 (Bear dsley et al., 2004), while ADCP results showed the presence of two eddies in George VI Sound during fall 2001 (Dorland and Zhou, 2008). These features in combination with the general circulation of the APCC, c ontribute to creating a favorab le retention habitat for zooplankton in Marguerite Bay. These circ ulation features can transport zooplankton within coastal Marguerite Bay, particularly smaller taxa such as copepods, which are unable to swim against currents. The northern sectors of the study area, such as southern Crystal Sound, had a more sluggish flow than that observed for the APCC (Zhou et al., 2004). Given the capabilities of euphasuiids to swim at cruising speeds of 10 15 cm sec-1 (Hamner et al., 1983; Zhou and Dorland, 2004), it is likely that these larger organisms are able to maintain their distribut ions preferentially in these quieter northern areas, resulting in the higher abundances observe d relative to the southe rn sectors. In a study from the Crystal Sound area using Acousti c Doppler Current Profiler (ADCP) data, Zhou and Dorland (2004) demonstrate that the swimming capabilities of E. superba can determine the maintenance of aggregations in th e mesoscale circulation field of the area. The effects of predation are cumulative over the summer productive period and
118 may have contributed to the different spat ial distributions observed between copepods and euphausiids in Marguerite Bay during fa ll. Predation during fall was likely not strong enough to significan tly reduce copepod abundances in the northern sectors, particularly given that the much of the e uphausiid and copepod populations occurred at different depths in the wate r column. A negative correl ation between abundances of copepods and euphausiids was reported for summer in waters around South Georgia, where predation pressure of euphausiids was f ound to determine, at least in part, copepod distributions (Atkinson et al., 1999). In su mmary, the combined effects of currents and retention features, seasonal predation, and behavior of macrozooplankton, likely determined the distinct spatial distributi ons observed for these taxa during fall in Marguerite Bay. Vertical distribution of euphausiids and other zooplankton, including copepods, amphipods and mysids, rarely overlapped th roughout Marguerite Bay during 2001. All euphausiids were consistently shallower than the most abundant copepods species, C. acutus and M. gerlachei and these vertical distributions are consistent with other results for euphausiids of the continental shelf of Marguerite Bay (e.g., Lawson et al., 2008; Ashjian et al., 2008). Considering that the fall 2001 cruise took place during May June, after C. acutus has normally started its seasonal des cent into deeper waters, the vertical separation of the two groups could be attributed to differing winter behaviors. However, the omnivore M. gerlachei which also comprised a major fraction of total copepods, does not undergo seasonal vertical migrations and is known to feed throughout the year. Thus, the deep distribution of this species rela tive to euphausiids co uld be the result of accumulated predation pressure. In contrast to the vertical depth pa rtitioning observed in
119 2001, euphasuiids and copepods had an overl apping distribution during 2002. The overall shallower distribution of copepods in 2002 could have resulted from a combination of factors, including (1) an earlie r fall cruise (April Ma y) relative to 2001 (May June), (2) a smaller contribution of C. acutus to the total copepod abundance with higher proportions of surfacedwelling species such as Ctenocalanus spp and Oithona spp., and (3) reduced predation pressu re during fall 2002, a consequence of E. superba not feeding on copepods (K. Daly, pers. comm. ), as well as lower densities of other omnivorous and carnivorous zooplankton, such as T. macrura chaetognaths and medusae. Studies of mysids from the Southern Ocean are few; however, Brandt et al. (1998) described 37 species of Antarctic mysids 19 of which are endemic, and reviewed information on their biogeography and verti cal distributions. Mysids are generally hyperbenthic or bathypelagic and prefer an omnivorous diet, feeding on phytoplankton and a variety of zooplankton, including copepods coelenterates, and euphausiid molts (Hopkins, 1985). During our study, mysids we re located consistently deeper than euphausiids and amphipods, with some individuals recorded at depths of up to 800 m, the maximum depth sampled during SO GLOBEC. These distributions are consistent with previous coastal studies, while in deep-waters, Antarctic species have been recorded at depths of up to 4500 m (Torres and Hopkins, 1988; Brandt et al, 1998 ). Although depths in Marguerite Bay reach ~ 1,600 m at a few locat ions, the mean depth of the area is ~ 400 m (Bolmer et al., 2004) (Fig. 1.1). In deeper area s of the WAP, such as the vicinity of the shelf break, vertical distribu tions of mysids may extend to greater depths than those reported herein.
120 Amphipods were located at intermediate depths between euphausiids and mysids in 2001, but were shallower than both groups during 2002. Although data on the species composition of the amphipod community during SO GLOBEC are not currently available, the distinct vertical distributi ons observed could be re lated to interannual variability in the percent cont ribution of different species. For example, in 2001 deeper living large gammarid amphipods appear to have dominated, while smaller hyperiid amphipods, such as Themisto gaudichaudii which are known to inhabit shallower waters (Hopkins, 1985; Torres and Hopkins, 1988), ma y have been more important in 2002. Relationship between zooplankton and fall environmental parameters There were no clear trends between zooplankton spat ial distributions and fall environmental variables in Marguerite Ba y. The positive correlation observed in 2001 between pigment concentrations and abundances of Ctenocalanus spp. could be the result of the mainly herbivorous diet of this species (Hopkins, 1985); however, no other herbivores had positive correlations with this parameter. The positive relationship between the carnivore Paraeuchaeta spp. and pigment concentrations during 2002 is even less evident. Paraeuchaeta spp. are raptorial predators, feeding primarily on smaller copepods, such as Oncaea spp. and copepodites of M. gerlachei (Hopkins 1985), which are omnivores and may ingest signi ficant amounts of phytoplankton. However, these prey species did not show positive re lationships with pigment concentrations. Moreover, only a weak positive trend (not st atistically significant) was observed between Paraeuchaeta spp. and their main prey, Oncaea spp. (r = 0.39; p = 0.216) and M. gerlachei (r = 0.32; p = 0.308) during 2002. The ove rall lack of relationship between
121 pigment concentrations and grazers is likely due to the fact that herbivorous/omnivorous copepods may not have been feeding on phytopl ankton during mid to late fall. Among the macrozooplankton, only the omnivores T. macrura and mysids showed positive correlations with pigment concentrati ons in 2001 and 2002 respectively, but these relationships were not co nsistent interannually. Relationships between copepod abundances and bottom depth are consistent for some copepod species (Table 4.4). For example, G. tenuispinus C. acutus and Oncaea spp. were classified as deep species during both years, and also showed a significant tendency for higher abundances at locations of deeper bathymetry, such as those in the vicinity of Marguerite Trough and s outhern Laubeuf Fjord (Fig. 1.3). Summary The variability observed in total ab undance and percent composition of the zooplankton of Marguerite Bay during fall can be linked directly to the contrasting environmental conditions that prevailed dur ing the preceding spring-summer seasons. During spring-summer 2000/2001 a combination of warmer sea surface temperatures, higher than normal chlorophyll concentrati ons, and low sea ice cover resulted in a favorable environment for zooplankton reproduc tion and larval growth, which led to the distinctive composition and overall higher abundances obs erved during fall 2001. Above average concentrations of ch lorophyll in many areas, including the Bellingshausen Sea and Marguerite Bay, led to high con centrations of copepods, juvenile T. macrura and larval Euphausia spp. during fall 2001, and subsequent elev ated numbers of juvenile and adult E. superba and E. crystallorophias in fall 2002. On the other hand, lower surface
122 temperatures, extensive and persistent sea i ce cover and, consequently, lower than normal chlorophyll concentrations dur ing spring-summer 2001/2002, re sulted in lower plankton abundances in Marguerite Bay during fall 2002, particularly copepods, larval euphausiids, herbivorous m acrozooplankton and juvenile T. macrura For other groups that do not rely heavily on phytoplankton for reproduction and survival, such as ostracods, pteropods, chaetognaths, am phipods and mysids, the interannual environmental variability did not have as la rge an impact on their populations, which is evidenced by the smaller interannual di fferences observed in their abundances.
