Report of

RVIB Nathaniel B. Palmer Cruise 01-03

to the

Western Antarctic Peninsula

24 April to 5 June 2001

 

United States Southern Ocean

Global Ocean Ecosystems Dynamics Program

Report Number 2

1)    to conduct a broad-scale survey of the U.S. SO GLOBEC study site to determine the abundance and distribution of the target species, Euphausia superba, and its associated flora and fauna;

2)    to conduct a hydrographic survey of the region;

3)    to collect chlorophyll data, nutrient data, and to make primary production measurements to characterize the primary production of the region;

4)    to collect zooplankton samples at selected locations and depths throughout the broad-scale sampling area;

5)    to survey the sea birds throughout the broad-scale sampling area and determine their feeding patterns;

6)    to survey the marine mammals throughout the broad-scale sampling area both by visual sightings and by passive listening techniques;

7)    to map the shelf-wide velocity field;

8)    to collect acoustic, video, and environmental data along the tracklines between stations using a suite of sensors mounted in a in a towed body;

9)    to collect meteorological data; and

10)    to deploy drifting buoys to make Lagrangian current measurements.

 

            The cruise track was determined by the locations of 84 stations distributed along 13 transect lines running across the continental shelf and perpendicular to the west Antarctic Peninsula coastline (Figure 1). The cruise consisted of a combination of station and underway activities (See Appendix 1, Event Log).  The along-track data were collected from the Bio-Optical Multifrequency Acoustical and Physical Environmental Recorder (BIOMAPER-II), the Acoustic Doppler Current Profiler (ADCP), the meteorological sensors, through-hull sea surface sensors, expendable bathythermograph (XBTs) probes, expendable conductivity-temperature-depth (XCTDs) probes, and Sonabuoys.  At the stations, a CTD/Rosette, with oxygen, transmissometer, and fluorometer sensors, was lowered to the bottom, and in depths greater than 500 m, a second cast to 50 m was made with a Fast Repetition Response Fluorometer (FRRF) until an electronic failure put it out of commission.  At selected stations, a 1-m² Multiple Opening/Closing Net and Environmental Sensing System (MOCNESS) was towed obliquely between the surface and the bottom or to 1000 m if the bottom was deeper for collection of zooplankton (335 Fm mesh).  A surface ring net tow was also made at some stations for collection of phytoplankton.  Satellite tracked drifters were deployed at selected stations.

   

   

Figure 1.  RVIB N.B. Palmer cruise track showing locations of stations and along-track observations.

 

CRUISE NARRATIVE

 

            We left the port of Punta Arenas, Chile around 0900 on 24 April 2001 and began the steam east down the Straits of Magellan.  Along the route, we conducted tests of the BIOMAPER-II handling system and conducted underway noise tests of the HTI acoustic system. In the early evening, members of the testing team, Bob McCabe, Terry Hammar, and Sam Johnston, along with the pilot, disembarked at the eastern approach to the straits, and we steamed in earnest for the Western Peninsula region of the Antarctic continent.

 

25 April

            On our second day out, we steamed along the eastern coast of Argentina and cleared the tip of South America, making about 12 kts.   The trip down along the east coast of Argentina was spectacular.  In the morning, there were high mountains to the west of our course with their tops shrouded with clouds.  Then in the afternoon, we steamed through a fairly narrow strait at the southern end of Argentina, where the mountains rose steeply out of the sea on both sides of the ship (really quite beautiful and wild looking). The winds were light and the seas quite flat.  There was a lot of sun mixed with clouds throughout the day.  The air was cool, around 5EC to 8EC.  There were many sea birds flying and the seabird ecologists were already at work counting them.  They used a small plywood enclosure (a wind break) out on the port wing of the bridge.  The marine mammal people were also active in surveying the area for whales.   They were using the bridge and the ice tower some two stories above the bridge.

 

26 April

            On the morning of 26 April, we were greeted with the roll of the ship in a long period swell.  Although the sea was only choppy in a 10+ kt wind, there was a large swell coming at the ship from the west.  The sky was heavily overcast all morning with the clouds coming down to the sea surface a few hundred meters out away from the ship. By afternoon, the relatively nice seas and weather disappeared and the seas were building along with the wind.  We had sustained winds around 40 kts out of the southeast for a good portion of the afternoon and early evening, and instead of a long period swell out of the west, we had a shorter period sea coming at us from off the port bow (133E) as we continued to steam nearly due south.  The air temperature also dropped from around 4EC or 5EC in the morning to -3.6EC in the evening.  Along with the wind was occasional mixed precipitation. A CTD was scheduled for the afternoon, but was scrubbed because of the wind and seas.  A Sonabuoy was deployed, however, about the time we left the 200-mile limit of Argentina.  Also starting at the 200 mile marker, Eileen Hofmann’s group began shooting XBTs at 10 nm intervals to get temperature/depth profiles to show when and where we crossed the Polar Front. We crossed the front in the evening at 60E 10.29 S; 66E 10.66 W. 

