23 April to 6 June 2001

Available from

U.S. Southern Ocean GLOBEC Planning Office

Center for Coastal Physical Oceanography

Crittenton Hall

Old Dominion University

Norfolk, VA 23529

Sponsored by Office of Polar Programs, National Science Foundation

LMG01-03 report prepared by Richard Limeburner, Robert Beardsley, Mark MacDonald, Sue Moore, and Deborah Thiele

This cruise report was prepared by Richard Limeburner, Robert Beardsley, Mark MacDonald, and Deborah Thiele. We especially thank Captain Sanamo and the officers and crew of the R/V Gould Cruise LMG01-03 for their skill and superb assistance, which allowed deployment of the moorings and drifters under difficult weather conditions, observations of the marine mammals, and completion of the hydrographic survey. Often, the moorings were deployed in heavy seas among icebergs and the ship’s officers were still able to navigate the ship to specified mooring deployment sites.

Special thanks to Skip Owens of Raytheon Polar Services (RPS) for coordinating the successful mooring deployment operations. The successful mooring planning, design, and deployments were primarily due to the efforts of Scott Worrilow, with assistance from Ryan Schrawder and Jim Ryder. Thanks also to Andy Nunn, Jenny White, and Jonette Tuft of RPS for helping on deck and maintaining and operating the shipboard scientific equipment and coordinating this instrumentation with the scientific party. Thanks to Claudia Cenedese for helping with the CTD stations and mooring deployments. This research effort is sponsored by the National Science Foundation, NSF research grant OCE-99-10092. All data and results in this report are to be considered preliminary.

Acknowledgements 4

1. Purpose 6
2. Accomplishment Summary 6

3. Cruise Results 7

3.1 Bathymetric Survey 7

3.2 Mooring Deployments 8

3.3 Hydrographic Data 10

3.3.1 Calibration 10

3.3.2 CTD Data 10

4. Meteorological Measurements 15

4.1 Introduction 15

4.2 Instrumentation 16

4.3 Data Acquisition and Processing 16

4.4 Description of Cruise Weather 19

5. Marine Mammal Observations 21

5.1 Acoustic Census of Mysticete Whales 21

5.2 Visual Survey, Photo ID And Biopsy Summary 22

6. Chief Scientist’s Log of Daily Events 23

7. Cruise Personnel 26

Appendix 1 Bathymetric Surveys of the Mooring Sites 28

Appendix 2 Drifter tracks 32

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

1. Purpose

The primary purposes of cruise R/V Lawrence M. Gould (LMG01-03) were to deploy a Woods Hole Oceanographic Institution (WHOI) array of six current meter moorings near Marguerite Bay, deploy a Scripps Institute of Oceanography (SIO) moored array of eight whale acoustic recording packages (ARPS) along the western Antarctic Peninsula, deploy six near-surface satellite-tracked drifters, and to conduct a marine mammal survey of the Marguerite Bay region. This report summarizes the events that occurred during cruise LMG01-3 on the R/V Gould.

A central hypothesis of the Southern Ocean Global Ocean Ecosystems Dynamics (SO GLOBEC) collaborative research program is that a unique combination of physical and biological factors contribute to the enhanced growth, reproduction, recruitment, and survivorship of Antarctic krill (Euphausia superba) on the central western Antarctic Peninsula (WAP) shelf. In particular, this region provides the following conditions that are especially favorable to winter survival of larval and adult krill: a) a clockwise shelf circulation that retains the krill population in a favorable environment for extended periods of time; b) an early and long-lasting ice cover that provides dependable food and protection for larval krill to grow and survive over winter; and c) on-shelf intrusions of warm, salty, nutrient-rich Upper Circumpolar Deep Water, which affect hydrographic and ice properties and enhance biological production. The moored array, drifter, and float component of the Southern Ocean GLOBEC program is investigating shelf circulation processes and their spatial and temporal variability using long-term moorings and satellite-tracked Lagrangian drifters and isobaric floats. Supporting data on the surface forcing (wind stress and heat flux) will also be obtained and the combined data set used to describe the shelf circulation and water property variability on vertical scales of 10s of meters and time scales from hourly to seasonal.

 

2. Accomplishment Summary

The severe Antarctic weather was a potential limiting factor during cruise LMG01-03. Winds were often greater than 20 kts during the cruise and all science operations were halted when the winds were greater than 35 kts. However, only one day was spent waiting for an improvement in weather conditions. During the cruise, we successfully deployed the six current meter and eight whale acoustic moorings along the western Antarctic Peninsula, deployed six near-surface drifters, made marine mammal observations of the region, and made six conductivity-temperature-depth (CTD) casts at the WHOI moorings. The LMG01-03 cruise track is shown in Figure 1 and mooring deployment locations are shown in Figure 2.



Figure 1. LMG01-03 cruise track

.

3. Cruise Results

We departed Punta Arenas, Chile at 1330 UTC on 18 March 2001, accompanied by 40 kt winds and steamed east through the Straits of Magellan, and then south to the Drake Passage. The passage to Antarctica took four days and the seas were typically rough with 20 to 40 kt winds but relatively mild air temperatures. We arrived at King George Island on 23 March and offloaded two scientists and equipment by evening, then steamed to the northern whale acoustic mooring site S8, where we deployed the mooring in rough seas. The R/V Gould then steamed to Palmer Station and arrived early on 24 March. The day at Palmer Station was spent offloading personnel and supplies and repositioning the mooring anchors and buoys on the ship in preparation for the upcoming mooring deployments.

1. Bathymetric Surveys

Digital depth data was collected with a Knudsen fathometer from Punta Arenas to Palmer Station, using a sound speed of 1500 m s^-1. The sound speed, based on CTD data collected in Marguerite Bay during the January 2001 Long-Term Ecological Research cruise, was changed to 1459 m s^-1 on 25 March just before leaving Palmer Station. The sound speed was changed to 1449 m s^-1 on 28 March at 1949 UTC and changed back to 1459 m s^-1 on 29 March at 2000 UTC. We made a small-scale bathymetric survey at each of the mooring sites prior to deployment and the results of these surveys are shown in Appendix 1.

Figure 2. Locations of the moorings and drifters deployed on LMG01-03.

The R/V Gould departed Palmer Station at 1400 UTC on 25 March 2001 and steamed to the whale acoustic mooring site S7. The mooring was successfully deployed at night under rough seas. A CTD station planned for the S7 site was cancelled due to 35 kt winds and the rough seas. The ship then arrived at the A1 mooring site at 1200 UTC on 26 March and we first made a detailed bottom survey of the local bathymetry. The moorings are designed for a specific depth and the A1 mooring site had very rough bathymetry, but a suitable site was found for the deployment (Appendix 1). The mooring locations are shown on Figure 2 and listed in Table 1.

The A1 mooring was deployed on 26 March under difficult working conditions. Strong winds were diminishing and the wind was from a different direction than the swells. The deck was awash during the deployment and we were informed that those conditions were as good as we should expect. After the deployment, we completed CTD 1 at the A1 mooring site and deployed a near-surface satellite tracked drifter.

Table 1. LMG01-03 mooring and CTD station locations.

During the night of 26 March, we steamed to the whale-acoustic mooring site S8 and made a brief bottom survey of the region before deploying the S8 mooring. The bathymetry at the S8 site is shown in Appendix 1. After the S8 mooring deployment, the barometer began to fall and the winds increased in strength. At sunrise on 27 March, the winds were over 50 kts from the northeast and the barometer, at 950 mb, was dropping 3 mb hr^-1. We spent the day at the A2 mooring site, waiting for the weather to improve. At 2000 UTC, the winds were not decreasing, the barometric pressure was 948 mb and holding, and the seas were building to over 6 meters, so the Captain decided to steam into Marguerite Bay and find some shelter from the seas.

On the morning of 28 March, we were in Marguerite Bay, off the south end of Adelaide Island. The winds were still above 40 kts and the temperature had fallen to - 5° C, but the seas were more moderate than offshore. We decided to bottom survey the three mooring sites located along the B line. Target mooring deployment positions were defined from the bathymetric surveys at the B line (Appendix 1).

The morning of 29 March was clear with 10 kt winds and mooring B3 was deployed at 1242 UTC. We then made a CTD cast, CTD 2, at the B3 mooring site. The B2 mooring was deployed at 2018 UTC on 29 March and CTD 3 was made at the B2 mooring site. The seas were still calm, so we decided to deploy a third mooring that day and mooring B1 was deployed in the evening and CTD 4 was made at the B1 mooring site. During the night of 29 March, we deployed three drifters near the B mooring line while steaming to the A line.

On the afternoon of 30 March, we made a bathymetric survey of the A2 mooring site. CTD 5 was completed in the late afternoon at the A2 mooring site and we then deployed mooring A2 and a drifter. That evening, we made a bathymetric survey of the A3 mooring site and, on the following mooring, we completed CTD 6 and deployed A3. During the afternoon of 31 March, we began deploying the remaining five SIO whale acoustic moorings. Moorings S5 and S6 were deployed on 31 March, S4 and S3 on 1 April, and S2 on 2 April. The mooring and CTD locations are listed in Table 1.

The ship’s CTD system consisted of a Seabird Electronics (SBE) Model 911⁺ CTD sampling at 24 Hz with a quartz crystal pressure transducer, two temperature and conductivity sensors pairs, a Photosynthetically Active Radiation (PAR) sensor, a fluorometer sensor, and a SeaTech 25 cm light transmission sensor. The CTD fish was lowered at 30 m min^-1 and the data were logged with DOS Seabird Electronics software. The CTD data was averaged into 1-m bins and the averaged downcast data is presented in this report. The two conductivity sensors on the CTD agreed with each other to within 0.003 mS cm^-1, and thus only one of these sensors was used to process the data in this report.

Approximately four water samples were collected at each station for calibration of the conductivity sensors on the SBE-911⁺ CTD. Figure 3 summarizes the comparison between the conductivity measured by the CTD and the in situ conductivity calculated from the measured salinity obtained from the water sample bottle using a Guildline Salinometer.

The RMS difference in conductivity between the CTD and the water sample bottles was 0.0048 and no correction was made to the raw conductivity data. All CTD data in this report will be transmitted to the National Oceanographic Data Center.

Temperature, salinity and sigma-t profiles and temperature-salinity correlations are shown next for the six CTD casts (Figures 4-9).

Figure 3. Differences between the CTD conductivity and the

seawater sample conductivity during the cruise.

Figure 4. CTD 1 profiles located at WHOI mooring A1.

Figure 5. CTD 2 profiles located at WHOI mooring B3.

Figure 6. CTD 3 profiles located at WHOI mooring B2.

Figure 7. CTD 4 profiles located at WHOI mooring B1.

Figure 8. CTD 5 profiles located at WHOI mooring A2.

Figure 9. CTD 6 profiles located at WHOI mooring A3.

1. Meteorological Measurements

4.1 Introduction

A good knowledge of the surface meteorological conditions during the U.S. SO GLOBEC program is essential to understand the role of surface momentum and heat flux forcing on the regional circulation and upper ocean properties. The surface meteorological data are also useful in interpreting other physical and biological data collected during the program.

The primary source of surface meteorological data during U.S. SO GLOBEC will be that collected aboard the L.M. Gould and N.B. Palmer during their cruises in the study area. LMG01-03 is the first U.S. SO GLOBEC cruise, and this report provides a preliminary description of the meteorological data collected on this cruise.

The L.M. Gould left Punta Arenas (PA) on 18 March and arrived at Palmer Station on 24 March (Leg 1). The Gould left Palmer Station on the next day and returned to Palmer Station from the U.S. SO GLOBEC area to the south on 8 April (Leg 2). The L.M. Gould left Palmer Station on 9 April for PA and arrived on 13 April (Leg 3). A full suite of meteorological data was collected during the cruise with two exceptions. Neither sea surface temperature (SST) nor sea surface salinity (SSS) data were collected while the L.M. Gould was docked at Palmer Station. The ship uses GMT year day (YD) as given by the Global Positioning System (GPS) for time. Leg 2 corresponds to YD = 84.61 to 98.52.

4.2 Instrumentation

The meteorological sensors are mounted on the ship’s main mast (Figure 10). The sensors include a pair of wind monitors and other sensors to measure air temperature (AT), relative humidity (RH), barometric pressure (BP), and incident shortwave (SW) and longwave (LW) radiation. Sea surface temperature (SST) was measured using a remote sensor in the intake manifold, and sea surface salinity (SSS) was measured using a thermosalinograph placed in the wet lab. The different sensors and their calibration history and installation dates are given in Table 2. Based on this table, the port wind monitor and the three radiation sensors (PAR, SW, and LW) are past their next scheduled calibration. In particular, the SW and LW sensors have not been recalibrated since September 1997.



Figure 10. Meteorological sensors mounted on platform railing on top of mast.

 

 

4.3 Data Acquisition and Processing

The raw L.M. Gould shipboard meteorological data were collected using the ship’s data acquisition system (DAS). A 1-minute processed subset of the raw data was saved at the end of each day in a flat ASCII text file on the ship’s DAS_DATA directory on drive Q (e.g., the data for YD=99 and YD=100 are located in Q:\.geopdata\JGOF\jg099.dat and jg100.dat respectively). This 1-minute time series was produced using a Joint Global Ocean Flux Study (JGOFS) code that merged the meteorological data with navigation and other data and combined the ship’s motion and the measured (relative to the ship) wind speed and direction data to make "true" wind speed and direction relative to the ground.

Table 2. LMG01-03 meteorological sensors, their calibration history, and

time of installation. The last column indicates if the sensor is to be

re-calibrated every year (A) or every two years (BA).