123 CHAPTER FIVE SUMMARY AND CONCLUDING REMARKS Validated satellite chlorophyll data offe r great insight into the temporal and spatial variability of phytoplankton dynamics in remote and difficult to access areas, such as the Southern Ocean. The high resolution and synoptic nature of these datasets are unattainable using traditional ship-based meas urements. Validation results confirm that SeaWiFS surface chlorophyll concentrations derived for the Southern Ocean are an accurate measure of in situ values between 0.05 1.5 mg m-3. These findings contradict previous results that report an underestim ation of SeaWiFS chlorophyll in Antarctic waters. The good agreement between SeaWiFS and in situ chlorophyll concentrations reported here was based on the use of chlo rophyll data determined by HPLC as ground truth, instead of chlorophyll c oncentrations estimated from fluorescence, which has been shown to introduce significant errors when cer tain accessory pigments are present in the water column. Because more than 90% of the Southern Ocean has chlorophyll values in the 0.05 1.5 mg m-3 range, it is not necessary to de velop an alternative bio-optical algorithm for this region. However, if co mputer models (e.g., to estimate primary production or eutrophic depth) have been developed using fluorometric methods as input, then the satellite estimates of chlorophyll concentrations will need adjustment to be consistent with these models. A major finding of this dissertation is th e presence of predictable and persistent
124 phytoplankton blooms in the Bellingshausen Sea and Marguerite Bay areas between 1997 and 2004, and the strong influence that environm ental variability in these areas has on the zooplankton community of the WAP. The l ack of field sampling in this region may partly explain why the magnitude of phytopl ankton aggregations has been previously overlooked. Chlorophyll concentrations in the southern sectors are consistently higher than in any other part of the WAP. The c limatology of chlorophyll c oncentrations for the period between 1997 and 2004 indicated that ch lorophyll values were elevated during 2001 in comparison to all other years, particul arly in the vicinity of Marguerite Bay. Zooplankton composition in the vicinity of Marguerite Bay was similar to that reported for previous WAP studies, and includ ed 12 species of copepods, three species of euphausiids, and ten other groups. The obser ved variability in total abundance and percent composition of the zooplankton during fall was strongly influenced by the contrasting environmental conditions that prevailed during the preceding spring-summer seasons. During spring-summer 2000/2001 warmer sea surface temper atures, higher than normal chlorophyll concentra tions, and low sea ice cover created a favorable environment for zooplankton reproduction and la rval growth. Above average chlorophyll concentrations in 2001 led to high c oncentrations of copepods, juvenile T. macrura and larval Euphausia spp. during fall 2001, and subsequent el evated abundances of juvenile and adult E. superba and E. crystallorophias during fall 2002. In contrast, lower surface temperatures, extensive and persistent sea i ce cover, and consequently lower than normal chlorophyll concentrations during spring-summer 2001/2002 re sulted in lower plankton abundances in Marguerite Bay during fall 2002, particularly copepods, larval euphausiids, herbivorous m acrozooplankton and juvenile T. macrura For other groups
125 that do not rely heavily on phytoplankton for re production and survival such as ostracods, pteropods, chaetognaths, amphipods and mysids, e nvironmental variability did not have a large impact on their populati ons, which is evidenced by the small interannual differences observed in their abundances. Zooplankton, particularly E. superba originating in the Be llingshausen Sea and southern areas along the WAP may be a significant source for populations found downstream in shelf waters of the northern An tarctic Peninsula, the Scotia Sea, and South Georgia (Fach and Klinck, 2006; Thorpe et al, 2007). Thus, variability in the timing and persistence of phytoplankton in the southern WAP will not only affect local zooplankton and predators, but ultimately may impact th e entire ecosystem of the WAP and adjacent Scotia Sea. The pivotal role that environmental variability in Marguerite Bay and the Bellingshausen Sea plays in structuring th e zooplankton community is demonstrated by the results presented here. Given the few datasets currently available, and the poor understanding of the processes than control environmental va riability in these regions, particularly chlorophyll dynamics, future st udies of ecosystem dynamics along the WAP should include the Marguerite Bay and Bellingshausen Sea regions.
126 REFERENCES CITED Ainley, D.G., and Demaster, D.P. 1996. Th e upper trophic levels in polar marine ecosystems. In: Smith, W.O. (Ed.), Polar Oceanography, Part B: Chemistry, Biology, and Geology. Academic Press, Inc., San Diego, pp 599-629. Arrigo, K.R., and McClain, C.R. 1994. Spring phytoplankton production in the western Ross Sea. Science 266, 261-263. Arrigo, K.R., and van Dijken, G.L. 2003. Phytoplankton dynamics within 37 Antarctic coastal polynya systems. Journa l of Geophysical Research 108, 3271. Arrigo, K.R., Worthen, D., Schnell, A., and Lizotte, M.P. 1998. Primary production in Southern Ocean waters. Journal of Geophysical Research 103, 15587-15600. Ashjian, C.J., Davis, C.S., Gallager, S.M., Wiebe, P.H., and Lawson, G.L. 2008. Distribution of larval krill and zoopla nkton in association with hydrography in Marguerite Bay, Antarctic Peninsula, in austral fall and winter 2001 described using the Video Plankton Recorder Deep Sea Research II 55: 455-471. Ashjian, C.J., Rosenwaks, G.A., Wiebe, P.H., Davis, C.S., Gallager, S.M., Copley, N.J., Lawson, G.L., and Alatalo, P. 2004. Di stribution of zooplankton on the continental shelf off Marguerite Bay, An tarctic Peninsula during austral fall and winter, 2001. Deep-Sea Re search II 51: 2073-2098. Atkinson, A. 1995. Omnivory and feeding sel ectivity in five copepod species during spring in the Bellingshausen Sea, Antarcti ca. ICES Journal of Marine Science 52: 385-396. Atkinson, A. 1998. Life cycle strategies of ep ipelagic copepods in the Southern Ocean. Journal of Marine Systems 15: 289-311.