 

27 April

            Our course took us first to Palmer Station located on Anvers Island.  In the early morning daylight, we entered the channel leading to Palmer Station to rendezvous with the R/V Gould.  The silhouettes of mountains rising out of the sea illuminated by the first light of the sun still below the horizon were breathtaking.  For many of us, it was the first time that we had seen a piece of the Antarctic continent.  The sky was mostly clear for the first time in several days and there was just enough light to give the rugged snow covered mountains a blueish tinge.  The ship moved into a deep harbor area a few hundred meters from the station dock where the L.M. Gould was tied up, dropped anchor, and shut down the main engines. From the deck of the ship, we could look into the face of a glacier only a few hundred meters away.  The winds were very much reduced, down to about 15 kts, leaving the sea fairly calm.

            The stay at Palmer Station was short and we steamed back out into the open seas of the shelf region of west Antarctic Peninsula around noon and headed for our first station in the SO GLOBEC grid.  The sunny sky gave way to overcast and then heavier cloud cover.  The winds, however, stayed down until early evening and the seas had only a moderate swell. Snow squalls started in the evening.  During the day, bird and marine mammal observations were made. This was also the first full day of Seabeam data ping editing and the UNIX and MAC workstations in the computer room were fully utilized for a good portion of the day.

 

29 April

            Work commenced at Consecutive Station #1 early in the morning (grid location 499.251) with a couple of CTD casts, a shallow one with the Fast Repetition Rate Fluorometer (FRRF) and a deep cast without the FRRF.   BIOMAPER-II was deployed into the water shortly after the last CTD cast and was towyo’d in a sawtooth pattern between the surface and 200 to 250 m as the ship steamed at about 5 kts towards the next station. For this first deployment, wind was from about 260E at 20 to 25 kts and there was a large swell under white capped seas.  Skies were mostly overcast, but occasionally the sun almost broke through.

            BIOMAPER was taken out of the water around 1130 at station #2 after doing two good towyos because of electrical problems.  We continued to steam to station #3 and the CTD work was continued, along with the bird and marine mammal observations.  We arrived at station #3 by late afternoon and following the CTD, the MOCNESS was brought out on deck and made ready for the tow.  The seas were very rough and fairly frequently the stern would dip under the sea surface and water would flood the deck.  With a light snow swirling across the deck, the MOCNESS frame was outfitted with the deflector flaps, the car batteries for the strobe light system, net response, and net bar traps.  The system was then tested. The deployment of MOCNESS was hampered some because the seas were large and the ship was pitching a lot, but the net went in reasonably well. During the lowering of the net system, electrical problems developed but were corrected before the start of the oblique section of the tow, where nets were sequentially opened and closed successfully. Just at midnight, BIOMAPER-II was deployed with the ship pitching, the wind blowing 30 kts, and a light snow still swirling around us.

 

30 April

            Electronic problems again forced the return of BIOMAPER-II back on board for more trouble-shooting.  During the recovery in high winds and seas, the cable jumped the sheaves in the slack tensioner. The arduous job of getting the wire back on the sheaves in the slack tensioner ensued.  It took about more than an hour to fairlead the wire back in place and finally recover the towed body.  Damage to the electro-optical towing cable made it necessary to cut out the bad section and re-terminate the end, which required a substantial amount of time.  Work, however, continued throughout the day with CTD casts made at stations #5 through #9, and bird and mammal observations taken during the daylight hours. The CTD work was by no means easy, given the sea state.  On several occasions, waves breaking against the ship would flood into the Baltic Room and launch and recovery was often difficult because of the ship’s motion.

 

1 and 2 May

            We completed work at 4 stations on 1 May and 3 stations on 2 May.  The work included eight CTD profiles, three 1-m² MOCNESS tows, twelve Sonabuoy deployments and three BIOMAPER II deployments.  In addition, the bird and mammal survey groups made a number of sightings. On 2 May, the wind slacked significantly over the previous highs in the 30 kt range during much of the previous days and was in the 5 to 10 kt range out of the west northwest (310E). The air temperature was about -0.1EC.  Large swells continued to make working on the stern difficult and wet, but with the reduced winds, they began to diminish.  During the late afternoon at station #15, we could see the rugged mountains of the western edge of Adelaide Island off to the east.

 

3 May

            This was a windy, cold, and snowy day.  Low clouds and fog dominated, cutting the visibility to a few hundred meters for much of the day.  More evidence of the impending winter conditions was the sighting of an iceberg as we approached station #19.  The station was offset to avoid coming to close to the iceberg and associated fledgling ice chunks.  None of us actually saw the iceberg; it was only visible on radar. Work was completed at stations #16, 17, 18 and 19.