 

Variable	Sensor	Serial Num.	Last Cal.	Next Cal.	Installed	Cal. Inv.
Star. Wind	RM Young 5106	WM 28393	11/07/00	11/07/01	12/28/00	A
Port Wind	RM Young 5106	WM 35061	2/04/99	2/04/00	8/26/99	A
AT, RH	RM Young 41372VC	2904	11/30/99	11/29/01	7/23/00	BA
BP	RM Young 61201	BP 00873	11/30/99	11/29/01	7/22/00	BA
PAR	Biosp. Inst. QSR-240P	6394	3/22/99	3/21/00		A
SW	Eppley PSP	28933F3	9/21/97	9/21/98		A
LW	Eppley PIR	32031F3	9/03/97	9/03/98		A
SST	Sea-Bird 3-01S	1619	7/27/99	7/26/00		A
SSS	Sea-Bird 21	219209-1577	12/30/99	12/29/00		A

The daily data were obtained from drive Q and converted into standard variables using the MATLAB m-file read_gould_met1m(YD). This program also edits the raw SST data and stores the final data set in a MATLAB mat-file for each day (e.g., jg100.mat for YD=100). The raw SST time series contains many values that were below the surrounding values by ~ 0.2^oC to 0.5^oC or more. These data dropdowns occurred through the cruise, with the amplitude and frequency of the dropdowns varying but the character of the dropdown remaining the same. A simple filter was used (clean_gould_sst.m) to replace the dropdowns with linear interpolation. An example of the raw and edited SST is shown in Figure 11. Any additional hand-editing of the SST and sea surface salinity (SSS) time series was done at this point and the final edited series stored in each day’s mat-file (e.g., jg100.mat).

The m-files start_gould_met1m(first_YD) and merge_gould_met1m(last_YD) were used to combine the 1-day jgxxx.mat files into a single 1-minute continuous time series for each variable. The merged data were then stored in gould_met1m.mat.

For further analysis, the 1-minute data in gould_met1m were low-pass filtered and subsampled using make_gould_met5m into 5-minute time series. The filter used is the pl66tn set with a half-amplitude period of 12 minutes. Both the shortwave radiation and PAR time series were corrected for small negative nighttime biases. The 5-minute data were then used to estimate the surface wind stress and heat flux components using bulk methods called by compute_gould_wshf5m. Two gaps in SST (when the ship was at the Palmer Station dock) were filled using linear interpolation so that the surface fluxes could be computed for the entire period of the cruise. The errors introduced by using the approximate SST at Palmer Station are small enough to ignore. The surface wind stress and heat flux data were then added to the gould_met5m, so that this 5-minute contains best versions of the surface meteorological conditions and forcing for the cruise.

 

Figure 11. Raw (blue) and edited (red) SST time series for a 2.4-hour period on YD 100.

 

As mentioned above, the JGOFS data file jgxxx.dat placed on drive Q presents only the "true" wind speed and direction. The actual raw wind measurements are stored each day in another directory and will be included in the cruise data CD-ROM. Without that data, it is difficult to make an accurate assessment of the JGOFS code that removes the ship’s motion from the relative wind measurements made on the mast. A preliminary look can be made by comparing wind speed and the ship’s speed over the ground (SOG), to see if sudden changes in SOG are reflected in the wind speed. The example shown in Figure 12 is not definitive but does suggest that the JGOFS "true" wind still contains some ship motion. A full analysis of the raw data will be needed to check the JGOFS code and see if a better code for "true" wind can be developed. Until this analysis is done, the wind speed and direction data and those variables that are functions of wind speed (e.g., wind stress) contained in gould_met1m and gould_met5m must be considered as preliminary values.

 

 

 



Figure 12. Plot of ship’s speed over the ground (SOG) from GPS in red and the

JGOFS "true" wind speed. Note how the wind speed tends to jump when the

ship stops and starts.

4.4 Description of Cruise Weather

Time series of the 5-minute surface meteorological data are shown in Figure 13 for the entire cruise. During Leg 2 (YD=84.61 to 98.52), a deep low pressure system passed over the region on YD 87, with strong winds above 40 kts towards the southwest to south for almost two days (peak winds exceeded 50 kts for several hours). As the low pressure system passed, the winds dropped and a short 3-day period (YD 88-91) followed with winds 10-20 kts towards the north and northeast. This good weather window allowed the completion of the WHOI and SIO mooring deployment work. Winds became strong again on YD 91-93, with maximum winds between 30 and 40 kts during YD 92. The L.M. Gould spent the rest of Leg 2 searching for and studying whales in the inland passages and inside Marguerite Bay until heading back to Palmer Station on YD 97. During Leg 2, the skies were usually overcast, with occasional snow, so the downward flux of shortwave radiation was generally small: mean 28 W m^-2, with a maximum of 296 W m^-2 on YD 94. The downward flux of longwave radiation dropped from values near 280-300 W m^-2 to roughly 200 W m^-2 during the few periods with clear skies. Using this drop as an indicator of clear skies, there were clear skies for only about 12% of Leg 2.



Figure 13. Surface meteorological measurements during LMG01-03.

2. Marine Mammal Observations

5.1 Acoustic Census of Mysticete Whales

Most large whales produce distinctive calls such that passive acoustic methods can be used to study their distribution, behavior, and minimum absolute abundance. The repetitive call of the ordinary blue whale is ideal for long-range signal transmission. Although blue whale calls vary by geographic region, each population produces a characteristic low-frequency (~20 Hz) repetitive call of roughly 10 to 20 seconds duration, sometimes with multiple components. The detection range for blue whale sounds from a simple bottom recorder is typically 20 km for a reliable consistent detection, and longer if one applies matched filter techniques, which require knowledge of the call character. The long-term nature of seafloor recorder deployments allows for a statistically significant number of acoustic encounters even with a widely dispersed whale population, assuming whales call roughly 10-50% of the time. The recordings will likely also include sounds from minke, right, fin, and humpback whales.

Mysticete whales will be detected via reception of their calls on passive, bottom-mounted acoustic recorders. Detection of whale calls via moored passive acoustic recorders has proven quite effective during recent studies, especially for blue and fin whales. New technology, that of long-term deployments of autonomous low power recorders, makes an acoustic survey of mysticete whales in remote locations practical. Deep water is desirable partly because the ambient noise, which is largely produced at the surface, is reduced at depth and also because acoustic travel paths will interact less with the seafloor, which absorbs acoustic energy.

Given the great uncertainty in the numbers of blue whales in the Southern Ocean (500 to 5000), and in the subspecies to which they belong, we believe the minimum census estimates which can be provided by acoustic monitoring is a key goal of the proposed project. If we were to have 800 acoustic contacts per season, for instance, it would be clear that the International Decade of Cetacean Research (IDCR) population estimates are severely underestimating the actual populations. Even in the case where we would obtain 80 acoustic contacts per season, in keeping with the IDCR estimates, we should be able to provide some estimate of which geographic areas and/or subspecies the animals are from. Application of the techniques of point transect theory to the results of the survey where each acoustic contact is assigned a range should allow a minimum census estimate, the primary factor which will remain to be answered from other combined visual and acoustic data being the percent of whales calling during a time constant.

The second fundamental goal of this work will be minimum population estimates and seasonal occurrence profiles for fin and humpback whales. Other species, such as minke and sperm whales, may be detected but are expected to be so infrequent as to make population density estimates unreliable. Perhaps the most important overall result of this work will be to establish an acoustic detection baseline from which to measure future changes in relative abundance of Southern Ocean mysticete whales.

Throughout the cruise, we deployed and recorded 36 directional sonobuoys and two broadband sonobuoys, both randomly and when whales were sighted, as a means to record and thereby "groundtruth" mysticete whale calls in this remote region. Recordings were obtained from humpback, minke, and fin whales during the cruise. No blue or right whales were sighted or heard during this cruise.

We have deployed eight seafloor acoustic recorders, which each record a hydrophone that is floated about 5 m above the seafloor continuously at 500 samples per second for 15 months, writing the data to 36 gigabytes of computer hard drive in each instrument. The recorders have a 16-bit dynamic range and are powered by lithium double-D size batteries, which are placed inside high tensile aluminum pressure cases. The seafloor recorders use a system of drop weights, benthos glass balls, and an acoustic release for recovery.

5.2 Marine Mammal Visual Survey, Photo ID, And Biopsy Summary

A visual survey for cetaceans was conducted on LMG01-03 during daylight hours on all days when weather conditions allowed. Searching was conducted by the International Whaling Commission (IWC) observer (Deborah Thiele) and Sue Moore, with assistance from the rest of the acoustic mooring team. Sightings were recorded on a laptop-based Wincruz Antarctic program, which also logged GPS position, course, and ship speed automatically. Seals, seabird concentrations, ice concentration, SST, and depth were also recorded on the program. Survey effort generally commenced at first light from the outside bridge wings and/or inside the bridge (weather dependent) and ceased at dark. Visual surveys alone were conducted during the oceanographic and acoustic mooring components of the cruise. Large numbers of humpback whales were observed during the afternoon transit across the Bransfield Strait. Five days of ship time were allocated to conduct cetacean survey (including closing on sightings, biopsy, photo identification, and sonobuoy deployment) from first light on Tuesday, 3 April to the end of day on Saturday, 7 April. The weather improved for this part of the survey, with calm seas and a lifting of the Peninsula cloud blanket, providing us with some rare sunny days! During this time, the ship traversed the inside channels to the north of, and including, Adelaide Island. In Matha Strait, two large feeding groups of the dark shouldered minke whale were sighted (total of 80 animals). Photographs were obtained and biopsies of four minkes from the largest group were taken. Upon entering the northern end of Marguerite Bay, just near Rothera Station, a large group (30) of killer whales was sighted. The killer whales broke into sub-groups and traveled north past the ship with many accompanying fur seals and a minke whale. One humpback was also seen in the vicinity. The ship took a southwest course to Neny Island and then headed across the Bay and towards Palmer Station on a course close to the coast and inside the islands from Matha Strait onwards. Two pairs of humpbacks were located late in the afternoon of 7 April, with photo identification and biopsy obtained from one pair. Total biopsies taken: 5 minke (one from a group of 7 and 4 from the feeding group of 50) and 5 humpback (one from one pair and both animals from another two pairs).

 

 

 

 

 

 

Table 4. LMG01-03 marine mammal observations.

Total species	Total sightings:animals
Fin whale	1:3
Minke ordinary	9:91
Like minke	5:14
Hourglass dolphin	1:4
Unidentified cetacean	1:1
Unidentified large whale	3:5
Killer whale	4:53
Ziphiidae	2:5
Humpback whale	38:81
Humpback whale and like fin whale (mixed group)	1:12
Undetermined minke whale	18:40
Like fin whale	1:3
Unidentified whale	4:4
Total	88:316

 

3. Chief Scientist’s Daily Log

All times local time, GMT-4

Sunday – 18 March 2001

Monday – 19 March 2001

Tuesday – 20 March 2001

Wednesday – 21 March 2001

Thursday – 22 March 2001, Day 81

Friday – 23 March 2001, Day 82

0306 Deployed SIO whale acoustics mooring S1 in a snowstorm

Steaming to Palmer Station

Saturday – 24 March 2001, Day 83

Sunday – 25 March 2001, Day 84

Monday – 26 March 26 2001

Tuesday – 27 March 27 2001

Wednesday – 28 March 28 2001

Winds > 40 kts, unsafe to work on deck

2000 Began bathymetry survey of B3 mooring site

Thursday – 29 March 2001

CTD 4

Deploy B1 mooring

Deploy drifter 26373

1. Deploy drifter 30459

Richard Limeburner Chief Scientist

Raytheon Polar Services

Dionito Sabinas Deck

Figure 1. Bathymetry at the A1 mooring site.

Figure 2. Bathymetry at the A2 mooring site.

Figure 3. Bathymetry at the A3 mooring site.

Figure 4. Bathymetry at the B1 mooring site.

Figure 5. Bathymetry at the B1 mooring site.

Figure 6. Bathymetry at the B3 mooring site.

Figure 7. Bathymetry at the S7 mooring site.

Figure 8. Bathymetry at the S8 mooring site.

Appendix 2. Drifter tracks

Report prepared by Jose Torres, Jennifer Burns, Kendra Daly, Bill Fraser, Sarah Marschall, Frank Stewart, and Meng Zhou, with considerable assistance from our colleagues in the science party, our Raytheon Polar Services crew, and the captain and crew of the R/V Lawrence M. Gould.

A grateful scientific party would like to extend its sincere thanks to our friends on the Lawrence M. Gould for making our cruise a successful one. To Skip Owen, Christian McDonald, and Josh Spillane for help on deck, in the Zodiacs, under the ocean, in the shop, with logistics and damn near everything else. To Sheldon Blackman and Bruce Felix for keeping our MOCNESS nets running, emails flowing, DAS system recording, and CTDs dunking. To Captain Warren Sanamo, chief mate Jesse Gans, 2^nd mate Tracy Ruhl, and 3^rd mate John Snyder for plain and fancy boat driving, even in unsurveyed areas, putting up with our schedule changes, and generally keeping us all safe. To Chief Engineer Mike Murphy, 1^st engineer Paul Waters, 2^nd engineer Russ Lesser, and our two oilers Noli Tamayo, and Donde Dasoy for keeping us afloat and moving, even when declutched on the starboard side. To Rudy Lucas, Romeo Agonias, and Luciano Albornoz for 45 days of great meals and great desserts. To Fernando Naraga and Roy Ninon for all their help on the winches, and Rafael Sabino and Dionito Sabinas for their general help on deck and keeping our boat spotless. We thank you one and all.