127 Atkinson. A., and Peck, J.M. 1988. A summer-w inter comparison of zooplankton in the oceanic area around South Georgia. Polar Biology 8: 463-473. Atkinson, A., Ward, P., Hill, A., Brierle y, A.S., and Cripps, G.C. 1999. Krill-copepod interactions at South Ge orgia, Antarctica, II. Euphausia superba as a major control on copepod abundance. Marine Ecology Progress Series 176: 63-79. Atkinson, A., Whitehouse, M.J., Priddle, J ., Cripps, G.C., Ward, P., Brandon, M.A., 2001. South Georgia, Antarctica: a productive, cold water, pelagic ecosystem. Marine Ecology Progress Series 216, 279-308. Barlow, R.G., Mantoura, R.F.C., and Cu mmings, D.G. 1998. Phytoplankton pigment distributions and associated fluxes in th e Bellingshausen S ea during the austral spring 1992. Journal of Marine Systems 17, 97-113. Beardsley, R.C., Limeburner, R., and Owe ns, W.B. 2004. Drifter measurements of surface currents near Marguerite Bay on the western Antarctic Peninsula shelf during austral summer and fall, 2001 and 2002. Deep-Sea Research II 51: 19471964. Bianchi, T.S., Lambert, C., and Biggs, D. C. 1995. Distribution of chlorophyll a and phaeopigments in the northwestern Gu lf of Mexico: A comparison between fluorometric and high-performance liquid chromatography measurements. Bulletin of Marine Science 56, 25-32. Bidigare, R.R., Frank, T.J., Zastrow, C., a nd Brooks, J.M. 1986. The distribution of algal chlorophylls and their degr adation products in the Southern Ocean. Deep-Sea Research 33, 923-937. Bolmer, S.T., Beardsley, R.C., Pudsey, C ., Morris, P., Wiebe, P., Hofmann, E.E., Anderson, J., and Maldonado, A. 2004. A hi gh-resolution bathymetric map from Marguerite Bay and adjacent west Antarc tic Peninsula shelf from the southern ocean GLOBEC program. Woods Hole Oceanographic Institution Technical Report, WHOI-2004-02, 78pp. Boyd, P.W. 2002. Environmental factors cont rolling phytoplankton processes in the Southern Ocean. Journal of Phycology 38, 844-861.
128 Brandt, A., Mhlenhardt-Siegel, U., and Si egel, V. 1998. An account of the Mysidacea (Crustacea, Malacostraca) of the Southern Ocean. Antarctic Sc ience 10(1): 3-11. Brinton. E., and Townsend, A.W. 1991. Development rates a nd habitat shifts in the Antarctic neritic euphausiid Euphausia crystallorophias Deep-Sea Research 38: 1195-1211. Brinton, E., Huntley, M., and Townsend, A.W. 1986. Larvae of Euphausia superba in the Scotia Sea and Bransfield Strait in March 1984 development and abundance compared with 1981 larvae. Polar Biology 5: 221-234. Cabal, J.A., Alvarez-Marqus, F., Acua J.L., Quevedo, M., Gonzalez-Quirs, R., Huskin, I., Fernndez, D., Rodrguez del Valle, C., and Anadn, R. 2002. Mesozooplankton distribution and grazing during the productive season in the Northwest Antarctic Peninsula (FRUELA cr uises). Deep-Sea Research II 49: 869882. Campbell, J.W. 1995. The lognormal distribution as a model for bio-optic al variability in the sea. Journal of Geophysical Research 100: 13,237-13,254. Campbell, J.W. Methods for quantifying the uncertainty in a chlorophyll algorithm as demonstrated for the SeaWiFS algorit hm. Remote Sensing of Environment (submitted) Carder, K.L., Steward, R.G., Harvey, G.R., and Ortner, P.B. 1989. Marine humic and fulvic acids: their effects on remote sensing of ocean chlorophyll. Limnology and Oceanography 34: 68-81. Cavalieri, D., Parkinson, C., Gloersen, P., and Zwally, H.J. 1996, updated 2005. Sea ice concentrations from Nimbus-7 SMMR and DMSP SSM/I passive microwave data, June to September 2001. Boulder, CO, US A: National Snow and Ice Data Center. Digital media. Chisholm, S.W., Olson, R.J., Zettler, E. R., Goericke, R., Waterbury, J., and Welschmeyer, N. 1988. A novel free liv ing prochlorophyte abundant in the oceanic euphotic zone. Nature 334: 340-343.
129 Conover, R.J., and Huntley, M. 1991. Copepods in ice-covered seas Distribution, adaptations to seasonally limited food, meta bolism, growth patterns and life cycle strategies in polar seas. Journa l of Marine Systems 2: 1-41. Costa, D.P., and Crocker, D.E. 1996. Ma rine mammals of the Southern Ocean. Foundations for Ecological Research West of the Antarctic Peninsula. Antarctic Research Series 70: 287-301. Croxall, J.P. 1984. Seabirds. In: Laws, R.M. (Ed.), Antarctic Ecology. Academic Press, London, 533-619 pp. Daly, K.L. 1990. Overwintering developmen t, growth, and feeding of larval Euphausia superba in the Antarctic marg inal ice zone. Limnology and Oceanography 35(7): 1564-1576. Daly, K.L. 2004. Overwintering grow th and development of larval Euphausia superba : an interannual comparison under varying environmental conditions west of the Antarctic Peninsula. Deep-Sea Research II 51: 2119-2138. Daly, K.L., and Macaulay, M.C. 1991. The influence of physical and biological mesoscale dynamics on the seasonal distribution and behavior of Euphausia superba in the antarctic marginal ice zone Marine Ecology Progress Series 79: 37-66. Daly, K.L., and Zimmerman, J. 2004. Co mparisons of morphology and neritic distributions of Euphausia crystallorophias and Euphausia superba furcilia during autumn and winter west of the An tarctic Peninsula. Polar Biology 28: 7281. Daly, K.L., Smith, W.O., Johnson, G.C., Di Tullio, G.R., Jones, D.R., Mordy, C.W., Feely, R.A., Hansell, D.A., and Zha ng, J. 2001. Hydrography, nutrients, and carbon pools in the Pacific sector of the Southern Ocean: Implications for carbon flux. Journal of Geophysical Research 106: 7107-7124. Darecki, M., and Stramski, D. 2004. An eval uation of MODIS and SeaWiFS bio-optical algorithms in the Baltic Sea. Remote Sensing of Environment 89: 326-350.
130 De Baar, H.J., De Jong, J.T.M., Bakker, D.C. E., Lscher, B.M., Veth, C., Bathmann, U., and Smetacek, V. 1995. Importance of iron for plankton blooms and carbon dioxide drawdown in the Sout hern Ocean. Nature 373: 412-415. Deibel, D., and Daly, K.L. 2007. Zooplankton pro cesses in Arctic and Antarctic polynyas. In: Smith, W.O., Jr., Barber, D., (Eds.), Polynyas: Windows into Polar Oceans. Elsevier Oceanography Series 74: 271-321. Dierssen, H.M., and Smith, R. 2000. Bio-optical properties and remote sensing ocean color algorithms for Antarctic Peninsula waters. Journal of Geophysical Research 105: 26,301-26,312. Dierssen, H.M., Vernet, M., and Smith, R. C. 2000. Optimizing models for remotely estimating primary production in Antarctic coastal waters. Antarctic Science 12: 20-32. Dinniman, M.S., and Klinck, J.M. 2004. A mo del study of circulation and cross-shelf exchange on the west Antarctic Peninsula continental shelf. Deep-Sea Research II 51: 2003-2022. Donnelly, J., Sutton, T.T., and Torres, J.J. 2006. Distribution and abundance of micronekton and macrozooplankton in the NW Weddell Sea: relation to a spring ice-edge bloom. Polar Biology 29: 280-293. Dorland, R.D., and Zhou, M. 2008. Circulation a nd heat fluxes during the austral fall in George VI Sound, Antarctic Peninsula. Deep-Sea Research II 53: 294-308. Ducklow, H.W., Fraser, W., Karl, D.M., Quetin, L.B., Ross, R.M., Smith, R.C., Stammerjohn, S.E., Vernet, M., and Dani els, R.M., 2006. Water-column processes in the West Antarctic Peninsula and th e Ross Sea: interannua l variations and foodweb structure. Deep-Sea Research II 53: 834-852. El-Sayed, S.Z., 1971. Observations on phyt oplankton bloom in the Weddell Sea. In: Llano, G.A., Wallen, I.E. (Eds.), Biology of the Antarctic Seas IV vol. 17, pp 301-312.