 

4, 5, and 6 May

            On 4 to 6 May, we surveyed along transect leg #5, headed southeast into Marguerite Bay.  During 4 May, the weather was really foggy and wet.  The temperature hovered around freezing and snow was melting on the upper decks of the ship and the melt water was running over their edges, splashing on the lower decks. The wind picked up again too and the seas began to build up from the relatively calm conditions of yesterday and to some extent the day before.  A light snow was off and on all day and usually it came on stronger at night.  We completed work at stations #20, 22, and 23, which were deep stations off the shelf or at the shelf break. May 5^th started out cloudy with low visibility, but the sun broke through the clouds and with moderate winds; it was a welcome change from the previous days overcast and snow.  During the evening, snow again fell. During the course of this day, work was completed at stations #24, 25, 26, and 27.  Early on 5 May, during the run to station #24, BIOMAPER-II suffered a failure of the echosounding system and was recovered a couple of hours later at station #24. Work to repair the echosounder took several days.  May 6^th also started out with moderate winds and low visibility, but as we steamed into the area south and east of Adelaide Island around mid-day, the clouds lifted to expose more and more of the rugged coastline of the Island and the Western Peninsula, as we approached station #29 around mid-day. Black rock outcropping from snow and ice blankets which covered most of the earth here provided a stark contrast.  Shortly before coming onto station #29, we steamed past an iceberg, the first actually seen on this cruise.  We completed work at stations #28, 29, and 30.

            Late in the afternoon of 6 May, a zodiac was launched from the RVIB Palmer to carry a party of six over to the R/V Gould.  A spare MOCNESS underwater unit on the Palmer was “loaned” to the Gould to replace one that had stopped working. And in return, the Gould sent back some other parts that were needed on the Palmer.  Later in the evening, a second rendezvous took place to pick up another spare part from the Gould, this time one needed for the repair of the echosounder in BIOMAPER-II.

 

7 May

            We were greeted on 7 May by a magnificent sunrise and a grand view of the Western Peninsula’s mountains coming right down to the eastern edge of Marguerite Bay with its snow fields and glaciers.  The sun, still not quite up at 1000, backlit the mountains and gave them a golden halo in the region of sunrise. The air felt cold, although it was still right around 0EC, and there was ice on the helicopter deck - frozen melt water from the snow of the last few days.  Because of the calm seas, the aft main deck was dry, no water sloshing back and forth as was usually the case.  The wind was out of the northeast (075E) at 10 to 15 kts. The partly cloudy skies and considerable sunlight made for good visibility. This, combined with low winds and seas, made for an optimal work environment.  We completed work at stations #31, 32, 33, and 34 and along the trackline between the stations, including deployment of one satellite tracked drifter and five Sonabuoys, five CTD casts, one MOCNESS tow, and bird and marine mammal survey observations.  Steaming between stations was slowed significantly because Marguerite Bay has many shoal areas that are poorly charted.  The Seabeam bathymetry data being collected on this cruise will help remedy this, but the officers on the bridge exercised appropriate caution as we moved through the uncharted areas.

 

8 May

            May 8^th was by far the roughest day of the cruise to this point.  Shortly after the deployment of BIOMAPER-II around 0100 at Station 35, the wind and seas began to build and continued building the entire day.  By early evening, sustained winds were over 40 kts, with frequent gusts in the 50-kt range and occasional gusts over 60 kts. CTD casts were not done at survey stations #37, 38, 39, and 40.  Instead,  XCTDs were substituted, but even they had to be deployed from the 01 deck because the main deck was awash.  BIOMAPER-II, however, was left in the water and continued to towyo between stations as we steamed along the survey line at speeds of 4 to 5 kts.

 

9 May

            It was not until the early morning of 9 May that the winds began to drop into the low 30-kt range and the seas started to lessen. At station #41, we were finally able to resume our normal station operations in spite of continued strong winds of 33 to 38 kts out of the north-northeast (020E). This was an improvement over the conditions that had prevailed the past 24 to 36 hours.  Skies were still darkly overcast with the clouds coming down to the sea surface such that visibility was only a few hundred meters.  There was a lot of sea spray as the ship quartered into the sea during the transect out across the shelf on survey transect 6 but very little precipitation. We completed work at the shelf stations #40 and 42 and the offshore deep station #41.  The station work included deployment of two sonabuoys, two XBTs, two XCTDs, two CTDs and one MOCNESS tow.  BIOMAPER-II was towyo’d between the stations, and bird and mammal observations were made along the trackline. The latter station had depths of about 3000 m, and the CTD to the bottom and the MOCNESS tow to 1000 m took about a quarter of the day (6 hours). 