(see facing page)

Kneeling in front, from left to right: Joel Bellucci, Meng Zhou, Jose Torres

Standing, from left to right: Christian McDonald, Skip Owen, Frank Stewart, Sarah Marschall, Scott Polk, Yiwu Zhu, Steve Trumble, Michelle Grigsby, Joe Donnelly, Ryan Dorland, Hyoung-Chul Shin, Jennifer Burns, Mark Hindell, Beth Chittick, Kendra Daly, Aaron Morello, Bill Fraser, Sheldon Blackman, Dan Mertes

Far back, left to right: Bruce Felix, Chris Denker

Not present in photo: Josh Spillane, Chris Simoniello, Tracey Sutton

PURPOSE OF THE CRUISE 39

Objectives 39

Historical perspective 40

Departures from the original plan 40

Projects represented on the cruise 42

Cruise overview 42

PROCESS STATION 1 43 Synopsis 43

Individual group reports 44

BG-232-0 Jennifer Burns and Dan Costa - seal ecology 44

BG-236-0 Kendra Daly - Krill ecology and physiology 47

BG-245-0 Jose Torres - krill and fish ecology, krill physiology 47

BG-248-0 Meng Zhou - krill ecology, behavior, and modeling 48

PROCESS STATION 5 49

Synopsis 49

Individual group reports 51

BG-232-0 Jennifer Burns and Dan Costa - seal ecology 51

BG-236-0 Kendra Daly - Krill ecology and physiology 55

BG-245-0 Jose Torres - krill and fish ecology, krill physiology 55

BG-248-0 Meng Zhou - krill ecology, behavior, and modeling 55

PROCESS STATION 4 58

Synopsis 58

Individual group reports 60

BG-232-0 Jennifer Burns and Dan Costa - seal ecology 60

BG-236-0 Kendra Daly - Krill ecology and physiology 62

BG-245-0 Jose Torres - krill and fish ecology, krill physiology 63

BG-248-0 Meng Zhou - krill ecology, behavior, and modeling 63

PROCESS STATIONS 2 AND 3 64

Synopsis 64

Individual group reports 68

BG-232-0 Jennifer Burns and Dan Costa - seal ecology 68

BG-236-0 Kendra Daly - Krill ecology and physiology 74

BG-245-0 Jose Torres - krill and fish ecology, krill physiology 74

BG-248-0 Meng Zhou - krill ecology, behavior, and modeling 75

PROCESS STATIONS 6 AND 7 - END OF CRUISE 77

Synopsis 77

Individual group reports 79

BG-232-0 Jennifer Burns and Dan Costa - seal ecology 79

BG-236-0 Kendra Daly - Krill ecology and physiology 81

BG-245-0 Jose Torres - krill and fish ecology, krill physiology 82

BG-248-0 Meng Zhou - krill ecology, behavior, and modeling 83

CRUISE PARTICIPANTS 85

APPENDIX 1- Event log

APPENDIX 2- Map and table of CTD locations

PURPOSE OF THE CRUISE

The overall goal of the Southern Ocean Global Ocean Ecosystem Dynamics (SO GLOBEC) program is to elucidate shelf circulation processes and their effect on sea ice formation and Antarctic krill (Euphausia superba) distribution, and to examine the factors that govern krill survivorship and availability to higher trophic levels, including seals, penguins, and whales.

The SO GLOBEC field effort in its first year consisted of a series of five cruises within the fall-winter time frame. The initial cruise during late March-early April (LMG01-03) was a stand-alone effort to deploy current meter moorings and passive acoustic devices for monitoring marine mammal populations. Moorings were successfully deployed across the mouth of Marguerite Bay on the western Antarctic Peninsula shelf. The program quickly followed the mooring cruise with a series of two cruises using two vessels working in tandem, sampling in a survey grid centered on Marguerite Bay (Figures 1A and B). Cruises took place during the austral fall period from late April through May (LMG01-04, NBP01-03), and during the winter from late July through August (LMG01-06, NBP01-04). The two vessels working in tandem had two distinctly different missions. The survey vessel was primarily charged with conducting a broad scale survey to provide a shelf-wide picture of hydrography, circulation, and krill distribution. The process cruise was designed to provide detailed information on selected sites within the survey grid (see Figure 1).

This is the report of the first process cruise (LMG01-04) which took place from 20 April to 5 June 2001. We were the sister cruise to the fall survey, NBP01-03, which is described in U.S. Southern Ocean GLOBEC Report Number 2. Our mission was to examine in detail several elements of the biology of the target species, Euphausia superba, at a series of approximately five previously chosen process stations nested within the survey grid. Our specific objectives were:

The original five process sites as depicted in Figure 1B were chosen either because they represented a habitat type, or because they were associated with important elements of bottom topography, or both. Process site 1 was located at the shelf break along the axis of the main across-shelf canyon that runs from the shelf break through the mouth of Marguerite Bay and into George VI Sound, an embayment at the southeastern corner of Marguerite Bay. Process site 1 allowed us to sample a typical oceanic fauna at the oceanic depths seaward of the shelf break, and by moving inshore, at depths more typical of the shelf. As such, it really had two parts: "A and B". Both sampling sites were at the mouth of the Marguerite Bay across-shelf canyon, a topographic feature that the program considered an important potential conduit for bringing oceanic water and fauna deep into the bay.

Site 2 was located in the mouth of Marguerite Bay over the axis of the across-shelf canyon. It allowed access to depths typical of the shelf (500 m) and the greater depths typical of the canyon itself (800-1000 m). Site 3 was located well south of Marguerite Bay at what was believed to be a potential site for newly forming sea ice. It was designed to be representative of the mid-shelf environment.

Process stations 4 and 5 were both created to sample habitats deep within the Bay. The location of site 4 gave the program access to the fast ice at the southern end of George VI Sound. At the same time it allowed sampling in the southernmost leg of the Marguerite Bay across-shelf canyon. Site 5 allowed access to the deep fjords east of Adelaide Island: areas where both seals and penguins were believed to spend time in the fall and winter.

Weather, personnel, and scientific considerations modified the original station plan. The first and most important modification was the order in which the stations were occupied. Process site 1 was completed first, as planned, but the stations were then occupied in the following order: 5, 4, 2, and 3. Two new abbreviated stations were created: 6 and 7. Process station 3 was moved from its original mid-shelf location to a new site in Lazarev Bay.

The main reason for our change in station timing was to allow our predator groups access to the substrate (glacial rubble, islands and shoreline) they needed to complete their work since annual sea ice was absent throughout the study area. Winds in excess of 35 kts for much of the first three weeks of the cruise were also a factor. Had we not been able to sample effectively in the smaller embayments within the study area that, happily, had been chosen as process stations 4 and 5 (and later, station 3 by the Gould science party) about 70% of the cruise would have been lost to bad weather. A listing of the activities that occurred at each process station is given in Appendix 1 and Appendix 2 gives a listing of the CTD stations occupied during LMG01-04.

Figure 1. Proposed locations of A) survey stations and B) process sites for U.S. SO GLOBEC. The arrows indicate the order in which the process sites were to be occupied. The survey station locations were modified as described in U.S. Southern Ocean GLOBEC Report Number 2. Modifications to the process site locations are described in this report.

BG-232-0 Dan Costa and Jennifer Burns - seal ecology

BG-234-0 Bill Fraser - seabird ecology

BG-235-0 Chris Fritsen - SIMCO and water column phytoplankton communities

BG-236-0 Kendra Daly - Krill ecology and physiology

OG-238-0 Laurie Padman - mesoscale circulation and hydrography

OG-240-0 Eileen Hofmann - circulation, hydrography, modeling

BG-245-0 Jose Torres - krill and fish ecology, krill physiology

BG-248-0 Meng Zhou - krill ecology, behavior, and modeling

20 April 01 RV L.M. Gould departed Punta Arenas, Chile

25 April 01 RV L.M. Gould arrived at Livingston Island. Lost a day to weather.

26 April 01 Dropped off Dr. Hall's team began transit to Palmer Station.

27 April 01 Arrive Palmer Station.

28 April 01 Depart Palmer Station for study site.

29 April 01 Began sampling at process station 1.

12 May 01 Concluded sampling at process station 5, transited to process station 4.

13 May 01 Arrived process station 4 and initiated sampling.

17 May 01 Departed process station 4.

18 May 01 Arrived process station 2 and initiated sampling.

20 May 01 Departed process station 2 due to extreme weather.

21 May 01 Arrived process station 3, Lazarev Bay.

25 May 01 Departed process station 3.

26 May 01 Arrived in George VI Sound for penguin survey in A.M.

27 May 01 Arrived Neny Fjord A.M.

28 May 01 A.M. Arrived Day Island for seal survey.

29 May 01 A.M. Seal survey , HTI drift station.

29 May 01 P.M. MOC 10, MOC 1.

30 May 01 01:30 End of science. Transit to Palmer Station.

31 May 01 08:00 Arrive Palmer Station.

31 May 01 13:30 Departed Palmer Station - transited to Livingston Island.

01 June 01 08:00 Arrive Livingston Island for pick up of field party (O-196: Dr. Brenda Hall).

01 June 01 11:30 Departed Livingston Island - transit to Punta Arenas, Chile.

05 June 01 08:00 Arrived in Punta Arenas, Chile.

Process station 1 was located at the Western Antarctic Peninsula (WAP) shelf edge in line with the axis of the main across-shelf canyon that runs from the shelf break into George VI Sound (Figure 2). It was bounded by the following coordinates: 66.411° S 70.637° W, 66.0453° S 71.5988° W, 66.6551° S 71.2118° W, 66.2718° S 72.1079° W. The station was initiated with an 18-hour ADCP survey that covered off-shelf and on-shelf locations. Discussions among the science party led us to divide process station 1 into an off-shelf (site A) and on-shelf (site B) component (Figure 3). We spent 3 days sampling off-shelf at depths of about 3000 m and 2 days on-shelf at depths of about 500 m (Figure 4). A typical sampling day was as follows:

0000-1030 MOC 1 (Zhou)

1100-1200 CTD (Fritsen Group)

1230-1430 LIVE NETS (Daly/Torres)

1430-1530 HTI (Daly)

1530-2130 MOC 10 (Torres)

2130-2330 LIVE NETS

Figure 2. Map showing the cruise track comprising process station 1. Note position of process station 1 relative to Marguerite Bay and Adelaide Island. Map created by Bruce Felix.

Figure 3. Map showing locations of process stations 1A and 1B within station 1 cruise track. Map generated by Meng Zhou.

Winds were moderately high during most of our occupation of process station 1. Twenty-five to 30 kt wind speeds were encountered daily and seas hovered at about 4 to 5 m much of the time. Despite the rough weather a considerable amount of sampling and laboratory experiments were successfully accomplished.

It should be noted that process station 1 was an open water station so that our two predator groups, Costa/Burns (BG 232-0) and Fraser (BG 234-0) were unable to take any samples. Their data acquisition will begin with our early occupation of process station 5.

Individual group reports.

1. BG 232-0 (Burns)

We arrived in Punta Arenas, Chile on 18 April and loaded onto the L.M. Gould without too much difficulty. All the essential gear was in place, and the group arrived in Chile largely intact. Mark Hindell lost his luggage in Santiago, but it was found, and was sent south on the N.B. Palmer. The stop in at Palmer Station (27 April) was extremely helpful, as we were able to obtain the last few supplies we needed, and use the science laboratory facilities to preweigh chemicals (the boat's movement prevents the use of balances at sea). We left Palmer Station eager to get started, but knowing we would have to wait through Process Site One (PS1) before our part of the cruise could begin.

Figure 4. Map showing bathymetry at station 1. Scale noted in inset. Map generated using ship’s echosounder with MATLAB by Meng Zhou.

On the way from Palmer to PS1, we did some whale observations for the International Whaling Commission (IWC), and organized all our field gear, so that when we arrived in an area with seals, we would be ready to go. We also started testing the satellite relay data loggers that we plan to deploy on the seals, and found a few programming glitches. Most of the troubles have been worked through, but we have a few tests remaining to complete. Bill Fraser's arrival on the boat has allowed us to talk about likely sites to see penguins and seals, and we are working together to plan the predator captures. We will help Bill Fraser with penguin handling when possible, and he and Chris Denker will help us with seal handling when possible. At PS1, we were able to start our collection of potential prey items from the tucker trawls and MOC-10 nets (Figure 5) that were deployed by Jose Torres' group. We have some Antarctic krill and Electrona antarctica in the freezer at this time. By the end of the cruise we are hoping to have at least 10 specimens from each of the fish species and krill size classes that crabeater seals may be foraging upon. We will use these specimens for the stable isotope ratio and fatty acid analyses of diet. We also worked with Jose Torres, Skip Owen, the Captain, and Bill Fraser to identify where we wanted to go after PS1 was completed. We have decided to move inside Marguerite Bay to process site 5 (PS5) in order to have the highest chance of finding predators while there is still some light available to work with. As PS1 comes to a close, we are looking forward to getting started.

Figure 5. Tracey Sutton (on left) and Chris Simoniello organize the MOC-10 while the sea looks on. Photo by Joel Bellucci.

2. BG-235 (Stewart and Marschall for Fritsen)

Vertical profiles of in vivo fluorescence, irradiance (PAR), salinity, and temperature were obtained daily from CTD casts conducted at or around local noon. Casts on 29 April and 4 May were done in on-shelf waters (~bottom depth of 550 m); all other casts were done off-shelf (~bottom depth of 2800 m). Water column samples were taken concurrent with CTD/Rosette deployment from depths of 0, 5, 10, 15, 20, 30, 50, and 100 m. Sub-samples were preserved for later determination of dissolved organic carbon (DOC), dissolved organic carbon (DIC), particulate organic carbon (POC), and bacterial and viral abundance, filtered for on-ship determination of chlorophyll a concentration, and assayed for estimates of bacterial production and photosynthesis-irradiance relationships (PE curves, at 5 and 30 m only). Not all parameters were sampled/measured on each cast. The depth of the thermocline was ~60-70 m at off-shelf stations and ~70-75 m for on-shelf stations. Fluorescence at all stations remained consistent down to the thermocline with peaks at or near the surface; chlorophyll concentrations averaged ~0.22 mg l^-1 throughout the euphotic zone at off-shelf stations and ~0.30 mg l^-1 at the same depths at the on-shelf station sampled on 29 April. Bacterial production was estimated at only two stations. Production was consistent down to 100 m with a peak at 5 m at the on-shelf station, and decreased linearly (R² = 0.77) from the surface to 100 m at the off-shelf station. Primary production over a range of light levels was elevated at 5 m relative to 30 m. Process site 1 was ice-free; sea ice in its earliest stages of formation (grease, pancake, nilas ice) was not observed.

3. BG 236-0 (Daly)

The HTI acoustic system was deployed after we first arrived on process station 1 to complete the noise tests and collect data. However, the rough seas caused a significant noise problem as well as causing some damage to the towed body. During the 5 days on station, the seas only calmed down enough to deploy the HTI on two occasions during a 10-m² MOCNESS tow. All told, about 4 hours of acoustic data were collected at the off-shelf site and about one hour at the on-shelf site.

At the off-shelf site, a thin layer was detected on both the 38 kHz and the 120 kHz systems at about 60 m, which coincided with the pycnocline. Both frequencies also detected a significant layer about 250 m deep. Net tows collected myctophid fishes and Euphausia triacantha from this depth region. On shelf, the sea state conditions were marginal (27 kt winds and very choppy sea state) and the data often were obscured by noise. Nevertheless, both frequencies again detected a thin layer at ca. 60 m and a deeper layer about 180 m. Net collections indicated that the deeper layer was composed of myctophids. The deteriorating weather prevented the HTI from being deployed with the 1-m² MOCNESS. Later in the early morning, the HTI tail fin was damaged by green water swamping the back deck during the transit to process station 5. Repairs are currently underway.