131 El-Sayed, S.Z. 1994. Southern Ocean ecology: the BIOMASS perspective. Cambridge University Press. 421pp. El-Sayed, S.Z. 2005. History and evolution of primary productivity studies of the Southern Ocean. Polar Biology 28: 423-438. El-Sayed, S.Z., and Taguchi, S., 1981. Pr imary production and standing crop of phytoplankton along the ice-edge in the Weddell Sea. Deep-Sea Research 28A: 1017-1032. Esaias, W.E., Abbott, M.R., Barton, I., Br own, O.B., Campbell, J.W., Carder, K.L., Clark, D.K., Evans, R.H., Hoge, F.E., Gordon, H.R., Balch, W.M., Letelier, R., and Minnett, P.J. 1998. An overview of MODIS capabilities for ocean science observations. IEEE Transactions on Ge oscience and Remote Sensing 36: 12501265. Fach, B.A., and Klinck, J.M. 2006. Transport of Antarctic krill ( Euphausia superba ) across the Scotia Sea. Part I: Circulati on and particle tracking simulations. DeepSea Research I 53: 987-1010. Fraser, W.R., and Trivelpiece, W.Z. 1996. Factor s controlling the distribution of seabirds: winter-summer heterogeneity in the di stribution of Adlie penguin populations. Antarctic Research Series 70: 257-272. Garibotti, I.A., Vernet, M., and Ferrario, M.E. 2005. Annually recurrent phytoplanktonic assemblages during summer in the seasonal ice zone west of the Antarctic Peninsula (Southern Ocean). D eep-Sea Research I 52: 1823-1841. Garibotti, I.A., Vernet, M., Fe rrario, M.E., Smith, R.C., Ross, R.M., and Quetin, L.B. 2003. Phytoplankton spatial distribution pa tterns along the western Antarctic Peninsula (Southern Ocean). Marine Ecology Progress Se ries 261: 21-39. Gibbs C.F. 1979. Chlorophyll b interference in the fluorometric determination of chlorophyll a and phaeo-pigments. Australian Journal of Marine and Freshwater Research 30: 597-606.
132 Gloersen, P., Campbell, W.J., Cavalieri, D.J., Comiso, J.C., Parkinson, C.L., and Zwally, H.J. 1992. Arctic and Antarctic Sea ice, 1978-1987: Satellite Passive Microwave Observations and Analysis. NASA Special Publication 511. Goericke, R., and Repeta, D.J. 1993. Chlorophyl ls a and b and divinyl chlorophylls a and b in the open subtropical North Atlantic Ocean. Marine Ecology Progress Series 101: 307-313. Gordon, H.R. 1992. Diffuse reflectance of the ocean: influence of nonuniform phytoplankton pigment profile. Applied Optics 31: 2116-2129. Gregg, W.W., and Casey, N.W. 2004. Global and regional ev aluation of the SeaWiFS chlorophyll dataset. Remote Se nsing of Environment 93: 463-479. Hamner, W.M., Hamner, P.P., Strand, S.W ., and Gilmer, R.W. 1983. Behavior of Antarctic krill Euphausia superba : chemoreception, feeding, schooling, and molting. Science 220: 433-435. Hempel, G. 1985. Antarctic marine food webs In: Siegfried, W.R., Condy, P.R., and Laws, R.M. (Eds.), Antarctic nutrient cycles and food webs. Springer-Verlag, Berlin, pp 266-270. Hempel, I., and Marschoff, E. 1980. Euphausiid larvae in the Atlantic sector of the Southern Ocean. Meeresforschung 28: 32-47. Hewitt, R.P., Demer, D.A., Emery, J.H. 2003. An 8-year cycle in krill biomass density inferred from acoustic surveys conducted in the vicinity of the South Shetland Island during the austral summers of 1991 1992 through 2001 2002. Aquatic Living Resources 16: 205-213. Hoffman, E.E., Capella, J.E., Ross, R.M., and Quetin, L.M. 1992. Mode ls of the early life history of Euphausia superba Part I. Time and te mperature dependence during the descent-ascent cycle. Deep-Sea Research 39: 1177-1200. Hofmann, E.E., Wiebe, P.H., Costa, D.P., and Torres, J.J. 2004. An overview of the Southern Ocean Global Ocean Ecosystems Dynamics program. Deep-Sea Research II 51: 1921-1924.
133 Holm-Hansen, O., El-Sayed, S.Z., Francesch ini, G.A., and Cuhel, R.L. 1977. Primary production and the factors controlling phytoplankton growth in the Southern Ocean. Adaptations within Antarctic ecosy stems: proceedings of the Third SCAR Symposium on Antarc tic Biology. pp 11-50. Holm-Hansen, O., Kahru, M., and Hewes, C.D. 2005. Deep chlorophyll a maxima (DCMs) in pelagic Antarctic waters. II. Relation to bathym etric features and dissolved iron concentrations. Marine Ecology Progress Series 297: 71-81. Holm-Hansen, O., Kahru, M., Hewes, C.D., Kawaguchi, S., Kameda, T., Sushin, V.A., Krasovski, I., Priddle, J., Korb, R., Hewitt, R.P., and Mitchell, B.G. 2004a. Temporal and spatial distribution of chlo rophyll-a in surface waters of the Scotia Sea as determined by both shipboard meas urements and satellite data. Deep-Sea Research I 51: 1323-1331. Holm-Hansen, O., Mitchell, B.G. 1991. Sp atial and temporal distribution of phytoplankton and primary production in th e western Bransfield Strait region. Deep-Sea Research 38: 961-980. Holm-Hansen, O., Mitchell, B.G., Hewes, C.D., and Karl, D.M. 1989. Phytoplankton blooms in the vicinity of Palmer Stat ion, Antarctica. Polar Biology 10: 49-57. Holm-Hansen, O., Naganobu, M., Kawaguchi S., Kameda, T., Krasovski, I., Tchernyshkov, P., Priddle, J., Korb, R ., Brandon, M., Demer, D., Hewitt, R.P., Kahru, M., and Hewes, C.D. 2004b. Factors influencing the distribution, biomass, and productivity of phytoplankton in the Sc otia Sea and adjoining waters. DeepSea Research II 51: 1333-1350. Hopkins, T.L. 1985. Food web of an Antarctic midwater ecosystem. Marine Biology 89: 197-212. Hopkins, T.L., and Torres, J.J. 1988. The zoopl ankton community in th e vicinity of the ice edge, western Weddell Sea, March 1986. Polar Biology 9: 79-87. Hu, C., Carder, K.L., and Mller-Karger, F.E. 2001. How precise are SeaWiFS ocean color estimates? Implications of digiti zation-noise errors. Remote Sensing of Environment 76: 239-249.