 

10 and 11 May

            A near repeat of the day before yesterday, 10 May had sustained winds in the mid-40-kt range and gusts to 50 kts. We were again in a strong gale and it was a classic Antarctic stormy day.  The cloud cover was 100%. Although daylight came with the skies clear overhead, 100% cloud cover enveloped the area by the afternoon.  Before noon at station #44, a large iceberg was drifting some three miles from where the CTD was being deployed.  From the bridge, it looked like a ghost-ship out in the mists.  The winds intensified during late afternoon and into the evening, and most of the work was centered on drops of XCTDs and the towyoing of BIOMAPER-II between 30 to 40 m and 250 m, although we were able to do a CTD at station #44.   Problems with the BIOMAPER-II towing wire around midnight on 10 May and continued high winds and seas caused the cancellation of the CTD work at stations #45 to 48 and the use of XCTDs instead.  By mid morning on the 11^th, however, the winds had died down and the seas were dramatically lower.  In part, this was because we had entered the southern end of Marguerite Bay.  Here, the sea surface was ice-covered and many smallish icebergs were present that dampened the underlying storm swell.  Low clouds and a fine mist, which froze to the metal surfaces of the ship, made visibility limited for most of the day.  Still, it was a spectacular introduction to the beauty of the Antarctic winter seascape. The calm conditions made for excellent Seabeam bathymetric data acquisition and the data exposed the steep walls of the canyon that cuts into the heart of Marguerite Bay.

 

12 and 13 May

            May 12 was a good day for making oceanographic observations.  It began with calm conditions and the steam between stations #52 and 53, which took about nine hours, was marked by sea ice, bergie bits, and icebergs.  By mid-afternoon, we had arrived at station #53, right in the middle of the most impressive ice field with huge icebergs that we had seen thus far this cruise. The ship had to thread its way through the monster icebergs for several hours and they increased with the approach to station #53.  Captain Mike Watson expressed the possibility that there might be a need to change the station location because of the “wall” of ice we seemed to be approaching. But we were able to reach the position and conduct the planned work. A large number of seals, whales, and sea birds were also present in the area around this station. As a result, in addition to a CTD, the Zodiac was used in an attempt to get close to one of the minke whales sighted in the area to do a biopsy and to deploy a sonabuoy, and to collect sea ice samples. 

            During the steam to station #54, while towyoing BIOMAPER-II, the calm conditions gave way to increasing winds and seas.  By the time we reached station #54, conditions had deteriorated to such a degree that we were unable to recover the towed body safely or to do a CTD and MOCNESS tow.  Only an XBT was deployed and a surface water sample was taken. BIOMAPER-II was put down to a safe depth and the ship commenced to steam into the seas for about 12 hrs while waiting for the gale (sustained winds of 40 to 45 kts) to blow through.

            In the early morning on 13 May, the winds had come down to a point where the ship could be turned and headed to station #55, although the recovery of BIOMAPER-II had to wait until we had nearly reached the station.  At station #55, we were able to do a CTD and MOCNESS tow, and then re-deploy BIOMAPER-II for the transit to stations #56 and 57. 

            In spite of the closely spaced wind events and a significant amount of down time, during the 12^th and 13^th, we were able to do five CTD casts, twelve sonabuoy drops, one MOCNESS tow, two surface water collections, and two ring net tows for phytoplankton.  Between stations, we continued to acquire acoustic, video, and environmental data with BIOMAPER-II and to make bird and mammal observations.

 

14 May

            May 14 began with BIOMAPER-II being towyoed to station #57 and the sea conditions again deteriorating.  A Terascan image of our region showed an intense low pressure system to our west and heading our way.  By the time we reached the station at about 0500, it was clear that the towed body should be brought on board because conditions were likely to get much worse.  Unlike the missed recovery at station #54, this time we were able to safely bring the towed body on board.  The MOCNESS scheduled for this station was scrubbed, but a CTD profile to the bottom was made.  Shortly after that as we started to steam out across the Antarctic Circumpolar Current,  the weather window closed and seas were too rough along the route to permit us to do our planned work with the CTD or BIOMAPER-II.   XBTs and XCTDs plus surface water samples were the only game in town for stations #58, 59, 60, 61, 62, and 63.  So we steamed on with BIOMAPER-II on deck while the offshore survey was being done in very rough seas. By late afternoon, we had reached the end point of the offshore stations and turned onto a much more comfortable course for the steam to the next shelf station # 64, some 37 miles away.