At process station 1, live tows collected large abundances of Euphausia superba larvae (i.e., calyptopis and furcilia) in the upper 50-60 m at both the off-shelf and on-shelf sites. Experiments to measure growth and molting rates, egestion rates, and ingestion rates were successfully completed. Several hundred larvae were frozen for dry weights and chemical composition.

4. BG 245-0 (Torres)

Five successful MOC 10 tows, 3 to 1000 m and 2 to 500 m, and 8 live net tows were executed during the occupation of process site 1. Depth strata sampled in the 1000-m MOC-10 tows were as follows: 0-1000 m, 1000-500 m, 500-200 m, 200-100 m, 100-50 m, and 50-0 m. MOC 10 tows in the upper 500 m sampled depths of 0-500 m, 500-300 m, 300-200 m, 200-100 m, 100-50 m, and 50-0 m. Our 1000-m MOC-10 tows revealed a typical oceanic fauna, with a rough breakdown as follows.

1000-500 m: Cyanomacurus piriei (rattail or grenadier fish), Pasiphaea scotiae, Gymno-scopelus braueri (myctophid), Gnathophausia, Bathylagus antarcticus, Euphausia triacantha, Thysanoessa macrura, Gigantocypris mulleri, Gennadas.

500-200 m: Periphylla, Electrona antarctica (myctophid), Gymnoscopelus braueri, Bathylagus antarcticus (deep-sea smelt)

200-100 m: Gymnoscopelus, Electrona, Euphausia triacantha. This depth stratum had highest fish numbers. E triacantha numbered about 50.

100-50 m: Salpa thompsoni, Euphausia triacantha, Thysanoessa macrura

50-0 m: Periphylla, beaucoups salps, Themisto gaudichaudii

Our samples on the shelf revealed a very similar faunal composition with depth. One of the dominant fish species in the 500-1000 m depth stratum, Bathylagus antarcticus, had largely dropped out at the shelf break. It should be noted that Euphausia triacantha was very abundant in our nets both on and off the shelf. The author considers it to be a good indicator species for the presence of Circumpolar Deep Water.

We have had good luck with our live animal captures. E. superba furcilia are very abundant and our physiological measurements are going well. We have about 100 respiration runs already complete.

5. BG 248-0 (Zhou)

The first ADCP survey started at process site 1 from 66 24.341° S, 70 45.512° W at 1725 local time on 29 April and finished at 1830 local time on 30 April. The study site was chosen at the shelf break where a relatively deep channel leads to the deep canyon into Marguerite Bay. The survey consisted of 4 across-shelf transects, and covered an area of 56 km at the across-shelf break direction and 28 km along-shelf break direction. The shelf break is centered at the middle of the cross shelf transects.

The measurements up to 310 m (the maximum depth of good ADCP measurements) showed a northeastward current of approximately 20-25 cm s^-1 representing the Antarctic Circumpolar Current (ACC) in the water deeper than the 3000 m isobath. A clockwise mesoscale eddy of 11-18 km was discovered in our study area off the shelf break. This eddy penetrated from the surface to the maximum depth of ADCP measurements, and it remained during our study period from 29 April to 4 May. On the shelf, the current was northeastward at the surface; it rotated to an across-shelf direction at 230 m. The measurements demonstrated an across-shelf intrusion of Upper Circumpolar Deep Water (UCDW) onto the west Antarctic Peninsula continental shelf, and into the deep canyon which links to Marguerite Bay.

The CTD measurements showed the surface water was at -0.617° C and 33.748 psu on the shelf. The mixed layer was up to 70 m. In the off-shelf area the surface water was at -0.894° C and 33.746 psu. The mixed layer depth was approximately 50 m shallower than that of on shelf. Below the mixed layer, it was the Antarctic Winter Water that was coolest (-1.31° C on the shelf and -1.71° C off the shelf). The maximum temperature was 1.52° C at 259 m on the shelf and 1.82° C at 210 m off the shelf.

The processed ADCP echo intensity measurements along the transects showed the highest-backscattering layer between 150-250 m and having a horizontal scale from 5 to 15 km. A second layer at a depth of 50-100 m was also present but weaker. There was no difference between on-shelf and off-shelf areas. The data did not show any diel difference in the vertical structure.

The surface zooplankton aggregations measured by the ADCP were consistent with the measurements of our Optical Plankton Counter (OPC). The OPC showed a maximum zooplankton abundance at the size of krill larvae, which was confirmed by samples from MOCNESS tows. However, what was responsible for the highest backscattering between 150 and 250 m cannot be confirmed by either the OPC or the MOCNESS. In this depth range, the OPC counts and sizes showed no difference from background, and MOCNESS samples showed a few large krill which would not contribute to such high backscattering. This question needs to be further investigated.

PROCESS STATION 5

Process station 5 (Figures 6 and 7) was occupied as the second process sampling site to maximize the probability of encountering predators: penguins and seals. It was clear from satellite images and the sea surface temperature that the probability of seeing pack ice was very low and we still needed to get some predator data. The logic was that if we could not get sea ice, at least we could get a rocky shoreline or calved-off bergie bits with a reasonable chance of access to crabeater seals and penguins. Also, our krill groups had some good trawling locales within striking distance of any predator concentrations that came to light.

Figure 6. Cruise track for process station 5. See text for details on sampling.

Our strategy paid off reasonably well. We wended our way up Laubeuf Fjord, which is the main channel that runs to the east of Adelaide Island. In the northeast corner of the channel that runs around Wyatt Island (Figure 8), we found some crabeater seals. The Burns seal team had a good day instrumenting and working up seals. Unfortunately, no penguins were to be found. To get access to penguins we transited to Avian Island, which is just south of Adelaide Island, in the hopes that we would find some AdJ lie penguins for Bill Fraser and Chris Denker to work with. The weather held off just long enough after our arrival at Avian Island for Fraser and Denker to get ashore and instrument and lavage a few penguins. We got a good full day of work done there and then shot back up toward Pourquoi Pas Island and Bourgeois Fjord to explore for seals and penguins, but that turned out to be barren of predators. In the interim, during dark hours we were successful in getting at least a MOC-10 and MOC-1 every night, an ADCP survey every other night, and ADCP surveys during weather bad enough to preclude any other data collection. As a consequence, we have mapped process station 5 quite well. In addition, we have multiple tows in proximity to predator concentrations.

Figure 7. Map showing bathymetry at process station 5. Scale noted in inset. Map generated by Meng Zhou with MATLAB, using output from ship’s echosounder.

Figure 8. Wyatt Island. One of the many spectacular examples of Antarctic shoreline in Laubeuf Fjord. Photo by Joel Bellucci.

Our concept of process station 5 as it evolved through our sampling is as follows. It is bounded on the south by the southern tip of Adelaide Island to include Avian Island; the southern boundary extends east to the continent (Figure 6). It is bounded on the north by the northern end of Wyatt Island and the channel that runs around it. Included within station 5 are the many fjords and islands that occupy the area between the east coast of Adelaide Island and the Antarctic Peninsula.

After a few science investigator discussions and assessing a couple of critical factors, notably daylength and ice cover, or lack thereof, we decided to head for our original process site 4 in the vicinity of George VI sound for the same reasons we went to process site 5. It was the best bet for everyone getting the data they needed. We have the head of the canyon right there for night sampling operations and testing our faunal concentration hypothesis and we have some islands near the canyon that provide potential habitats for penguins and seals. Also, it was a pretty likely spot for sea ice formation.

1. BG-232-0 Burns/Costa

On 5 May, we completed our testing of the Platform Transmitter Terminals (PTTs) that we will deploy on crabeater seals in the hopes of deploying tags once we were on site at Process Site 5. However, on 6 May, it was snowing and windy, and so we were unable to conduct Zodiac operations. We moved further north up Laubeuf Inlet in the hopes of finding calmer seas, some protection from the wind, and some ice. It is clear that there will be little sea ice in the region, but we are hoping that the seals haul out on shore or on glacier ice. On 7 May, we deployed our first PTT on an adult female crabeater seal. We left the boat at 0900 and worked our way in north up the Lawrence Channel (east side of Wyatt Island) looking for seals. We immediately saw Antarctic fur seals hauled out on the southeast shore of Wyatt Island (Figure 8) and as we moved north up the channel, we saw more and more fur seals (mainly on the shore of Wyatt and Arrowsmith Penninsula), and then some Weddell Seals on small floes in the channel itself. Towards the head of the channel we saw many crabeater seals, and moved through the brash ice to a small floe, where we caught an adult female at 1130. We had planned to do a reduced suite of measurements on this animal to get the feel for working on small floes and with gas anesthesia, which has never been used on pack ice seals before. The anesthesia went very well, and we successfully deployed our first PTT. We hoped to return to the bay the next day as there were many crabeater seals around, and this was the first location where we had seen any. Unfortunately, on 8 May, the weather was poor, with high winds from the northeast and large waves. We remained on board, and planned for our work the next day. On 9 May, we searched the area around Avian Island for crabeater seals while Bill Fraser's team worked on AdJ lie penguins. We saw many southern elephant seals and Weddell seals hauled out on Avian Island and a few crabeater seals were spotted in the water, but we saw none on land. On the adjacent shore of Adelaide Island there were many southern elephant seals, but the nearby reefs were awash, and so we were unsuccessful at finding crabeater seals. Large swells prevented our searching further, or working far from the vessel. On the 10 May, the boat moved northeast up Bourgeois Fjord, and at first light we were at the north end of Ridge Island. For once, the sky was clear, wind was low (15 kts), and temperatures warm. We (seal team plus Frank Stewart and Skip Owen) headed off the boat at 0900 and went west between Blaiklock and Pourquoi Pas Islands. We saw one minke whale and the odor and mess of a shag or penguin rookery. The narrows were full of old ice (large bergs) but no sign of seals or sea lions on the beach or ice. We got to the inlet on the south side of Blaiklock before the wind turned us back. We then headed north up the fjord along the side of Blaiklock, but did not make much headway against the wind and chop, and decided to move back downwind, as there were no sign of seals in this region either.

We searched along the west side of the low tip of Ridge Island, and then cut across to Dog Leg Fjord. The fjord was calm and filled with brash ice and larger bergie bits, and Frank Stewart was able to collect some newly forming ice, the first we had seen since 7 May. We spotted 3 crabeater seals on very small floes, and far in towards the bay. The heavy ice made it difficult to work in very far, as sea ice hampers the Zodiac. The first seal we approached entered the water while we were moving in, the second was too deep into the fjord and sea ice to reach, and the third was on a very small floe (8 x 10 m). We were able to get the seal in the net, but it escaped into the water before we could secure it. So, with no other seals in the area, we headed back across the channel, and down the east side of Ridge to the boat. Dog Leg Fjord looks to be a good site for seals with a protected bay and old floes and is a good site to examine next season. Blind Bay might also be good, depending on the wind. We were unable to check it because the northeast wind made the approach in a Zodiac difficult. We were all cold and wet by the time we got back to the boat.

On 11 May, we returned to Wyatt Island (Figure 9), since it has been the only place were we have seen a plethora of pinnipeds. Today was no exception. The wind was high, but the Captain was willing to navigate along Hinks Channel (no previous soundings) to get us in to the bay at the northeast end of Wyatt. Here we found many fur seals and crabeaters, and saw several minke whales. The winds that have been plaguing us had moved several large icebergs out of the bay since we were last here, but there seems to be a current pattern that retains sea ice in this area. The first animals we saw were the fur seals, but as we moved deeper into the sea ice, we saw several crabeaters on floes, and many more in the water. We launched the Zodiac, and were able to rapidly deploy two PTTs: one on an adult male and one on an adult female. Each deployment took about an hour and a half, and the last one went very smoothly. The wind packed the sea ice together, and after the second seal, we were losing light, and the remaining visisble seals were deep within the packed floes. We returned to the boat (1530) to work up the samples collected.

Figure 9. Two crabeater seals on a piece of glacial rubble in the vicinity of Wyatt Island, process station 5.

It seems to us that the crabeater seals were hauling out in the morning, and then in late afternoon, most of the seals we saw were in the water. This suggests nocturnal feeding, and we are anxious to retrieve data from the tags we deployed. We were able to reconstruct the movements of the female we tagged on 7 May. She spent a few days working along the coast of Arrowsmith Peninsula, and then moved north to the northeast corner of Day Island (the top of Hinks Channel). That area is very similar geographically to the area we have been working (although we have not been there yet), and it looks to be a likely spot to look for seals in the future. The dive data we have recovered from that female shows that she is spending a lot of time in the water, and making a reasonable proportion of dives between 100-200 m. This is the depth where the net tows in the area are finding lots of krill, mysids, and fishes. By remaining in this area and working here, we have been able to document (for the first time) the diving and movement patterns of a seal and the depth distribution and abundance of its prey. This is very exciting. The female seal also appears to be spending most of her time near shore. As this is where all the remnant floes are, it seems that she is preferentially using areas where there is suitable ice for haul out.

2. BG -236-0 Fraser

The SO GLOBEC seabird component aboard the Lawrence M. Gould had a most successful week. Working in calm waters greatly facilitated the analysis of AdJ lie penguin diet and scat samples collected at Palmer Station since October 2000. This aspect of our GLOBEC work couples directly with the activities of the Palmer Long-Term Ecological Research program as we seek to obtain a continuous seasonal record of AdJ lie penguin diets in the sampling grids shared by these two programs. After a long search, we were also finally successful in finding AdJ lie penguins in the Marguerite Bay region. Finding these birds gave us an opportunity to deploy five of the ten satellite tags we are carrying with us. We were also able to obtain ten diet samples from birds recently emerged from feeding at sea and collected the scats of 20 birds that had overnighted at the sampling site. We are currently receiving excellent data from the satellite tags deployed in Marguerite Bay and also from tags deployed earlier at Palmer Station.