134 Hu, C., Muller-Karger, F.E., Taylor, C., Carder K.L., Kelble, C., Johns, E., and Heil, C. 2005. Red tide detection and tracing us ing MODIS fluorescence data: A regional example in SW Florida coastal waters. Remote Sensing of Environment 97: 311321. Hu, C., Nababan, B., Biggs, D.C., and Mlle r-Karger, F.E. 2004. Variability of biooptical properties at sampling stations a nd implications for remote sensing. A case study in the NE Gulf of Me xico. International Journa l of Remote Sensing 25: 2111-2120. Huntley, M., Karl, D.M., Niiler, P., and Holm-Hansern, O. 1991. Research on Antarctic Coastal Ecosystem Rates: an interdis ciplinary field experiment. Deep-Sea Research 38: 911-942. Ikeda, T. 1984. Development of th e larvae of the Antarctic krill ( Euphausia superba Dana) observed in the labor atory. Journal of Experime ntal Marine Biology and Ecology 75: 107-117. Ikeda, T. 1986. Preliminary observations on the development of the larvae of Euphausia crystallorophias Holt and Tattersall in the laboratory. Mem. Natl. Inst. Polar Res., Special Issue 40: 183-186. Jeffrey, S.W. 1976. A report of green algal pigm ents in the Central North Pacific Ocean. Marine Biology 37: 33-37. Jeffrey, S.W., Mantoura, R.F.C ., and Bjornland, T. 1997. Data for the identification of 47 key phytoplankton pigments. In: Jeffrey, S.W., Mantoura, R.F.C., and Wright, S.W. (Eds.), Phytoplankton pigments in oceanography. UNESCO Publishing, Paris, 661pp. Kawaguchi, S., and Satake, M. 1994. Relationshi p between recruitment of the Antarctic krill and the degree of ice cover near the South She tland Islands. Fisheries Science 60, 123-124. King, J.C. 1994. Recent climate variability in th e vicinity of the Antarctic Peninsula. International Journal of Climatology 14: 357-369.
135 Kirk, J.T.O. 1983. Light and Photosynthes is in Aquatic Ecosystems. Cambridge University Press, Cambridge, pp 49-50. Klinck, J.M., Hofmann, E.E., Beardsley, R.C., Salihoglu, B., and Howard, S. 2004. Water-mass properties and circulation on th e west Antarctic Peninsula continental shelf in austral fall and winter 2001. Deep-Sea Research II 51: 1925-1946. Klinck, J.M., and Nowlin, W.D., Jr. 2001. Antarc tic Circumpolar Curren t. In: Steele, J., Thorpe, S., and Turekian, K., (Eds.), En cyclopedia of Ocean Sciences. Academic Press, pp 151-159. Korb, R.E., Whitehouse, M.J., and Ward, P. 2004. SeaWiFS in the Southern Ocean: spatial and temporal variability in ph ytoplankton biomass around South Georgia. Deep-Sea Research II 51: 99-116. Lancraft, T.M., Reisenbichler, K.R., Robis on, B.H., Hopkins, T.L., and Torres, J.J. 2004. A krill-dominated micronekton and m acrozooplankton community in Crocker Passage, Antarctica, with an estimate of fish predation. Deep-Sea Research II 51: 2247-2260. Lascara, C.M., Hofmann, E.E., Ross, R.M., and Quetin, L.B. 1999. Seasonal variability in the distribution of Antarctic krill, Euphausia superba west of the Antarctic Peninsula. Deep-Sea Research 46: 951-984. Laws, R.M. 1985. The ecology of the southern ocean. American Scientist 73: 26-40. Lawson, G.L., Wiebe, P.H., Ashjian, C.J., and Stanton, T.K. 2008. Euphasuiid distribution along the Western Antarctic Peninsula Part B: Distribution of euphasuiid aggregations and biomass, and associations with environmental features. Deep-Sea Research II 53: 432-454. Loeb, V., Siegel, V., Holm-Hansen, O., Hew itt, R., Fraser, W., Trivelpiece, W., and Trivelpiece, S. 1997. Effects of sea ice exte nt and krill or salp dominance on the Antarctic food web. Nature 387: 897-900.
136 Lohrenz, S.E., Weidemann, A.D., and Tuel M. 2003. Phytoplankton spectral absorption as influenced by community size struct ure and pigment composition. Journal of Plankton Research 25: 35-61. Lopez, M.D.G., and Huntley, M. 1995. Feeding and diel vertical migration cycles of Metridia gerlachei (Giesbrecht) in coastal waters of the Antarctic Peninsula. Polar Biology 15: 21-30. Lorenzen, C.J. 1981. Chlorophyll b in the eastern North Pacific Ocean. Deep-Sea Research 28: 1049-1056. Lowry, L.F., Ward Testa, J., and Clavert, W. 1988. Notes on winter feeding of crabeater and leopard seals near th e Antarctic Peninsula. Polar Biology 8: 475-478. Makarov, R.R. 1979. Larval distribut ion and reproductive ecology of Thysanoessa macrura (Crustacea: Euphausiacea) in the Scotia Sea. Mari ne Biology 52: 377386. Makarov, R.R. 1981. Larval development of the Antarctic euphausiids. BIOMASS Handbook No.3. Maritorena, S., Siegel, D.A., and Peterson, A.R. 2002. Optimization of a semianalytical ocean color model for global-scale a pplication. Applied Optics 41: 2705-2714. Marr, J.W.S. 1962. The natural history and geography of the Antarctic krill ( Euphausia superba ). Discovery Report 32: 33-464. Marrari, M., Daly, K.L., and Hu, C. 2008. Spa tial and temporal variability of SeaWiFS chlorophyll a distributions west of the Antarctic Peninsula: Impli cations for krill production. Deep-Sea Research II 55: 377-392. Marrari, M., Hu, C., and Daly, K.L., 2006. Validation of SeaWiFS chlorophyll a concentrations in the Southern Ocean: A revisit. Remote Sensing of Environment 105: 367-375.
137 Marschall, H. 1988. The overwintering strategy of Antarctic krill under the pack-ice of the Weddell Sea. Polar Biology 9: 129-135. Mauchline, J. 1981. Key for the identifica tion of Antarctic euphausiids. BIOMASS Handbook No.5. McClain, C.R., Cleave, M.L., Feldman, G.C ., Gregg, W.W., Hooker, S.B., and Kuring, N. 1998. Science quality SeaWiFS data for global biosphere research. Sea Technology 39: 10-16. Measures, C.I., and Vink, S. 2001. Dissolved Fe in the upper waters of the Pacific sector of the Southern Ocean. Deep -Sea Research II 48: 3913-3941. Meyer, B., Atkinson, A., Blume, B., and Bathmann, U. 2003. Feeding and energy budgets of larval Antarctic krill Euphausia superba in summer. Marine Ecology Progress Series 257: 167-177. Moffat, C., Beardsley, R.C., Owens, B., and van Lipzig, N. 2008. A first description of the Antarctic Peninsula Coastal Curre nt. Deep-Sea Resear ch II 55: 276-293. Moline, M., and Przelin, B.B. 1996. Long-term monitoring and analyses of physical factors regulating variability in coastal Antarctic phytoplankton biomass, in situ productivity, and taxonomic compositi on over subseasonal, seasonal and interannual time scales. Marine Ecology Progress Series 145: 143-160. Moline, M.A., Przelin, B., Schofield, O., a nd Smith, R.C. 1997. Temporal dynamics of coastal Antarctic phytoplankton: environmen tal driving forces and impact of a 1991/92 summer diatom bloom on the nutri ent regimes. In: Battaglia, B., Valencia, J., and Walton, D.W.S. (Eds.) Antarctic communities species structure and survival. Cambridge University Press, pp 67-72. Moore, J.K., and Abbott, M.R. 2000. Phytopla nkton chlorophyll distributions and primary production in the Southern Ocean. Journal of Geophysical Research 105: 28,70928,722.