 

15 & 16 May

            May 15 was a relatively benign day compared to the preceding few days and a welcome respite from the marginal working conditions.  Winds were steady at 10 to 20 kts throughout the day and low clouds continued to prevail, but there was only limited precipitation in the form of snow flurries, usually at night. There was still a large swell running - a remnant from yesterday’s high winds. May 16^th was the second day of reasonably good working conditions.  Winds for the most part remained under 20 kts until the evening when they began to increase, and the swell was down from yesterday.  Much of the day was spent near the coast of Alexander Island, working in the vicinity of stations #68 and 69 and steaming in between them. An entourage of about 80 seals stayed swimming alongside the ship for several hours after leaving station #68.

            Work over the course of these two days included seven CTDs, three MOCNESS tows, seven sonabuoy and one satellite-tracked drogue deployment, the bird and mammal surveys when possible, and BIOMAPER-II towyo’s between each of the stations.

 

17 May

            The good working conditions of the last two days gave way to less favorable conditions and a good portion of 17 May was spent with near gale conditions (28 to 33 kts) as we steamed out to the edge of the continental shelf on the tenth transect of the broad-scale survey.  The anemometers were not giving very accurate readings because of ice build-up on the propellor blades last night.  In the early evening, the winds dropped to 20-25 kts and working conditions improved enough to permit the CTD to be deployed.  BIOMAPER-II, which was deployed at station #69 on the 16^th of May, remained in the water throughout the day and was towyo’d between the stations.  During the day, work was completed at stations #71 to 74 and included two CTDs, two XCTDs, four XBTs, two sonabuoy deployments, two ring net tows (one ended up in the ship’s propellor and the net was destroyed), three bird observation sessions, and the BIOMAPER-II towyo’s.

 

18 May

            By 18 May, a certain monotony had set in as the around-the-clock survey work at and between the stations continued along with the seemingly endless cloudiness and fog, often accompanied by freezing mist or snow flurries.  The shortness of the day was also the subject of conversation, especially for the bird and mammal surveyors, who need the ship to be steaming between stations during a period when there is daylight to make their observations. The weather had moderated significantly from yesterday and the seas were much reduced, enabling all of the programmed activities to take place. Winds were out of the southeast (137E) at about 20 kts and the air temperature was -1.5EC.  We came close to Alexander Island during the early afternoon, but it was cloudy and foggy and there were snow flurries falling, so there was not much to see.  From the bridge, it was possible to get a glimpse of the Island through binoculars while there was still some light (the sun was up for about 3.5 hours on the 18^th).

 

19 and 20  May

            The weather continued to be overcast with no breaks to let a little blue sky through during the short daylight period. The winds, which were in the less than 20 kt range for most of the morning, were up around the 30 kt range in the afternoon, giving rise to a rougher sea and marginal working conditions.  Still no work was canceled. On 20 May, the weather was quite good for a change.  Some blue sky was even showing for a portion of the daylight hours. Winds were light and the temperature hovered around freezing.  In the early afternoon on the 20^th, the broad-scale survey ended at station #84.  The last activity was a MOCNESS tow (#18).  Around noon from the bridge, Charcot Island was visible to the east with its black rock surfaces showing where it was not covered with a mantle of snow and ice.  Also barely visible was an ice shelf that extended out into the sea away from the island to the southeast.

            The end of the survey also brought an end to the systematic way in which the scientific effort was conducted.  The list of tasks to be completed in the remaining days was large, but when and where they could be accomplished depended to a large extent on finding the right conditions.  The first priority was to find an area with penguins and whales that could be approached using the Zodiac inflatable boats.  To this end, we decided to steam into an embayment east of Charcot Island, which offered some protection from the prevailing northeast wind and was a likely place for some pack ice.  On the way to the embayment east of Charcot Island, the CTD group took the opportunity to define the coastal current structure and hydrography a bit more by taking two CTDs along the way, including one at a location occupied by Stan Jacobs (Lamont-Doherty Earth Observatory) in March 1994, and a series of XBTs.  At the Jacobs’ location, patches of ice chunks coalesced around the ship.  By midnight, as we steamed into the bay from the Jacobs’ location, many large icebergs were visible on radar and the sea ice was thickening.

            During the 19^th and 20^th , work was completed at stations #79 to 84 of the survey and two additional stations on the way to the embayment east of Charcot Island, including eight CTDs, five XBTs, five sonabuoy deployments, three ring net and two MOCNESS tows, two sets of bird and one set of whale observations, and BIOMAPER-II towyos between each of the stations.