2. BG-235-0 Stewart and Marschall (representing Fritsen)

The fifth of May was spent in transit to process site 5 at the northern end of Marguerite Bay. Following arrival at process site 5, CTD casts were conducted daily at or around local noon from 6 to 10 May. Vertical profiles of CTD parameters (in vivo fluorescence, irradiance (PAR), salinity, and temperature) extended to maximum depths ranging from ~300 m on 9 May to ~840 m on 6 May. Water column samples were taken concurrent with CTD/Rosette deployment from depths of 0, 5, 10, 15, 20, 30, 50, and 100 m. Sub-samples were preserved for later determination of dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), particulate organic carbon (POC), bacterial and viral abundance, and spectral-absorption by phytoplankton, filtered for on-ship determination of chlorophyll a concentration, and assayed for estimates of bacterial production and photosynthesis-irradiance relationships (PE curves, at 5 and 30 m only). Not all parameters were sampled/measured on each cast. A thermocline was well developed at a depth of ~80 m at the deepest station (~840 m) surveyed. At other locations a thermocline was poorly developed and water temperatures increased gradually from -1.2 to -0.6^oC at the surface to ~1 to1.2^oC at the bottom. Fluorescence at all stations was elevated in the euphotic zone relative to waters below ~80-100 m. The rate of bacterial production was estimated across the range of sample depths at one station (6 May) and at 5 and 30 m at two additional stations (7 and 9 May). Production rates fluctuated little from 0 to 50 m but were noticeably depressed at 75 and 100 m. Temperatures above freezing and a strong north-northwesterly wind seemingly prohibited the formation of new sea ice in most regions of process site 5. Grease ice was observed from the ship during a period of calm winds on 7 May, and, in concert with Zodiac operations supporting BG-232-O (Burns), was sampled from the sea surface in a protected area at the mouth of Dog Leg fjord on 10 May. New-ice samples were processed for later determination of chlorophyll concentration, bacterial and viral abundance, and dominant algal taxa.

4. BG-236-0 Daly

Process station 5 was a study in contrast relative to process station 1. Laubeuf Fjord had large concentrations of larval Euphausia superba, as did process station 1. However, the larvae in this picturesque fjord were primarily late stage furcilia (F5 and F6), whereas the larvae on the outer shelf were a mixture of calyptopis, early furcilia (F1-2) and some later furcilia stages (F4-6). The larvae in the fjord were also somewhat larger at the same stage than those on the shelf. The relative size differences suggested that the fjord was a more favorable environment for larvae, assuming that they had been in the region for some time. In addition, Laubeuf Fjord had large concentrations of large krill (juveniles and immature females and males), fishes, mysids, copepods, and amphipods. Other euphausiids, Euphausia crystallorophias and Thysanoessa macrura, were also present. During our occupation of process station 5, we measured growth and molting rates, ingestion rates, egestion rates, and assimilation efficiency in both furcilia and large E. superba. All stages were lively and actively feeding. The adults have sequestered lipid for overwintering.

Acoustic data were collected during four 10-m² MOCNESS tows, two 1-m² MOCNESS tows and 1 Tucker trawl. Diffuse layers near surface appeared to be larval krill. Deeper dense layers and individual targets were detected at the same depths where large krill and fish (e.g., Pleuragramma) were collected in nets. We also calibrated the acoustic system using a standard tungsten carbide sphere. The target returned sound intensity on the 38 kHz frequency that was close to the theoretical value, while the 120 kHz system was about 3 dB hot. The difference in the efficiency of the 120 kHz transducer will be adjusted in the post-processing of data.

5. BG-245-0 Torres

The physiognomy of the catches changed substantially from the shelf break to the process station 5 area. The most obvious change was the presence of large numbers of adult krill in the 50-100 m and 100-200 m depth strata. Some new species made an appearance, notably the mysid Antarctomysis ohlinii. It was the dominant species in the 200-300 m depth stratum and was very abundant in the 300-500 m stratum as well. Pleuragramma antarcticum became the dominant fish and we found it from 50-500 m, with more individuals in the 200-500 m depth range. Pareuchaeta became a major player in the upper 100 m and krill furcilia, while present, were not nearly as abundant as at the shelf break.

The euphausiid, Euphausia triacantha, a dominant at the shelf break, was completely absent from our catches at process station 5 as were other members of the classical midwater fauna including Gnathophausia, Gigantocypris, and the coronate scyphomedusa Periphylla. At a couple of the process station 5 trawling sites we picked up considerable siphonophore biomass at the 200-400 m depth range, and in one, we picked up a huge swarm of thimble jellies in our 200-300 m net.

6. BG-248-0 Zhou

Both ADCP echo intensity measurements and MOC-1 tows showed significant gradients between process station 1A (66° 5.9¢ S, 71° 3.7¢ W) off the shelf break and at process station 1B (66° 5.7¢ S, 70° 59.3¢ W) on the shelf. In ADCP measurements, the vertical structures were similar, but the strengths were different between process stations 1A and 1B. The strongest backscattering layer was centered at 200 m with a thickness of 100 m. A second, weaker backscattering layer was observed at the surface. The catches of MOC tows at process station 1B were at least 10 times the biomass of those at process station 1A. In all hauls, there was a large amount of yellow-green material with the appearance of green grass when viewed under the microscope, though the fluorometer measurements did not show any significant change in fluorescence concentration (voltages) between process stations 1A and 1B.

We had to abandon our plan to run a transect following the deep canyon into Marguerite Bay during our transit from process sites 1 to 5 because of heavy seas. The captain took a short direct transit from our shelf break station (1B) to the south of Adelaide Island, and took Woodfield Channel into the north end of Marguerite Bay, the entrance of Laubeuf Fjord. The transect showed a contrast from patchy mid-depth aggregations and a weak surface layer in blue water and at the shelf break, to very dense near-surface aggregations and mid-depth aggregations in Marguerite Bay. MOC-1 tows, similar to MOC-10, caught very large numbers of E. superba adults and larvae in the deep channel at the entrance of Laubeuf Fjord (68° 3.7¢ S, 68° 3.1¢ W). In a MOCNESS tow, we typically found:

25-50 m: mostly larvae with a few adults

50-100 m: A lot of E. superba adults (5-6 cm)

100-200 m: A lot euphausiid larvae, some E. superba adults.

The ADCP is mounted approximately 4 m below the surface; the bin length is set at 8 m and the blank after transmitting is 8 m. From these settings, the center of the first measurement is at 14 m, which should be an integration between 10 and 18 m. We missed the first 10 m close to the surface. Because E. superba adults were caught between 50-150 m, the ADCP echo intensity measurements should not miss swarms of E. superba adults.

The cross-talking between the OPC and MOC-1 was greatly reduced by reducing the SFK voltage on the OPC from 6 volts to 1 volt. The OPC behaved normally and the MOCNESS communication quit very occasionally, but otherwise worked normally. We should thank our electronics technicians, Sheldon Blackman and Bruce Felix, for their tireless efforts to solve these problems.

The meteorological data collected on the Palmer and Gould were compared for the period when they met. The long wave radiation measurements on the Palmer and Gould showed a difference of 25 W m^-2 during the period when both of them were at Palmer Station. The comparison for the period when they met in Marguerite Bay showed a difference of 50 W m^-2.

The mean meteorological parameters for the period from 5 to 10 May (when the Gould was in the vicinity of process station 5):

These parameters were calculated after applying a 60-minute lowpass filter to the 1-minute meteorological data.

The water below 200 m in the deep basin at the entrance to Laubeuf Fjord showed the same temperature-salinity characteristics as the water below 200 m on the shelf. The temperature of the surface mixed layer is similar to the surface water on the shelf, but was much fresher. The temperature and salinity in the mixed layer were 0.547^oC and 33.363 psu, respectively. The mixed layer depth was approximately 70 m. The temperature and salinity in the deep water below 250 m were 1.151^oC and 34.624 psu, respectively. Moving further into Laubeuf Fjord, the temperature-salinity signature west of Wyatt Island was similar to that in the deep basin at the entrance of Laubeuf Fjord. However, the mixed layer was shallow, approximately 50 m, and the temperature was 0.12^oC cooler.

The CTD cast at 67^o 7.6¢ S, 68^o 9.3¢ W just south of Adelaide Island, west of the entrance to Laubeuf Fjord, showed different temperature-salinity characteristics. The surface temperature was 0.17^oC, and the salinity was 33.045 psu. The upper water column was continuously stratified. The water column between 175 m and the bottom (~300 m) was very uniform. The temperature was 0.238^oC and the salinity was 34.258 psu, which indicated that the water was much cooler, fresher and lighter than the water at the entrance of Laubeuf Fjord.

An ADCP survey over the deep canyon showed a mean southward flow of 20-40 cm s^-1 at all depths. There are northward currents of 5-10 cm s^-1 on both sides of the deep canyon. The survey took over 11 hours, and covered an area of 15 x 15 nautical miles. We did not see any reversal current in the deep canyon. The lack of vertical structure in currents suggested that current is mostly barotropic. The survey also covered almost a semidiurnal tidal period in the first survey. The measurements show a persistent southward flow in the deep canyon. This current has been repeatedly measured throughout the study period at process station 5.

Corresponding to the current observed around process station 5A, the E. superba adults were concentrated at depths of 50-150 m over a horizontal scale of 2-5 miles. More aggregations were found along the sides of the deep basin where currents were weaker or reversed to the north. Such relation may indicate a retention mechanism for krill to remain in this area. Of course, this is very preliminary. More detailed data analyze will be conducted.

The ADCP transects made along the principal axis of Laubeuf Fjord and Cole Channel showed a mean southward flow plus a near tidal period oscillation (This was a very rough view from moving ADCP data). The ADCP echo intensity transect shows a spatial variability of krill distribution. The largest E. superba swarm of 20 nm along the fjord is located in Laubeuf Fjord, east of Rothera Station (the U.K. land base). E. superba adults were still abundant in the area west of Wyatt Island, but distributions were patchy. For example, right after a MOC-10 tow which caught a lot of E. superba adults, a MOC-1 tow caught very few along the same track. The processed ADCP echo intensity data showed us that we just missed the swarm during the MOC-1 tow.

It is not known whether E. superba adults perform diel vertical migrations. We saw evidence that the backscattering layer occupied by E. superba adults moved deeper from 75-100 m to 150 m, or simply disappeared in the mid-layer typically occupied by euphausiid larvae in the morning. It was hard to interpret such change because it can be produced by spatial variability, current advection and vertical migration. Whether E. superba adults vertical migrate between day and night needs to be further investigated.

Several ADCP transects showed the highest abundance of E. superba adults yet recorded during the cruise. The aggregations of E. superba adults were found mostly on the slopes of the deep channel where currents were relatively weak or northward. An ADCP/ XCTD survey was abandoned in midstream because of an anticipated need to for more time to transit to Wyatt Island under heavy wind conditions. Ten XCTDs were launched; five were bad.

The current in this portion resembled an estuarine circulation, with a surface southward flow and a deep returning flow. E. superba adults were abundant, but less than process stations 5A and 5B. The patch size was also reduced to 1-2 km.

The MOC-1 tow samples showed that there was a huge amount of salps, mysid and large copepods below 150 m. The codends were overfilled. The ADCP showed a weak vertical uniform oscillation flow, suggesting barotropic tidal currents.

PROCESS STATION 4

Synopsis

Process station 4 was located in the southeastern corner of Marguerite Bay and included the entire George VI Sound (Figures 10 and 11). It was bounded roughly by the tip of Alexander Island near 68.8^oS 70.6^oW on the northwest, 68.8^oS 70.0^oW on the northeast and the southern end of George VI sound on the south. It included the southeastern end of the large across-shelf canyon that runs south-southeast through Marguerite Bay (Figure 11). The canyon runs right down the middle of George VI sound. The L.M. Gould arrived at process station 4 at about 1000 on 13 May in the vicinity of the Bugge Island group but wind and visibility prevented close inspection of the island. A trackline was developed for towing and two MOC 10s and two MOC 1s were executed on our first day of sampling. Each day thereafter the Gould spent the daylight hours surveying for seals and penguins and the night hours towing either MOC 1s, MOC 10s, or executing an ADCP/CTD survey down the length of the canyon to map current patterns. Nets were fished at two points along the length of the canyon. The southern end of the canyon was well sampled; depths were about 1000 m all the way to the perennial ice at the southern end of the canyon. There was little or no new sea ice in George VI sound and the coasts contained much brash and many grounded icebergs (Figure 12). More specific information is available in the individual group reports.

A few seals and numerous penguins were spotted but were inaccessible to our predator teams due to the character of the sea ice, which was largely unconsolidated glacial rubble. Process station 4 was departed 17 May for process station 2, which is located in the mouth of the Bay on the canyon axis.

Figure 10. Cruise track for process station 4: George VI Sound. Map generated by Bruce Felix.

Figure 11. Map showing bathymetry at process station 4. Scale noted in inset. Map generated by Meng Zhou with MATLAB, using output from the ship's echosounder.

Figure 12. The end of the road. Icebergs and fast ice at the southern end of George VI Sound. Photo by Joel Bellucci.

1. BG 232-0 Burns/Costa

This week we did not have much success at sighting seals, and were unable to conduct any research. The few seals we sighted were deep in the pack ice and inaccessible by Zodiac (n=4) or sighted late in the day on floes too small to support the seal and us (n=2). The oceanography has shown that the prey availability in this area is lower than that of the previous process station. In addition, there has not been much suitable sea ice for seals in the area, as most of the sea ice we have seen has been either large icebergs or very small brash. We conducted a thorough search of lower George VI sound and the western coastline, and do not think that the seal abundance in the area is high. It is possible that there are seals in the area, but the low light, low abundance of prey, and lack of suitable haulout substrate makes it unlikely that we could capture any, even if more time for searching were available. We were able to obtain a few krill from this station for our fatty acid and stable isotope work on seal diets, but there were few fish caught in the trawls, and so we do not have any fish samples from this area of the Bay.

In contrast to the (lack of) work in this area, we have been receiving data from the three seals we tagged last week by Wyatt Island. All three are now at the northeast tip of Day Island and foraging in the same locations and to similar depths. The two seals we tagged on 11 May both headed to Day Island on the evening of 11 May, which coincided with the clearing of the bay by wind (see 12 May report below). The seals were diving routinely to ~ 200 m, which was just above the bottom (as determined from ETOPO 2) but were also spending time foraging at about 50 m. The area they were frequenting is characterized by a variable bathymetry and several deep holes. We are very interested in returning to that area to deploy the remaining 5 tags and to obtain some more detailed information about the prey field in the area. The area to the south has high prey abundance and did contain some suitable ice.

Detailed Report: 12 May: We remained at Wyatt Island overnight, following our success on 11 May. However, it blew hard in the night, and in the morning as we rounded the north tip of the Island, it became clear that the wind had blown the Bay free of sea ice. The ship then circumnavigated the Island, and headed to George VI Sound.

Detailed Report: 13 May: We arrived near George IV Sound in snow and fog, no possibility of seal work as we were in open water.