138 Moore, J.K., Abbott, M.R., Richman, J.G., Smith, W.O., Cowles, T.J., Coale, K.H., Gardner, W.D., and Barber, R.T. 1999. SeaW iFS satellite ocean color data from the Southern Ocean. Geophysical Research Letters 26: 1465-1468. Morel, A. 1988. Optical modeling of the upper ocean in relation to its biogenous matter content (case 1 waters). Journal of Geophysical Research 93: 10,749-10,768. Mujica, A. 1989. Distribucin del Orden Euphaus iacea en el Estrecho Bransfield y zonas adyacentes (SIBEX-INACH, Enero-Frebrer o, 1984). Pesq. Antart. Bras. 1(1): 3544. Mller-Karger, F.E., Varela, R., Thunell, R., Astor, Y., Zhang, H., Luerssen, R., and Hu, C. 2004. Processes of coastal upwelling and carbon flux in the CARIACO basin. Deep Sea Research 51: 927-943. Murase, H., Matsuoka, K., Ichii, T., and Ni shiwaki, S. 2002. Relationahip between the distribution of euphausiids and baleen whales in the Antarctic (35E 145W). Polar Biology 25: 135-145. Nemoto, T., and Nasu, K. 1958. Thysanoessa macrura as a food of bale en whales in the Antarctic. Sci. Rep. Whales Re. Inst. 13: 193-199. Nicol, S., De la Mare, W.K., and Stolp, M. 1995. The energetic cost of egg production in Antarctic krill ( Euphusia superba Dana). Antarctic Science 7: 25-30. Nordhasuen, W. 1992. Distribution a nd growth of larval and adult Thysanoessa macrura (Euphausiacea) in the Bransfield St rait region, Antarctica. Marine Ecology Progress Series 83: 185-196. Nordhausen, W. 1994. Distribut ion and diel vertical mi gration of the euphausiid Thysanoessa macrura in Gerlache Strait, Antarctica.Polar Biology 14: 219-229.
139 OReilly J.E., Maritorena, S., OBrien, M.C., Siegel, D.A., Toole, D., Menzies, D., Smith, R.C., Mueller, J.L., Mitchell, B. G., Kahru, M., Chavez, F.P., Strutton, P., Cota, C.F., Hooker, F.B., McClain, C.R ., Carder, K.L., Mller-Karger, F.E., Harding, L., Magnuson, A., Phinney, D., M oore, G.F., Aiken, J., Arrigo, K.R., Letelier, R., and Culver, M. 2000. SeaWi FS postlaunch calibration and validation analyses: Part 3. SeaWiFS postlaunch techni cal report series, 11. In: Hooker, S.B., and Firestone, E.R. (Eds.), NAS A Technical Memorandum 2000-206892, 49pp. Orsi, A.H., Whitworth III, H., and Nowli n, W.D. 1995. On the meridional extent and fronts of the Antarctic Circumpolar Current. Deep-Sea Research 42: 641-673. Park, T. 1994. Taxonomy and distribution of the marine calanoid copepod family Euchaetidae. Bulletin of the Scripps In stitution of Oceanography University of California, San Diego v. 29, University of California Press, Berkeley, CA. 203pp. Parkinson, C.L. 2002. Trends in the length of the Southern Ocean sea-ice season, 19791999. Annals of Glaciology 34: 435-440. Parsons, T.R., Takahashi, M., and Ha rgrave, B. 1984. Biological Oceanographic Processes. Pergamon Press, Oxford. 330pp. Przelin, B.B., Hofmann, E.E., Mengelt, C., and Klinck, J.M. 2000. The linkage between Upper Circumpolar Deep Water (UCDW) and phytoplankton assemblages on the west Antarctic Peninsula continental shel f. Journal of Marine Research 58: 165202. Przelin, B.B., Hofmann, E.E., Moline, M., a nd Klinck, J.M. 2004. Physical forcing of phytoplankton community structure and pr imary production in continental shelf waters of the Western Antarctic Peninsul a. Journal of Marine Research 62: 419460. Quetin, L.B, and Ross, R.M. 2003. Episodic recruitment in Antarctic krill, Euphausia superba in the Palmer LTER study region. Ma rine Ecology Progress Series 259: 185-200. Quetin, L.B., Ross, R.M., Frazer, T.K., Amsler M.O., Wyatt-Evens, C., and Oakes, S.A. 2003. Growth of larval krill, Euphausia superba in fall and winter west of the Antarctic Peninsula. Ma rine Biology 143: 833-843.
140 Rakusa-Suszczewski, S. 1984. Krill larvae in the Atlantic sector of the Southern Ocean during FIBEX 1981. Polar Biology 3: 141-147. Rodriguez, F., Varela, M., and Zapata, M. 2002. Phytoplankton assemblages in the Gerlache and Bransfield Straits (Antar ctic Peninsula) determined by light microscopy and CHEMTAX analysis of HP LC pigment data. Deep-Sea Research II 49: 723-747. Ross, R.M., and Quetin, L.B. 1983. Spawning fr equency and fecundity of the Antarctic krill Euphausia superba Marine Biology 77: 201-205. Ross, R.M., and Quetin, L.B. 1986. How productiv e are Antarctic kri ll? BioScience 36: 264-269. Ross, R.M., and Quetin, L.B. 1989. Energetic cost to develop to the first feeding stage of Euphausia superba Dana and the effect of delays in food availability. Journal of Experimental Marine Bi ology and Ecology 133: 103-127. Ross, R.M., Hofmann, E.E., and Quetin, L. B. 1996. Foundations for Ecological Research West of the Antarctic Peninsula. Am erican Geophysical Union, Washington, DC. 448pp. Sarmiento, J.L., Hughes, T.M.C., Stouffe r, R.J., and Manabe, S. 1998. Simulated response of the ocean carbon cycle to anthropogenic climate warming. Nature 393: 245-249. Savidge, G., Harbour, D., Gilpin, L.C., a nd Boyd, P.W. 1995. Phytopl ankton distributions and production in the Bellingshausen Sea, Austral spring 1992. Deep-Sea Research II 42: 1201-1224. Schnack-Schiel, S.B. 2001. Aspects of the study of the life cycles of Antarctic copepods. Hydrobiologia 453/454: 9-24. Schnack-Schiel, S.B., and Hagen, W. 1995. Life cycle strategies of Calanoides acutus Calanus propinquus and Metridia gerlachei (Copepoda: Calanoida) in the eastern Weddell Sea, Antarctica. ICES Journal of Marine Science 52: 541-548.