 

21 May

            The Antarctic is known for its snow- and ice-covered mountains, its grand ice shelves, and its frozen embayments, and on 21 May, we had the pleasure of working in the latter and having distance views of the former in remarkably good weather (easterly winds at less than 10 kts and an air temperature about -3.8EC).  Most of the effort was devoted to finding a site where bird and mammal observations could be made from Zodiac inflatable boats and where the first of the ROV deployments could take place for under-ice krill studies.  The area chosen was the embayment bounded by Charcot Island on its north, the Wilkins Ice Shelf to the east, and Latady Island to the south.  We had intended to do a station in the vicinity of the Wilkins Ice shelf edge where it was hoped that we might find whales and penguins, as well as do a CTD survey along its face.  In the early morning of 21 May, however, we found ourselves ploughing ahead at about 3.5 kts into heavier and heavier pack ice and we were still some distance from our intended destination.  Earlier in the transit as we entered the ice field around midnight, there were a lot of sea birds in the high intensity spot lights which are always on and focused ahead of the ship after dark, but in the early morning light we were not seeing any birds or mammals. The bird and mammal researchers did not see any sense in going further, so after entering a lead that had probably recently opened and then frozen, we stopped to take an XBT and get ice samples.  The latter were obtained by putting the personnel carrier with three people aboard over the side with the large main crane and landing it on the ice.  Then we turned around and headed back the way we came, but a mile or two to the north.  By about 1100 hours, we again reached open water and the ice edge. The zodiacs were readied and then the first was deployed for the whale group to use and the second ferried the bird party, after some delaying outboard engine problems were resolved.  During the Zodiac deployments, the ROV was being readied and upon their return, the ROV was deployed for testing and trial runs under the ice late in the afternoon.  Unfortunately, a leak developed in the underwater unit of the ROV and fried some of the electronics.  Fortunately, the parts that fried were not essential and could be by-passed.  By late evening as we were steaming to a new sampling location on the north side of the Wilkins Ice Shelf, the ROV was repaired and ready to again be used.

           

22 May

            This was another day for observing the grandeur of the Antarctic ice-scape. In the pre-sunrise light, we steamed towards the Wilkins Ice Shelf between the northern part of Charcot Island and the southern end of Rothschild Island. The first light was before 0900 and the glow in the sky was just behind the mountains of Rothschild Island, so that the mountains were silhouetted as black against a rose-hued band of light just above their crests and the dark grey of the water leading up to their base.  The visibility was very good with high thin clouds overhead and even a bit of clear sky. The winds remain light and out of the southeast and the air temperature was around -5.3EC.  It was an ideal period for looking for bird and mammal krill predators to study. At a pre-dawn (0845 local) meeting on the bridge of the Palmer, a consensus was reached that we should head towards the Wilkins Ice Shelf between the northern part of Charcot Island and the southern end of Rothschild Island in search of penguins and whales to study from the Zodiacs.  In spite of the good meteorological conditions, we spent most of the day in unconsolidated pack ice that was too thick to effectively operate the Zodiacs, yet too unstable to allow a person to walk on.  We did go deep into the ice pack and got to a place where there were few if any flying birds around the ship and only some occasional seals (leopards and crabeaters) lying on ice chunks.  So around noon, we headed towards Rothschild Island and the site where scientists on the R/V L.M. Gould reported seeing quite a few penguins and seals. Only a couple of penguins were sighted during the daylight period and also a pair of minke whales. 

            In the mid-afternoon, the ROV was again ready for deployment after its electronics had been repaired from the damage caused by a seawater leak. The ship’s track was altered to put us back into the heavier pack ice.  At a location just off some very large icebergs, we stopped and after clearing a hole in the ice pack, the ROV went into the water.  The ROV deployment went very well for an hour or so, and krill larval forms and adolescents were observed moving around on the underside of the ice pack.  The ROV was brought back on the deck in order to adjust the stereographic video cameras to provide a better view of the underside of the pack ice, but on deck, it was observed that the ROV again had leaked water in the electronics housing.  This put an end to the deployment.  We steamed at modest speed during the night through varying degrees of pack ice cover to a rendezvous point with the R/V L.M. Gould in Lazarev Bay.

 

23 May

            Lazarev Bay is bounded on the southwest by Rothschild Island, on the southeast by a small section of the Wilkins Ice Shelf, and on the northeast by Alexander Island.  It had been the work site for the R/V Lawrence M. Gould for the past several days.  Based on information from scientists on the Gould, the Bay provided us with the possible opportunity for conducting bird (especially Adélie penguin) and mammal studies, and ROV studies of under-ice krill abundance and behavior.  We arrived in the Bay late in the night on 23 May with winds out of the west-southwest (145E) at about 12 kts and the air temperature at -2.6EC.  Work began immediately with the ROV to study the distribution and abundance of krill living in the vicinity of the sea ice bottom surface.  This work went smoothly, in spite of a number of seals that found the ROV a subject of interest, until the operation was brought to a halt by water leaking into the underwater housing.  For the daylight operations, we were again in search of penguins and whales, but with the continuous pack ice in the Bay and the low possibility of finding whales present, the focus was on the penguins.  Based on information from Bill Fraser, the Adélie penguin expert on the R/V Gould, we decided to go deeper into the Bay where he said the ice was thicker and we might be able to get to an ice shelf where penguins were likely to come up out of the water around noon after having finished their daily feeding.  About 0900, we headed into the brash ice which flowed around large icebergs distributed throughout the Bay.  Along our route, we did encounter Adélie penguins in ones or twos, occasionally more, but the unconsolidated brash ice they were on was too thick for the Zodiacs and too unstable for a person to walk on. So we could not get to them. Finally about 1300, we got to a point where there were too many icebergs to maneuver around and so we turned to start the trek back toward the entrance of the Bay.  On the way, we came past an immature Emperor penguin on a small flow.  Also present in moderate numbers were several species of seals.  Late in the day, the ROV was again deployed and a series of transects originating from a central point were run under the ice to access krill distributions. The nighttime was used to steam to the next station approximately 50 nm northeast of Lazarev Bay, shooting XBTs at 10 nm intervals along the route.