Detailed Report: 14 May: We heard from the N.B. Palmer that they had seen a large concentration of seals at the tip of Alexander Island, and a smaller concentration two-thirds of the way up the coast (west side of the Sound). So, during the day we searched for seals along the western coast of Alexander Island (from Damocles Point to Cape Brown in the north). Snow and low visibility hampered our search range, but we did not see any sea ice or any old ice of the sort seals seem to prefer. Daylight is limited, so search hours were limited to 1000-1500. We saw no seals of any species, no whales, and only a few birds. The weather forecast suggested a large storm was on its way into the Bay and so instead of continuing north to the tip of Alexander Island (which would be exposed) the decision was made to head back down into the Sound. As daylight was waning, we crossed through a strip of brash ice that had crabeater seals hauled out. The floes were too small to work and the light was fading fast (1500). We were encouraged to see seals in the area, and had high hopes for the search the next day.

Detailed Report: 15 May: The L.M. Gould conducted a CTD grid all day; no seal research or sightings carried out.

Detailed Report: 16 May: We searched the bottom of George VI Sound for seals from 1030 onward. We started at the southeastern side near Niznik Island, and followed the ice edge along the bottom to the west until it petered out at Marr Bluff. The ice was glacial in nature, and packed tightly against the ice shelf and coastline. The few seals we saw (1 leopard, 1 Weddell, 2 fur seals, 3 crabeaters) were deep within the sea ice, and inaccessible by Zodiac or the L.M. Gould. However, we did see a large number of AdJ lie penguins. The search track crossed several CTD stations, so we stopped our search activities twice to conduct CTD stations for Meng Zhou's group. This reduced our search time from 5 hours to 3 hours.

Detailed Report: 17 May: The ship repeated the cruise track from yesterday, this time looking for penguins. We looked for seals during the daylight, but again did not see any that were accessible (total count 2 fur seals, 3 Weddell seals, 1 crabeater seal, no whales).

Detailed Report: 18 May: We spent the day in open water, conducting trawls and CTD stations. We saw no sea ice, nor seals. Science work completed to date: three SRDL tags deployed on crabeater seals; krill and mysid samples collected for dietary analyses. Remaining science to be conducted: deploy 5 SRDL tags on crabeater seals; obtain samples of fish species collected in trawls (currently we have 3 Electrona and 4 Pleuragramma)

2. BG 234-0 Fraser

The SO GLOBEC seabird component aboard the Lawrence M. Gould had a mixed week. Generally poor to dismal weather conditions, combined with poor visibility due to decreasing daylengths, greatly limited all attempts to locate penguins in areas where our remaining instruments could be deployed and more diet samples obtained. These failures have nevertheless been quite instructive in that the search for these predators has revealed some previously unknown patterns related to movements and migration and autumn habitat choices. On a more positive note, our previously deployed instruments are providing an excellent and informative database on AdJ lie penguin feeding and haul-out locations, thus providing our first glimpses of how these predators respond to circulation and bathymetry in their search for prey.

3. BG 235-5 Stewart and Marschall (for Fritsen)

Water column samples were taken concurrent with CTD/Rosette deployment at or around local noon from depths of 0, 5, 10, 15, 20, 30, 50, and 100 m at process station 5 on 12 May, at process station 4 on 13-17 May, and at process station 2 on 18-19 May. Sub-samples were preserved for later determination of dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), particulate organic carbon (POC), bacterial and viral abundance, and spectral-absorption by phytoplankton, filtered for on-ship determination of chlorophyll a (chla) concentration, and assayed for estimates of bacterial production and photosynthesis-irradiance relationships (PE curves, at 5 and 30 m only). Not all parameters were sampled/measured on each cast. Bacterial production rates and PE curves for process station 5 were obtained the week prior, and were consequently not obtained from 12 May samples. Chlorophyll a concentrations at process station 1 and process station 5 averaged between 0.20 and 0.25 m g chla l^-1 while chla concentrations in two new ice samples collected from Dog Leg Fjord at process station 5 were 3.7 and 4.95 m g chla l^-1. Rates of both bacterial and primary production at process station 4 were approximately 50-60% of those measured at process station 1 and process station 5. Bacterial production at process station 4 remains consistent throughout the euphotic zone but declines below 75 m. Production and chla data from process station 2 are not yet available. Strong winds and warm temperatures (above freezing) continued to inhibit large-scale new ice formation at process station 4. Newly-formed grease and pancake ice was observed only on 12 May; no ice was sampled. Perennial land-fast ice may be present in George VI Sound. Unfortunately, a persistent northerly wind led to the accumulation of large fields of brash ice that hindered observations of ice conditions along the southern and western shores of the sound. Process station 2 is free of sea ice.

4. BG 234-0 Daly

Euphausia superba at process station 4 in George VI Sound appeared to have shifted to an overwintering mode. Analyses for most experiments will not be completed until after we get back to our laboratories at our home institutions. A preliminary assessment, however, suggests that both larval and adult krill in this southern region have shifted to a lower metabolism possibly in response to the diminishing day length. During this past week, our group completed molting/growth, ingestion, and egestion rate experiments, and an assimilation efficiency experiment for larval and adult krill. Acoustic transects were curtailed to some extent due to high winds and many large icebergs in the region. Acoustic data were collected with one 1-m² MOCNESS tow and one 10-m MOCNESS tow. Two layers, about 55 m and 90-125 m, were detected by the 120 kHz system. Some individual targets were detected by the 38 kHz system at depths > 300 m.

5. BG 245-0 Torres

Five MOC 10 tows and four live net tows were executed during the occupation of station 4. Two MOC 10s were located a third of the way down the Sound vic 69° 15¢ S, two were located about halfway down vic. 69° 25¢ S, one was at the southern end vic 69° 47¢ S. Krill were captured in moderate abundance in the 0-50 m, 50-100 m, and 100-200 m depth strata. Adults were mainly found in the 100-200 m stratum, smaller size classes and Euphausia crystallorophias in the 50-100 m stratum and some larvae in the 0-50 m stratum. Gross estimates of krill abundance at process station 4 were about 10% of those at process station 5. In the deeper tows executed at this station (200-500 m, 500-800 m) an oceanic fauna was present, including the lanternfish Gymnoscopelus, the bathylagid Bathylagus and the pasiphaeid shrimp, Pasiphaea scotiae. At the southernmost trawling site, a new eelpout was captured, Seleniolychus, as was the icefish Cryodraco. Pleuragramma larvae were common in the upper 50 m, but Pleuragramma adults were not captured. Overall, our trawls in the canyon here showed a decidedly oceanic flavor with moderate krill abundance in the upper 200 m. In addition to the work on distribution and abundance we completed 100 individual determinations of respiration and excretion on krill adults and larvae.

6. BG 248-0 Zhou

George VI Sound and its vicinity:

The bottom topography (Figure 11) was surveyed during the whole study period in George VI Sound. The measurements indicated:

Temperature and salinity fields were surveyed by 23 CTD stations in the Sound and its vicinity (Appendix 2). The deep water below 400 m in George VI Sound shows similar temperature-salinity features to the deep water below 400 m on the shelf. The CTD measurements showed the permanent stratification between 150 and 400 m. The water within this layer is cooler and less salty than the water on the shelf, which indicates local influence. Winter water lies between 75-90 m, however in some stations, the surface mixed layer water is even cooler than this layer. The salinity in the surface water is relatively low, keeping the water column gravitationally stable. In some stations, the Winter Water layer formed last year has been eroded, which signals the arrival of the winter season.

The shelf water follows the deep channel into George VI Sound. At the center of the basin, the water is saltier than the surrounding water. Because the temperature variation is small throughout the region, the density field is primarily determined by the salinity field. Associated with the denser water feature at the center, a clockwise circulation is measured by the ADCP. In contrast, the water is less salty at the mouth of the Sound, forming a low-density center, which leads to an anti-clockwise gyre.

The circulation in George VI Sound is characterized by the two gyres described above associated with the two high- and low-density centers determined from the CTD and ADCP measurements. Further north, the flow field is dominated by the clockwise circulation in Marguerite Bay. The ADCP measurements show the westward flow between 69° 10¢ S and 69° 15¢ S. This flow bifurcates between 69° 30¢ W and 69° 40¢ W. One branch follows Alexander Island northward, and exits on the shelf slope. Another branch turns southward and then southeastward as a part of the gyre at the mouth of George VI Sound. A part of the flow intrudes into the Sound following the eastern shelf.

The wind direction sensors were down on days 134 and 135. We thank Bruce Felix for spotting the problem, and then making ceaseless efforts to fix the problem.

Mean STD Min Max

Ta(° C) -0.07 0.64 -1.70 2.41

Ts(° C) -1.41 0.31 -1.76 0

Relative humidity (%) 95.14 4.76 73.69 99.62

Atmospheric pressure (mb) 997.3 3.3 987.9 1005.1

Surface Fluorescence (volts) 0.78 0.10 0.65 5.86

Longwave Radiation (W m^-2) 283.2 28.4 184.1 309.0

Four MOC-1 tows were conducted in the process station 4 region. Samples showed a significant difference in species composition between process stations 4 and 5. There were fewer adult E. superba caught in this area. Overall, samples showed this to be a biologically active area. Large numbers of copepods and krill larvae, as well as moderate amounts of adult krill were caught in every net tow.

Process station 2 was located in the mouth of Marguerite Bay along the axis of the main across-shelf canyon that runs roughly from the shelf break to the southern end of George VI Sound (Figure 13). It was a large box bounded roughly in the northwest by 67^o 40¢ S, 72^oW, in the northeast by 67^o 40¢ S, 69^o 30¢ W, in the southwest by the northern tip of Alexander Island vic 68^o 40¢ S, 71^o 30¢ W and in the southeast by 69^oS, 69^oW. It was sited to cover the mid-canyon region and because of its location was amenable only to open water water sampling. We completed 2 CTDs, 2 Tucker trawls for live specimens, 5 MOC-10 tows, and 3 MOC-1 tows. All tows were accompanied by 120/38 kHz acoustic sampling when weather permitted. In addition to the net sampling a partial ADCP survey was also completed before sampling was curtailed by severe weather. The science party elected to move south to Lazarev Bay to process station 3 (Figure 13) during the time that would have otherwise been lost to weather.

Figure 13. Cruise track for process stations 2 and 3. Station 2 included all sampling in the mouth of Marguerite Bay and a CTD transect running southwest toward the northern end of Alexander Island. Station was wholly within Lazarev Bay. See station synopsis for more detail. Map generated by Bruce Felix.

We arrived in Lazarev Bay during first light on 21 May (Figure 14). It was decided by the science party that the Bay would replace the original process site 3 that was sited southwest of the bay in a mid-shelf location (Figures 15 and 16). It was in Lazarev Bay that we encountered the first new sea ice of the cruise. Moreover, the character of the sea ice in the Bay included some older, larger floes that were reasonable platforms for predator handling. Our sampling strategy in the Bay focused on daylight operations, making every effort to capitalize on the very limited daylight available. It consisted of visual predator surveys from the bridge for targeting seals and penguins. Zodiac operations followed for seal and penguin handling. Our predator teams were very successful in the Bay; all remaining satellite tags were deployed.

We inspected the underside of the new sea ice in the Bay with the first dives of the cruise and found krill furcilia already in residence under the mix of new pancake and glacial rubble (Figure 17). Individuals were collected for physiological work and a rough visual census was taken. Sea ice specimens were collected by our Sea Ice Microbial Communities (SIMCO) team for evaluation of microbial production and biomass.

Figure 14. Dawn in Lazarev Bay. Photo by Kendra Daly.

Figure 15. Detailed process station 3 cruise track in Lazarev Bay. Map by Bruce Felix.

Figure 16. Map showing bathymetry at process station 3. Scale noted in inset. Map generated by Meng Zhou with MATLAB, using output from the ship’s echosounder.

Figure 17. Diver collecting krill underneath the new pack ice in Lazarev Bay. Note the aquarium net used to capture krill larvae that is in the diver’s hand.

Night operations consisted of ADCP surveys for mapping the circulation and plankton biomass in the Bay, live tows with the Tucker trawl to supply specimens for our ongoing physiological experiments, and one MOC-1 for quantitative assessment of the krill populations within the Bay. Our sojourn in Lazarev Bay was highly productive. It filled a primary goal of the cruise, which was to examine the biological and physical characteristics of a newly forming sea ice regime.

1. BG 232-0 Burns/Costa

This week has been one of remarkable, if somewhat unexpected success. At the start of the week we had yet to deploy 5 of our 8 PTTs, and were becoming concerned over the success (or lack thereof) of our project. Some of the anxiety was relieved at the science meeting on 19 May, where we all agreed to set aside several days at the end of the cruise for dedicated predator work. However, the best luck of this cruise came from our diversion to Lazarev Bay on 20 May, when we left the mouth of George VI Sound due to bad weather.

Lazarev Bay was selected due to its protected location, proximity to the Wilkins Ice Shelf, and the potential for penguins, seals, and sea ice. For once everything worked as planned, and we arrived in Lazarev Bay on the 21 May to find new forming sea ice, many large remnants of second year sea ice, and many ice bergs and fragments. One of the more interesting discoveries during the transit and once we began to work in the Bay was that the charts and maps of this area are remarkably poor. Rothschild Island is actually 11 miles to the south of its charted position, and we have found and named (at least for the duration of this cruise) two new small islands. This will make interpretation of the seal movement data a little complicated, as we have no faith that the charts are correct. Resolving some of the charting issues is critical because over the course of this past week, we were able to deploy our remaining 5 PTTs on 3 adult female and 2 adult male seals.

All the seals we handled were captured on remnant second year sea ice, and floe size ranged from approximately 50 m² to 100 m². The floes were not covered with snow, but rather with consolidated bergie bits, and so were slick and difficult to maneuver on. In any given area, we found that the seals seemed to prefer the largest floes, but to select only remnant sea ice, and not any of the abundant small icebergs. During the evening transit on 21 May from Bill's Island (one of the islands named during the cruise) to the eastern side of the Bay towards Umber Island we noticed several crabeater seals hauled out on large floes late at night. This was encouraging, and we were hopeful that we could deploy tags in the Bay.

Indeed, on 22 May, we deployed satellite tags on two seals. In the morning, we captured and tagged a large adult male on a rugged piece of very slick ice. There was a bit of wind, and the ice was moving around in the bay, so that by the time we had finished, we could not find any other seals in the area. However, we continued to observe from the bridge of the L.M. Gould, and at around 1500, just as it was getting dark we started to see more seals hauling out. Since time was getting short (only 2 weeks left in the cruise) we decided to see what it was like to work in the dark, and left the L.M. Gould to attempt a capture of an adult female in the twilight.