141 Schnack-Schiel, S.B., and Mujica, A. 1994. Th e zooplankton of the Antarctic Peninsula region. In: El-Sayed (Ed.), Southern O cean Ecology: the BIOMASS prespective. Cambridge University press, pp 79-92. Schnack-Schiel, S.B., Hagen, W., and Mizdalski, E. 1991. Seasonal comparison of Calanoides acutus and Calanus propinquus (Copepoda: Calanoida) in the southeastern Weddell Sea, Antarctica. Marine Ecology Progress Series 70: 17-27. Sedwick, P.N., and DiTullio, G.R. 1997. Regulat ion of algal blooms in Antarctic shelf waters by the release of iron from melti ng sea ice. Geophysical Research Letters 24: 2515-2518. Shapiro, S.S., Wilk, M.B., and Chen, H. J. 1968. A comparative study of various tests of normality. Journal of the American Statistical Associ ation 63: 1343-1372. Siegel, V. 1987. Age and growth of Antarc tic Euphausiacea (crust acean) under natural conditions. Marine Biology 96: 483-495. Siegel, V, and Harm, U. 1996. The composition, abundance, biomass and diversity of the epipelagic zooplankton co mmunities of the Southern Bellingshausen Sea (Antarctic) with special reference to kril l and salps. Arch. Fish. Mar. Res. 44: 115-139. Siegel, V. 1992. Assess ment of the krill ( Euphausia superba ) spawning stock off the Antarctic Peninsula. Arch. Fisch. Wiss. 41: 101-130. Siegel, V. 2005. Distribution and population dynamics of Euphausia sup erba: summary of recent findings. Polar Biology 29: 1-22. Siegel, V. 1988. A concept of seasonal variation of krill ( Euphausia superba ) distribution and abundance west of the An tarctic Peninsula. In: Sahr hage, D., (Ed.), Antarctic Ocean and Resources Variability. Springer, Berlin, pp 219-230. Siegel, V., Bergstrom, B., Mhlenha rdt-Siegel, and U., Thomasson, M. 2002. Demography of krill in the Elephant Island area during summer 2001 and its significance to stock recruitment. Antarctic Science 14: 162-170.
142 Siegel, V., and Loeb, V. 1995. R ecruitment of Antarctic krill Euphausia superba and possible causes for its vari ability. Marine Ecology Pr ogress Series 123: 45-56. Siegel, V., Loeb, V., and Grger, J. 1998. Krill ( Euphausia superba ) density, proportional and absolute recruitment and biomass in the Elephant Island region (Antarctic Peninsula) during the period 1977 to 1997. Polar Biology 19: 393-398. Siegel, V., Ross, R.M., a nd Quetin, L.B. 2003. Krill ( Euphausia superba ) recruitment indices from the western Anta rctic Peninsula: are they representative of larger regions? Polar Biology 26: 672-679. Smith, D.A., Hofmann, E.E., Klinck, J.M., and Lascara, C.M. 1999. Hydrography and circulation of the west Antarctic Peninsul a continental shelf. Deep-Sea Research I 46: 925-949. Smith, R.C., Baker, K.S., and Dustan, P. 1981. Fluorometric techniques for the measurement of oceanic chlorophyll in th e support of remote sensing. SIO Ref. 81-17. Smith, R.C., Baker, K.S., Byers, M.L., and Stammerjohn, S.E. 1998b. Primary productivity of the Palmer Long Term Ec ological Research Area and the Southern Ocean. Journal of Marine Systems 17: 245-259. Smith, R.C., Baker, K.S., Dierssen, H.M ., Stammerjohn, S.E., and Vernet, M. 2001. Variability of primary produc tion in an Antarctic marine ecosystem as estimated using a multi-scale sampling strategy. American Zoologist 41: 40-56. Smith, R.C., Baker, K.S., and Vernet, M. 1998a. Seasonal and intera nnual variability of phytoplankton biomass west of the Antarc tic Peninsula. Journal of Marine Systems 17: 229-243. Smith, R.C., Dierssen, H. and Vernet, M. 1996b. Phytoplankton biomass and productivity in the Western Antarctic Peninsula re gion. In: Ross, R.M., Hofmann, E.E. and Quetin, L.B. (Eds.), Foundations for Ecol ogical Research West of the Antarctic Peninsula. AGU Antarctic Research Series 70, 233-256.
143 Smith, R.C., Stammerjohn, S.E., and Baker, K.S. 1996a. Surface air temperature variations in the wester n Antarctic Peninsula regi on. In: Ross, R.M., Hofmann, E.E., and Quetin, L.B. (Eds.), Foundation for Ecological Research West of the Antarctic Peninsula. AGU Antarc tic Research Series 70, 105-121. Smith, W.O., and Nelson, D.M. 1985. Phytopl ankton bloom produced by a receding ice edge in the Ross Sea: spa tial coherence with the dens ity field. Science 227: 163166. Sullivan, C.W., Arrigo, K.R., McClain, C.R., Comiso, J.C., and Firestone, J. 1993. Distribution of phytoplankton blooms in the Southern Ocean. Science 262: 18321837. Thorpe, S.E., Murphy, E.J., and Watkins, J. L. 2007. Circumpolar connections between Antarctic krill ( Euphausia superba Dana) populations: Investigating the roles of ocean and sea ice transport. Deep-Sea Research I 54: 792-810. Trees, C.C., Kennicutt, M.C., and Brooks, J.M. 1985. Errors associated with the standard fluorometric determination of chlorophylls and phaeopigments. Marine Chemistry 17: 1-12. Trivelpiece, W.Z., and Fraser, W.R. 1996. The breeding biology and distribution of Adlie penguins: adaptations to environmental variability. Antarctic Research Series 70: 273-285. Vaughan, D.G., Marshall, G.J., Connelley, W. M., Parkinson, C., Mulvaney, W., Hodgson, D.A., King, J.C., Pudsey, C.J., and Turner J. 2003. Recent rapid regional climate warming on the Antarctic Peninsula. Climate Change 60: 243-274. Vaughan, D.G., and Doake, C.S.M. 1996. Recent atmospheric warming and retreat if ice shelves on the Antarctic Pe ninsula. Nature 379: 328-331. Voronina, N.M. 1998. Comparative abundance and distribution of majo r filter-feeders in the Antarctic pelagic zone. Journal of Marine Systems 17: 375-390.