 

24 May

            On 24 May, we worked in the vicinity of station #53, where we previously had observed many seabirds and seals and a number of whales and what we thought was a dense acoustic scattering layer of krill 80 to 120 m below the surface.  Humpback and minke whales were both heard (via sonabuoy) and sighted a couple of miles before reaching the station location, so we steamed back towards that site until we came across them again. By about 1000, both Zodiac inflatable boats were away, one headed to where the whales were to try to get biopsy samples and the other went over to a patch of brash ice where there was an attempt to catch petrels and see what they were eating. The weather was ideal with no wind, little swell, and for most of the day, a glassy sea surface.  Air temperature was -3.4EC.  Visibility was excellent with high clouds overhead and the work on this day was done with the looming peaks of the mountains of Alexander Island and a large glacier coming down to the shore line just a few miles to our east.  When we were here during the broad-scale survey, the clouds were down to the water and the mountains were not visible. It was a good day for the whale group.  Ari Friedlaender obtained biopsy samples from three humpbacks and one minke whale during the Zodiac forays out into the still waters where the whales were diving among the scattered patches of brash ice.  The bird people did not have as much luck. Chris Ribic and Erik Chapman went out to brash ice patches to try and trick petrels to come close to a cod liver oil-soaked red cloth out on a piece of sea ice so they could net them.  However, the petrels were too quick and none were caught. They did, however, make some interesting biological observations.

            The Zodiacs were out of the water as the last of the daylight faded around 1515.  BIOMAPER‑II went into the water shortly after that (1530) at a location centered where the work with the Zodiacs had been done in order to map a subsurface layer of krill that we had observed earlier in the cruise. This work went until 2130 and was followed by a CTD.  Around 2300, we got underway for the first Automated Weather Station (AWS) deployment site.

 

25 May

            The work on 25 May began with a XBT/CTD section from the northern part of Alexander Island across to the Kirkwood Islands to better define the origins of the coastal current which flows along the coast to the southwest of Marguerite Bay.  This section was finished in the area where the first of two Automated Weather Stations were to be installed (Appendix 6).  We arrived at the Kirkwoods about 0930 and in the dim first light, we started looking at the various islands through binoculars and assessing their prospects as the installation site.  There was only one big piece of rock and it was mostly ice- and snow-covered. We very slowly made our way into a position about a mile or so away from the biggest island after we ascertained that the other smaller pieces of exposed rock were almost certainly covered with water during periods of high winds and waves.  While we were making the observations from the bridge and coming up with a decision about the AWS prospects, the Marine Technicians were loading up the AWS equipment into the Zodiac to be ready if given the go.  The weather was cooperative with winds in the 10 to 15 kts range out of the south (170E) and good visibility (high broken clouds most of the day).  It was cold, however, with the temperature around ‑3.0EC.  Around 1000, the decision was made to launch the Zodiac and make the trip over to the largest island.   A party of six made the trip to the island with Bob Beardsley in the lead and after some difficulty, they found an acceptable landing site and made it onto land. Some five to six hours later, they returned having successfully moved the equipment to the top of the island and done the AWS installation. The second Zodiac was also launched to scout the area around the islands looking for whales and to launch a sonabuoy away from the Palmer, but no whales were seen.

            Once the Zodiac parties were back on board, we set sail for survey station #37, arriving about 2230.  A MOCNESS was tow was completed there just after midnight with winds in the 25 to 30 kts  range coming from the southwest.  This MOCNESS tow and a subsequent one was done to fill a gap in the broad-scale survey tows that resulted from cancellations caused by high winds and seas.