The female seal we handled was deeply asleep during our approach to the floe and remained asleep through all the procedures. The waning twilight was sufficient to administer the initial drug dose, and the bridge lights from the L.M. Gould provided much of the illumination we needed to work at night. With an extremely successful evening deployment, we felt confident that if the weather held we could work at night. In fact, we repeated this pattern of searching for seals in the morning and late afternoon on the next two days.

On 23 May, we deployed two PTTs on animals in the late afternoon and evening, after failing to capture animals in the morning. On 24 May, we deployed our last tag on an adult female (Figure 18) that was captured in the dark, after failing to sight any seals during daylight hours. Night work has proven less difficult than expected. With excellent help from Skip Owen and the bridge, and optimal weather conditions, we have had good success with this technique.

Figure 18. Crabeater seal "Flo" with recently applied satellite tag. Flo was still under anesthesia when the photo was taken. Photo by Jennifer Burns.

Over the course of our time in Lazarev Bay, we confirmed that the seals were showing a remarkable diel pattern of activity (Figure 19). We have seen very few seals hauled out in the daylight hours, but have seen many animals active in the water in the late afternoon, and animals seem to haul out just as the light fails after 1530. This pattern of hauling out at around 1600 until just after midnight that we are seeing is similar to that being recorded by the tags deployed on instrumented animals. We are not sure why the animals are hauling out in the late afternoon, but the pattern seems widespread. As a result, we believe that the ability to work in the dark has been crucial to our success this cruise. All 8 tags are on animals and transmitting (Figure 20, Table 1). Of the three animals that were tagged during week three, two are still in the area around Day and Hansen Islands, and the third has moved north past Laird Island, and is now more than 300 km from where it was initially tagged (e.g., Figure 21). As yet, we have only recovered a small amount of data from the seals tagged in Lazarev Bay, but the data that we do have shows a high degree of individual variation. Two animals moved deeper into the Bay before one moved out to the north and the other moved northwest along the eastern coast of Rothschild Island before heading south along the outer (west) coast. A third animal moved directly out of the bay to the north, while the fourth moved west into the open ocean. We do not have sufficient data from the fifth animal yet to determine its movements. All have shown extensive movements in the few days since tagging (50 km or more).

At this point it is clear that crabeater seals do not move passively with the drifting sea ice, but instead are capable of long-distance directed movements. What drives these movements, and how the animals are selecting foraging locations, is the overall goal of our research project. In the few remaining days of the cruise we hope to handle a few more animals to trouble-shoot procedures for the winter cruise and to bolster our sample size. The colder nights in Lazarev Bay highlighted some of the difficulties we will face in July, and the better experienced we are now, the more successful that cruise will be.

Figure 19. Flo's dive depths and swim speeds as reported by her satellite tags.

Figure 20. Locations of seal tagging.

Table 1. Summary of seal tag deployments.

Figure 21. Flo's locations as reported by her satellite tags.

2. BG 234-0 Fraser

The U.S. SO GLOBEC seabird component spent much of the week searching for AdJ lie Penguins at our most southern station in Lazarev Bay. Our objectives were to obtain diet samples and deploy our last five remaining PTTs. Although we were unable to locate penguins in conditions suitable for diet sampling work to proceed in the field, we succeeded in our efforts to deploy the PTTs (Figure 22). As a result, we now have a tagged population of penguins at both the northern and southern ends of the SO GLOBEC grid, and a nice experiment in progress to examine aspects of the foraging ecology of this species in relation to sea ice development and bathymetry as winter conditions advance in this region.

Figure 22. AdJ lie penguin showing recently applied satellite tag. Photo by Joel Bellucci.

3. BG 235-0 Stewart and Marschall (for Fritsen)

The efforts of BG-235 during week 4 have been divided between water column and new sea ice sampling for the first time during the cruise. Water column samples were taken concurrent with CTD/Rosette deployment at or around local noon from depths of 0, 5, 10, 15, 20, 30, 50, and 100 m on 19 May at process station 2 and on 21 and 23 May in Lazarev Bay. A CTD cast scheduled for 20 May at process station 2 was cancelled due to rough seas. Sub-samples were preserved for later determination of dissolved organic carbon (DOC), dissolved inorganic carbon (DIC), particulate organic carbon (POC), bacterial and viral abundance, and spectral-absorption by phytoplankton, filtered for on-ship determination of chlorophyll a (chla) concentration, and assayed for estimates of bacterial production and photosynthesis-irradiance relationships (PE curves, at 5 and 30m only). Sea ice was not observed at process station 2 but was present in varying stages of early formation on all sampling days in Lazarev Bay. The newly-developed sea ice field consisted primarily of unconsolidated grease or brash ice interspersed with small (< 50 cm diameter) to large (>5 m diameter) pancakes. Sea ice coverage (by all ice types) over the water surface ranged from ~5% in wind swept areas (i.e., near "Bill's Island" on 25 May) to ~90% in fields of densely-packed pancakes (e.g., evening, 24 May). Cohesive pack ice was not yet present in any region sampled in Lazarev Bay. Samples of newly formed sea ice were collected on 21-23 May and 25 May in concert with Zodiac-supported seal and penguin surveys. Ice types collected ranged from grease ice and unconsolidated pancake ice (22 and 25 May) to brash ice chunks and consolidated pancakes (pancake holds form when handled; 21-23 May). A single sample of ice algae was scraped from the under surface of a floe by J. Torres on 23 May during diving operations beneath ice where krill were observed to be feeding. Sea ice samples were melted in the dark and processed according to protocol for measurement of the same parameters measured for water column samples. Early observations indicate that microbial biomass is elevated in newly-formed ice of all stages (grease to consolidated pancake ice) relative to the water column while production rates (primary and bacterial) of organisms aggregated in newly-formed sea ice are approximately equivalent to or slightly diminished from (i.e., bacterial production) rates determined for planktonic microorganisms.

4. BG 236-0 Daly

We completed investigations at process stations 2 and 3. Station 2 was located over a deep canyon in Marquerite Bay. Acoustic surveys were made with two 10-m MOCNESS tows. The highest concentrations were detected near surface, consistent with the relatively abundant larval Euphausia superba collected in surface layer nets. Growth, molting, ingestion, and egestion rates were measured for late stage furciliae (Furcilia 4-6). Growth and molting rates also were determined for early stage larvae (Calyptopis 3 to Furcilia 4).

Process station 3 located in Lazarev Bay provided access to early forming sea ice (Figure 14). Our first net tow for live animals collected large numbers of larvae and some adult E. superba, E. crystallorophias, and many Pleuragamma larvae. Few adult krill were collected in subsequent tows, possibly because towing was restricted to certain regions of the Bay due to the large numbers of icebergs. Growth, molting, ingestion, and egestion rates were measured for late stage furcilia. Clouds of furciliae were observed near the undersurface of sea ice by divers. A planned under-ice experiment was aborted with the radio call that a leopard seal was seen heading our way. Furciliae collected by divers were measured for gut fluorescence. In addition, egestion rates and assimilation efficiency were determined for these larvae. No acoustic surveys could be conducted in this Bay due to the sea ice conditions.

5. BG 245-0 Torres

Process site 2. Five MOC 10 tows were completed at two different locations in the mid-canyon region of site 2. Both sites exhibited high fish biomass relative to tows executed in other regions of the Marguerite Bay study area. The faunal composition at process site 2 was decidedly oceanic, with strong representation by the lanternfishes Electrona, Gymnoscopelus, and Protomyctophum and the bathlagid, Bathylagus. Krill were in low abundance in all five tows. No Pleuragramma were collected at process site 2.

Process site 3. Lazarev Bay saw a change in our sampling scheme as we switched from net tows to diving as our major sampling tool. We completed 3 dives in the five days on site and observed krill under the new ice in all three dives (cf. Figure 17). Our highest numbers were in a glacial rubble/new sea ice mix at the eastern edge of the Bay. The eastern side of the bay has a narrow, but very deep canyon that runs along part of its eastern margin and our most productive dive was located very close to it. Our Tucker tows for live specimens revealed a robust krill population in the bay with both adults and furcilia well represented. The high sea ice concentrations in the Bay precluded MOC-10 tows. About 125 individual determinations of respiration and excretion were made on krill at process site 3.

6. BG 248-0 Zhou

Process station 2: The entrance into Marguerite Bay

Bottom topography was surveyed during the whole study period. Our soundings matched the ETOPO 3 map quite well in most of the process site 2 area except near the coast of Alexander Island. There are several unmarked shallows and deep canyons near the northern tip of the island.

A CTD transect comprised of 9 stations (Figure 23) was made crossing the exit current region from approximately 68^o 30¢ S, 70^o 00¢ W to the northern tip of Alexander Island. Measurements indicate that the water near the Island was significantly less salty, producing a horizontal density gradient and a baroclinic component in the exit current.

A 20 by 20 nautical mile CTD and ADCP survey grid was designed covering mooring stations B1, B2, and B3 after we established the site of process station 2 at the middle of the entrance into Marguerite Bay. One of the purposes was to compare shipboard and mooring physical data for a better understanding of spatial and temporal variability. The survey started just as a north wind increased to 30 kts. CTD stations were eliminated and replaced by XCTDs launched from the 02 deck. After 4 successful launches the wind increased to 40 kts and the survey was called off.

The ADCP measurements consistently showed an exit current along the shelf of Alexander Island. The current measurements on the east side of the deep canyon showed a southward intruding current.

We finished 4 ADCP transects at process station 2, crossing mooring B1 once and B2 twice. Though the planned CTD stations were nearly abandoned, the ADCP measurements we acquired were reasonably good. The ADCP measurements showed a current at 30 m; the current was always southwestward and varied from 5 to 40 cm s^-1. The survey was longer than 13 hours, which is the semidiurnal tidal period.

iv) Meteorological conditions

Mean STD Min Max

Ta(^oC) -0.34 1.13 -4.05 3.46

Ts(^oC) -0.94 0.34 -1.76 -0.12

Relative humidity (%) 86.03 9.61 57.33 98.28

Surface Fluorescence (volts) 0.99 0.11 0.76 2.50

Longwave Radiation (W m^-2) 282.3 22.64 204.0 315.2

Three MOC-1 net tows were conducted in the process station 2 area. Net samples showed low numbers of krill larvae, few krill adults, and low numbers of copepods. The ADCP echo intensity measurements showed weaker backscattering in this area which agreed well with the net tow samples.

Process Station 3: Lazarev Bay

Bottom topography in the Bay was surveyed down to the fast ice edge for the first time. Rothschild Island was approximately 11 miles due south of its charted location, and the west coast of Alexander Island was approximately 5 miles due west. Our survey showed that the water depth in most of the bay varied between 100 m and 300 m except for some uncharted islands and rocks. A narrow deep canyon of 800-1000 m depth extends into the bay along the coast of Alexander Island. Because the canyon lies so close to the Island, and because there were so many icebergs in its vicinity, we could not survey its full extent. Thus, we do not know its full width and how far it extends into the Bay. The canyon walls are nearly a 200-600 m straight drop.

The water below 400 m in the canyon showed similar temperature-salinity characteristics to the deep shelf water seen elsewhere, which indicated the connection of this deep canyon to the shelf region. The water above 400 m was modified: it was much fresher and cooler.

The ADCP current measurements at the mouth of the Bay show a southwesterly flowing shelf current as an extension of the current along Alexander Island. A small branch turns into the Bay at the deep canyon. In the western shallow region, the flow is northwesterly, forming a clockwise circulation, which then joins the southwest shelf current.

Mean STD Min Max

Ta(^oC) -2.48 1.91 -5.90 6.08

Ts(^oC) -1.59 0.25 -1.78 -0.44

Relative humidity (%) 76.52 9.40 45.95 90.59

Surface Fluorescence (volts) 1.11 0.15 0.83 1.47

Longwave Radiation (W m^-2) 245.1 24.82 190.2 294.2

One MOC-1 net tow to 300 m was conducted at the deep canyon near the entrance. A MOC-1 net tow in the bay was precluded by icebergs. Net samples showed a moderate amount of krill larvae, 2 adults, and large number of copepods between 150-300 m. The OPC measurements showed a pattern similar to the net samples.

PROCESS STATIONS 6 AND 7 - END OF CRUISE

Our primary goal for the last 4 days of the cruise was to locate concentrations of penguins and seals for diet analyses. To this end we departed Lazarev Bay on 25 May and returned to George VI Sound. Observations during our occupation of process station 4 suggested that very high numbers of AdJ lies penguins were residing in the sound. We arrived on 26 May at first light, intending to make our way to the Rhyolite Islands, but were stymied by newly formed ice, ca. 6 inches in thickness, and a sound peppered with growlers, bergie bits, and icebergs. It became clear that the Rhyolite Islands were unreachable in a reasonable amount of time. We surveyed the north margin of George VI Sound and decided to move on to Neny Fjord about midway up the eastern shore of Marguerite Bay. We did a MOC-1 on a trackline southwest of Neny Fjord to get a snapshot of krill abundance in its vicinity and arrived in the fjord proper at first light on 27 May (Figure 24).

Figure 24. Final complete cruise track for LMG01-04. Process station 6, Neny Bay, is the small embayment in the middle of the eastern margin of Marguerite Bay. The ship track takes a jog to the east at that point. It is also readily discerned as CTD station 56 in the figure showing CTD locations in Appendix 2. Map by Bruce Felix.

Neny Fjord was free of new sea ice; sea surface temperatures were hovering in the range of -1.0^oC. We did a Zodiac-based penguin survey of the bay and attempted a blue water dive to survey the upper 1 m of the water column for evidence of krill furcilia in an ice-free environment. The dive was cut short by equipment problems but no krill were observed. Neny Fjord was dubbed process station 6.

At nightfall on the 27 May, the L.M. Gould transited to the south end of Adelaide Island for an ADCP survey during the hours of darkness and then moved up Laubeuf Fjord (process station 5) to the vicinity of Day Island for a seal survey, arriving there at first light on 28 May (Figure 24). We found a very changed environment there, with much of the glacial rubble and small floes gone, presumably blown out by the prolonged high winds that plagued us all during the first three weeks of May. The fjord was ice- and seal-free so the L.M. Gould moved further north into Hanusse Bay, vic Laird Island, for the last science day. We completed an ADCP survey of the bay during darkness, surveyed for seals and penguins at first light on 29 May, completed an HTI drift station, and finished the science operations with a successful MOC-10 and MOC-1. Hanusse Bay was dubbed process station 7 (Figure 25).