144 Watkins, J.L., Hewitt, R., Naganobu, M., and Sushin, V. 2004. The CCAMLR 2000 Survey: a multinational, multi-ship bi ological oceanography survey of the Atlantic sector of the Southern O cean. Deep-Sea Res earch II 51: 1205-1213. Welschmeyer, N. 1994. Fluorometric analysis of chlorophyll a in the presence of chlorophyll b and phaeopigments. Limnology and Oceanography 39: 1985-1992. Wiebe, P.H., Ashjian, C.J., Gallager, S.M., Davis, C.S., Lawson, G.L., and Copley, N.J. 2004. Using a high powered strobe light to increase the cat ch of Antarctic krill. Marine Biology 144: 493-502. Wright, S.W., Jeffrey, S.W., Mantoura, R.F.C ., Llewellyn, C.A., Bjor land, T., Repeta D., and Welschmeyer, N. 1991. Improved HPLC method for the analysis of chlorophylls and carotenoids from ma rine phytoplankton. Marine Ecology Progress Series 77: 183-196. Zapata, M., Rodriguez, F., and Garrido J.L. 2000. Separation of chlorophylls and carotenoids from marine phytoplankton: a new HPLC method using a reversed phase C8 column and pyridine-conta ining mobile phases. Marine Ecology Progress Series 195: 29-45. Zar, J.H. 1984. Biostatistical Analysis Prentice Hall Inc., New Jersey. 717pp. Zhang, C., Hu, C., Shang, S., Mller-Karger, F.E., Li, Y., Dai, M., Huang, B., Ning, X., and Hong, H. 2006. Bridging between SeaWiFS and MODIS for continuity of chlorophyll a concentration assessments o ff Southeastern China. Remote Sensing of Environment 102: 250-263. Zhou, M., and Dorland, R. 2004. Aggregation and vertical migration behavior of Euphausia superba Deep-Sea Research II 51: 2119-2138. Zhou, M., Zhu, Y., and Peterson, J.O. 2004. In situ growth and mortality of mesozooplankton during the austral fall a nd winter in Marguerite Bay and its vicinity. Deep-Sea Research II 51: 2009-2118.
A ( r m A ppendix 1. V r ight colum n m aximum sa m V ertical dis t n ) in Margu e m pling dept h t ribution of e e rite Bay du r h when a ne t Nor t Sou t 146 e uphausiids r ing fall 200 t haul was s a Crystal So u t hern Laube u t hern Laube u (left colum n 1. The blac k a mpled to d e u nd u f Fjor d u f Fjor d n ) and cope p k broken lin e e pths shallo w p ods (ind m 3 e indicates t h w er than 50 0 3 ) h e 0 m.
A A ppendix 1. ( ( Continued) B Nort h 147 B ourgeois F j Neny Fjo r h ern Alexan d j or d rd d er Islan d
A A ppendix 1. ( ( Continued) Nort h Nort h 148 h ern Alexan d h ern Alexan d Lazarev B a d er Island d er Islan d a y
A A ppendix 1. ( ( Continued) North South South 149 ern George V ern George V ern George V V I Soun d V I Soun d V I Soun d
150 Appendix 2. Vertical distribu tion of euphausiids (left co lumn) and copepods (ind m-3) (right column) in Marguerite Bay during fa ll 2002. The black broken line indicates the maximum sampling depth when a net haul was sampled to depths shallower than 500 m. Color legend as in Appendix 1. Crystal Sound Crystal Sound Northern Laubeuf Fjord
151 Appendix 2. (Continued) Southern Laubeuf Fjord Southern Laubeuf Fjord Northern Alexander Island
152 Appendix 2. (Continued) Northern Alexander Island Northern Alexander Island Marguerite Trough
153 Appendix 2. (Continued) Marguerite Trough Southern Adelaide Island Inner Marguerite Bay
154 Appendix 3. Vertical distribu tion of macrozooplankton (ind m-3) in Marguerite Bay, in relation to environmental parameters. Pigment (mg m-3) represents chlorophyll + phaeopigment concentrations. Southern Laubeuf Fjord, fall 2001
155 Appendix 3. (Continued) Bourgeois Fjord, fall 2001
156 Appendix 3. (Continued) Southern George VI Sound, fall 2001
157 Appendix 3. (Continued) Crystal Sound, fall 2002
158 Appendix 3. (Continued) Southern Laubeuf Fjord, fall 2002
159 Appendix 3. (Continued) Inner Marguerite Bay, fall 2002
160 Appendix 4. Vertical distribu tion of macrozooplankton (ind m-3) in Marguerite Bay, in relation to environmental parameters. Pigment (mg m-3) represents chlorophyll + phaeopigment concentrations. Northern Laubeuf Fjord, fall 2001
161 Appendix 4. (Continued) Northern Laubeuf Fjord, fall 2002
ABOUT THE AUTHOR Marina Marrari received a degree in Bi ological Sciences from the Universidad Nacional de Mar del Plata, Argentina, in 2001. She did her thesis research at the Zooplankton Laboratory of th e Instituto Nacional de I nvestigacin in Desarrollo Pesquero (INIDEP) under the guidance of Dr. Mara Delia Vias, and later worked at Scripps Institution of Oceanography for severa l months in 2002. During this time, Marina participated in numerous research cruises in continental shelf waters off the coast of Argentina and California. Marina came to the College of Marine Scie nce of the University of South Florida as a Fulbright Scholar in the fall of 2002, and joined the Zooplankton Ecology Laboratory led by Dr. Kendra Da ly. While in the Ph.D. program, she participated in several international research meetings a nd coauthored four publications in peerreviewed scientific journals.
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 002001128
007 cr mnu|||uuuuu
008 090429s2008 flu s 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0002598
Characterization of the western Antarctic Peninsula ecosystem :
b environmental controls on the zooplankton community
h [electronic resource] /
by Marina Marrari.
[Tampa, Fla] :
University of South Florida,
Title from PDF of title page.
Document formatted into pages; contains 161 pages.
Dissertation (Ph.D.)--University of South Florida, 2008.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
ABSTRACT: The zooplankton community of Marguerite Bay, western Antarctic Peninsula, was investigated in relation to variability in chlorophyll concentrations and sea ice dynamics, using a combination of satellite remote sensing techniques and plankton net data. SeaWiFS chlorophyll data were validated with concurrent in situ data measured by HPLC and fluoromentric methods, and results indicate that SeaWiFS chlorophyll is an accurate measure of in situ values when HPLC data are used as ground truth. Climatology data of SeaWiFS chlorophyll west of the Antarctic Peninsula showed that the Bellingshausen Sea and Marguerite Bay usually had higher and more persistent chlorophyll concentrations compared with northern regions. These predictable phytoplankton blooms could provide the Antarctic krill, Euphausia superba, with the food required for successful reproduction and larval survival.^Unusually high krill reproduction in 2000/2001 was coincident with above-average chlorophyll concentrations throughout the study area and was followed by the largest juvenile recruitment since 1981. High larval densities at the shelf break along the Antarctic Peninsula may have resulted, in part, from krill spawning in the Bellingshausen Sea. Interannual differences in sea ice also probably contributed to the variability in larval krill abundances. Interannual differences were observed in the species composition of the zooplankton of Marguerite Bay during fall, and these were linked to variability in the environmental conditions. Thysanoessa macrura was the most abundant euphausiid in 2001, while Euphausia crystallorophias dominated in 2002, and E. superba had intermediate densities during both years.Copepods were more abundant in 2001 by a factor of 2.6. Copepods and T. macrura showed a rapid population response to unusually high chlorophyll concentrations in the Bellingshausen Sea and Marguerite Bay during spring-summer 2000/2001, whereas E. superba and E. crystallorophias had a longer term response and showed increased recruitment in fall 2002. There were no clear associations between the distribution of zooplankton and environmental conditions in fall; however there was a significant relationship between chlorophyll concentrations in the Bellingshausen Sea during the preceding spring and zooplankton patterns during fall.
Mode of access: World Wide Web.
System requirements: World Wide Web browser and PDF reader.
Advisor: Kendra L. Daly, Ph.D.
x Marine Science
t USF Electronic Theses and Dissertations.