Figure 25. Detailed cruise track of process station 7: Hanusse Bay. Map by Bruce Felix.

1. BG 232-0 Burns/Costa

This week we headed north from Lazarev Bay with the hope of capturing a few additional crabeater seals to increase sample sizes and refine procedures before the cruise ended. However, weather and ice conditions were not optimal and we did not spot any seals in George VI Sound (lots of thin, new sea ice, 26 May) or in Neny Bay (no sea ice, 27 May). On 28 May, we returned to Wyatt and Day Islands, where two of our tagged seals were still diving. However, the weather pattern had shifted to snow with high winds from the northeast, and conditions were far from optimal for sighting or working on seals. We sailed up Lawrence Channel to the head of Wyatt Island and from there up Hinks Channel to the head of Day Island without sighting any sea ice or seals. At the head of each channel we mapped the coastline with the ship's radar. This was to confirm that the glaciers had receded from their charted positions and that the many locations we were receiving from seals "on land" were really due to errors in the charts themselves.

Following the mapping exercise, we continued north through the Gullet and into Crystal Bay. While technically outside the study area, one of our animals (PTT 23100) had moved on a direct track through Crystal Bay between the 24 and 26 May, and we were interested in surveying the area. We did not see any seals until late in the afternoon, when several were sighted hauling out in the glacial rubble that was wind-packed against the western coastline of the channel. This pattern of hauling out late in the afternoon fit with the behavior of our tagged animals and was a repeat of the pattern we saw in Lazarev Bay. On 30 May, we searched for seals in the fjords and bays in the southeastern corner of Crystal Bay, but again were unsuccessful at spotting any animals, despite the fact that we knew at least one seal had moved past this area just a few days before. The minimal amount of sea ice suitable for a seal to haul out upon likely contributed to our poor sighting success, for on 31 May during our transit to Palmer Station, we passed through extensive areas with good sea ice and second year pans, and saw many crabeater, leopard, and Weddell seals hauled out, suggesting that sighting success is dependent on both sea ice and weather conditions.

As we leave the study area, all eight of our tags are still transmitting (e.g., Figure 21, Table 1). Two of the seals remained in the area around Day Island, one has continued to track north and is now about 200 km south of Anvers Island, and the five that we tagged in Lazarev Bay are using the open ocean habitat to the west of Lazarev Bay. While these five seals are all using slightly different areas at any given time, there are remarkable overlaps in their foraging locations, which suggests some persistent oceanographic features that may be concentrating prey in certain areas (e.g., Figure 21). We look forward to integrating the seal data with that obtained by the N.B. Palmer during its survey of the area. This cruise has been successful, and we are looking forward to the next one in July. Then we hope to deploy another eight tags and discover more about what drives seal behavior in the wintertime.

2. BG 234-0 Fraser

The U.S. SO GLOBEC seabird component aboard the L.M.Gould had a most successful last week, thus achieving most of its planned objectives for this cruise. Earlier gaps in our spatial coverage of AdJ lie penguin diets were in part filled by Dr. Chris Ribic and her group aboard the N.B. Palmer, who were able to find birds in a few key locations not investigated by the L.M. Gould. Analyses of these diet samples suggest striking differences in the types of prey being taken by penguins in the Marguerite Bay region when compared, for example, with the diets of penguins north of this region. We have also obtained the first evidence that AdJ lie penguin diets in the Marguerite Bay region may be sex-specific, suggesting that males and females engage in different foraging strategies during the autumn/early winter period. And finally, the PTT data are suggesting that AdJ lie penguins are highly localized foragers in autumn, and tend to focus their search efforts around areas where circulation and bathymetry provide a retentive environment for prey. These findings have arisen in part through on-board collaborations with Dr. Meng Zhou and his group, who provided valuable ADCP data. This collaboration is expected to continue as the analyses of respective data are completed in the U.S. We conclude this brief report by thanking Captain Warren and his crew and the Raytheon staff aboard the L.M. Gould for their excellent support of our program.

3. BG 235-0 Stewart and Marschall (for Fritsen)

Substantial concentrations of newly-forming sea ice were encountered for the last time during the cruise at the southern end of George VI Sound on 26 May. The sea ice field was contained primarily within a belt paralleling the coast/ice sheet and consisted of patches of newly-consolidated clear pack ice (without snow cover; gray in appearance), interspersed within patches of open water, mixed-age pancake ice, new grease ice, and small (<5 m) to large (>100 m) icebergs. New pack ice was sampled from the deck by using a net to collect chunks broken from the pack by the hull of the ship. The new pack ice was 12.5 cm in thickness. Pack ice samples were diluted with filtered seawater and melted in the dark prior to sub-sampling for chla concentration, bacterial biomass, dissolved organic carbon, nutrients, and dominant algal taxa. Chlorophyll concentrations in George VI Sound and Lazarev Bay ice samples were variable (~0.1-1.0 mg l^-1) but, in general, elevated relative to water column concentrations, reflecting the tendency for plankton cells to be "scavenged" by newly-formed ice crystals rising through the water column and coalescing at the surface. Vertical profiles of in vivo fluorescence, irradiance (PAR), salinity, and temperature were obtained from CTD casts conducted at or around local noon on 26 May in George VI Sound, on 27 May en route to Neny Island, on 28 May in Lawrence Channel (near process site 5), and on 29 May in the northern part of Lallemand Fjord (north of process site 5). An assay designed to measure bacterial production over time within new sea ice and surface seawater from Lazarev Bay was terminated on 30 May. Production over an eight day time period was best fit by an exponential equation (r² = 0.7323). Both 31 May and 1 June were devoted to waste disposal and laboratory cleaning.

4. BG 236-0 Daly

This past week we completed experiments on krill rate measurements from process station 3. We spent one day exploring Neny Fjord, as a potential location for new sea ice formation and predators. This region turned out to be much warmer than at the southern end of the Marguerite Bay, with air temperatures about 1.8° C and surface sea water at -0.99° C. There was no sea ice and few seals or penguins. We ran three acoustic surveys, two with the Tucker Trawl and one with the 1-m MOCNESS. The 38 kHz frequency indicated fish targets throughout the water column and the 120 kHz frequency detected a small near-surface layer at about 15 m and a relatively dense layer between 50 and 300 m. Net catches included Pleuragramma, larval and adult Euphausia superba, E. crystallorophias, many copepods (such as Paraeuchaeta) and siphonophores. By the time we reached Avian Island later in the evening, winds had increased, preventing us from collecting acoustic data during the ADCP survey of that region.

On our last science day, we did a standard target sphere calibration in Lillemand Fjord and then collected acoustic data in Hanusse Bay during 10-m and 1-m MOCNESS tows. This region is upstream from Marguerite Bay and, therefore, may act as a source population for krill and fish. Relatively dense layers were detected near surface, in the upper 30-40 m, and between 100-200 m depth. Net catches included Electrona and a large number of larval and adult E. superba.

Several preliminary observations for this cruise are noteworthy, with the caveat that many samples have yet to be analyzed.

1. Larval E. superba were exceptionally abundant throughout the study area and included individuals in many stages of development, from Calyptopis 3 to Furcilia 6. These observations suggest that the past summer was a strong year for krill reproduction, that reproduction extended over a relatively long period, and that circulation was favorable for advecting and/or retaining larvae on the shelf.

1. Larval krill molted about every 18-20 days, whereas adults molted about every 30-37 days. Growth rates were measurable, but require further analyses. Gut fluorescence, a relative measure of feeding on phytoplankton or ice algae, and the production of fecal pellets, indicated that both larvae and adults were feeding during the study period. The much smaller larvae, however, had gut fluorescence levels similar to that in adults. Gut fluorescence in larvae collected by divers under sea ice was about two-fold higher at one site and similar at another site, relative to that found in larvae collected from the water column in net tows. These combined observations suggest that larval krill were actively feeding and growing during autumn in Marguerite Bay in preparation for winter.

5. BG 245-0 Torres

Three Tucker trawls (live tows) were completed during our last 4 days, two in the vicinity of Neny Fjord and one in Hanusse Bay (Figure 24). In addition, we had one successful MOC-10 within Hanusse Bay. Catches were quite different in the two regions. In Neny Fjord, we observed a fairly typical shelf fauna including Pleuragramma larvae and adults, E. superba larvae and adults, E. crystallorophias, the bathydraconid Psilodraco, many Pareuchaeta and Calanoides acutus, and a surfeit of siphonophore bracts. In Hanusse Bay, the krill populations were very much higher. Our very rough MOC-10 break down was as follows:

Note also that our 0-net oblique haul captured 2 specimens of the lanternfish, Electrona antarctica.

Our krill catches in Hanusse Bay were about equal to those in Laubeuf Fjord, the highest of the cruise. We made 100 individual determinations of metabolism and excretion in krill and other species in the last four days of science.

6. BG 248-0 Zhou

Bill Fraser raised a question to Jose Torres, Kendra Daly, and me (Meng Zhou) one day: how much food do the 30,000 pairs of AdJ lie penguins on Avian Island need in a day? We found there were a lot adult krill in Laubeuf Fjord from our intensive mesoscale surveys and process studies in the second week of the cruise. Though we found a fair amount of adult krill in George VI Sound and Lazarev Bay, the catches cannot be compared to those in Laubeuf Fjord. Questions we raised: Is there another place like Laubeuf Fjord, and where are the sources of those adult krill? We examined the records of penguin sites, circulation patterns we just acquired and analyzed, and topographic features. The plan was: 1) to conduct an exploratory survey of krill, penguins, and seals in the vicinity of Neny Fjord, 2) to conduct an ADCP survey in the vicinity of Avian Island for krill population and circulation patterns for a better understanding of krill spatial distributions and circulation patterns utilizing the night hours in the transit from Neny Fjord, and 3) to explore Hanusse Bay north of Laubeuf Fjord for the potential physical and biological linkages between these two, if there was time available after the seal searching operation.

Neny Fjord (68° 37.2¢ S, 67° 38.0¢ W) We learned in our first two weeks that adult E. superba were concentrated between 50 m and 120 m; larvae were distributed mainly near the surface. It was quite easy to acoustically identify layers of adult E. superba on the ADCP. However, they were often found near the walls of deep canyons containing abrupt and unpredictable shallows, and their horizontal scale was relatively small, typically varying between hundreds of meters and 2-4 km in extent. The exception was the very large patch in the deep canyon at the mouth of Laubeuf Fjord. A typical MOCNESS tow lasts 1.5-2 hours and covers approximately 3-4 nm. It was easy to miss the main concentrations of the patchily distributed E. superba adults, and we often did. One MOC-1 tow and one CTD were conducted utilizing the available time during the night transit between George VI Sound and Neny Fjord, as well as the period during Zodiac operations for seal and penguin surveys. The MOC-1 samples showed a large number of copepods, salps, fish larvae, and amphipods, some krill larvae, and several krill adults between 300 and 150 m, a large number of krill larvae between 150 and 25 m, and a few krill larvae between 0-25 m. We missed the main concentrations of adults.

A CTD cast was conducted at the entrance of Neny Fjord, a shallow embayment of approximately 300 m depth. Because the water depth of the Bay is shallower than that of intruding deep water, the vertical temperature and salinity structure was influenced mainly by local and regional effects. Compared with CTD data elsewhere at similar depths, the water in the upper 300 m of Neny Bay is relatively saltier, warm and heavier than the water near the entrance of George VI Sound. Though the water is saltier in Neny Fjord than that of Laubeuf Fjord, the temperature is similar, approximately -1° C. The ADCP measurements on 26 and 27 May do not show a clockwise circulation in the eastern part of Marguerite Bay. Along the tracks from George VI Sound to Neny Fjord and from Neny Fjord to Laubeuf Fjord, there were several mesoscale eddies with a spatial scale of 20-30 km.

Vicinity of Avian Island. The objective of this ADCP survey was to further investigate the relationship between predator and prey. Data from tagged penguins showed that they had been feeding in the vicinity of Avian Island. We hypothesized that a more complete survey of adult krill fields would bring us added insight into the penguin behavior. In the second week of our cruise, we had a very detailed ADCP-MOCNESS survey covering Laubeuf Fjord, Lawrence Channel, Bourgeois Fjord, and the open area below Pourquoi Pas Island. We missed the vicinity of Avian Island because of time constraints and the complicated bottom topography consisting of many shallows. In an effort to increase our understanding of the current regime at the southern end of Adelaide Island the survey started from the western slope of the deep canyon at the entrance of Laubeuf Fjord. We found adult krill to be very abundant on the slope. We passed the Elliott Passage into the deep basin south of Cape Alexandra. We again found a lot of adult krill. We crossed a shallow ridge, and surveyed the deep basin from the Avian Isalnd along the Woodfield Channel. To our surprise, there was no adult krill aggregation in the Woodfield Channel. Finally, we took a straight run from Woodfield Channel, through Laubeuf Fjord to Lawrence Channel. The measurements again showed a large number of adult krill.The coastal current turns into Marguerite May around the southern tip of Adelaide Island. It joins a southward flow out of Laubuef Fjord at the deep canyon. A branch turns to the south along the western slope of the deep basin south of Cape Alexandra, and flows around the Consul Reef to form a clockwise circulation. The shoal area there is too shallow for adult krill, but it can produce a topographically-induced eddy that might act to retain krill in the nearby deep basins of Laubeuf Fjord. An acoustic survey in a topographically complex area such as that near Avian Island, consisting of many shallows and reefs, was possible only after careful planning and a good dialogue with the bridge. A forward looking sonar would have aided the survey considerably. Hanusse Bay. It is not surprising that Hanusse Bay is a very productive area. We had the biggest adult krill catches of the entire cruise there. The most abundant area for adult krill was south of Laird Island. Adult krill were concentrated between 50 m and 120 m, behind seamounts and in the deep canyon located there. The ADCP measurements showed an anticlockwise circulation. The water was saltier in the upper 100 m of Hanusse Bay than in the Barlas Channel. If there is a density driven current across the narrows between Laubeuf Fjord and Hanusse Bay, the top layer should have a southward mean flow, which is favorable for krill to be advected into Laubeuf Fjord.