 

26 and 27 May

            The weather early on 26 May made working conditions difficult.  The winds were still around 30 kts during a MOCNESS tow that was completed at survey station #44.  They died down by 0900 and they remained in the 15 to 20 kts range during a survey of the bathymetry around the B-line of the current meter moorings spanning the canyon in the outer portion of Marguerite Bay, which took most of the daylight period.   Following the end of the mooring survey around 1830, a quick steam brought us to the position of survey station #27.  This location was where the BIOMAPER-II echosounder had failed a couple of weeks earlier. The towed body went into the water about 1930, but once in the water, one of the VPR cameras showed up as not being adjusted properly.  So the towed body was brought back on board to adjust the camera.  After the second launch, the towyo trackline went on a course from the vicinity of grid survey station #27 to station #28 and then over to station #31, a distance of about 50 nm.  There was a turn back towards the Faure Islands before it was brought back on board around 0700 on 27 May to make the steam over to the Islands where the second AWS was to be installed. The bathymetry along the transect line was extremely variable ranging from 700 meters to 140 meters. This made it impossible to towyo deeper than about 200 meters.

            The Faure Islands are located  in the northern end of Marguerite Bay and one of the largest is named Dismal Island.  Dismal Island was first charted in 1909 by Charcot and named in 1949 because of its desolateness and loneliness. On the day of the AWS installation, it fit that description.  The Automated Weather Station was installed at the top of a small island (approximately -68E05.5 S; -68E 48.75 W) just to the east of  Dismal Island (Appendix 6).  Winds during the day were in 18 to 22 kts range out of the north-northeast (020). While the heavy cloud layers and occasional snow kept the region looking gloomy during the short period of daylight, working conditions were reasonably good and the installation was done before the last light of the day had disappeared.  The small group of islands was also a good working site for those interested in the diet of penguins.  A couple of hours after the first Zodiac load of gear and people were offloaded onto the island, a second group was ferried over to a second small island adjacent to the first where penguins had been sighted. They managed to catch six Adélie penguins and obtain stomach samples before darkness forced their return to the Palmer. 

            The nighttime period was devoted to surveying the Lebeuf Fjord with BIOMAPER-II, looking for high concentrations of adult krill. Once located, an adaptive sampling approach was taken in an attempt to define the boundaries of the patch.

 

28 May

            During the 28^th of May, the krill patch study continued in the northern portion of Marguerite Bay throughout the day and into the evening. The winds for the most part were in the 10 to 15 kts  range out of the north (011E) and air temperature was -0.7EC.  In addition to using BIOMAPER-II to define the krill patch structure with the acoustics and Video Plankton Recorder, two horizontal MOCNESS tows were done within a portion of the krill layer that was well defined.  A CTD profile was also made within the area where high concentrations of krill were observed.  During the day, the winds were a steady 25 kts out of the north-northeast (020E) and this precluded putting the Zodiacs over the side to enable the whale group to attempt additional whale biopsies.  There were several humpbacks in the area that were sighted from the vessel and were heard on the sonabuoy transmissions close to where the highest concentrations of krill were observed.

 

29 May

            Between midnight and 0500 on 29 May, the krill patch study was concluded with the completion of two additional MOCNESS tows.  Then the ship steamed over to the eastern side of Marguerite Bay to the San Martin Station manned by Argentina. Two days earlier, we received an invitation by Base Leader, Captain Carlos Martin, to visit the station. They had discovered our presence in the area by chance and welcomed the opportunity to have us see their station. We arrived around 1030 and anchored about a mile from the station (-68E08.765 S; -67E06.374 W).  Then the Zodiacs were used to ferry most of the scientists and a number of the crew to the base for a 3 to 4 hour visit.  The winds were quite light and the sea nearly smooth and the air temperature was -3.8EC.  But we could not see much of the mountains around the station because of the snow which was coming down in a light to moderate fashion. 

            The station has a series of reddish buildings which serve a variety of purposes and there are a number of antennas distributed as an array throughout the station area.  It has been in existence since 1951.  The work of the station is mainly geophysical and astrophysical. The nineteen personnel, all male, are there for a year, and they had just completed the first two months of their stay, which ends next March.  Most are in the military and the two civilians present represent the scientific contingent.  The others provide support and maintain the station. Communication now is by radio, but a satellite system will be in place by July to enable data telemetry and voice communication.

            Our hosts gave us a tour of the station that included the science laboratory space, living quarters, the game room, the food storage buildings (freezer and dry stores), the garage with the skidoos and 4-wheel drive snowmobile, the carpentry shop, and the helicopter pad.  All of their supplies come in once a year when the exchange of personnel takes place.  There is also a medium-sized two story structure which the Captain referred to as their “House”.  The downstairs portion had a “mud” room for taking off boots and winter outer clothing, an exercise room, and other storage spaces. Upstairs was a cozy environment for eating, socializing, and entertainment (movie viewing, etc.). Wood-lined ceiling and walls gave the place a warm feeling. And in the middle of the biggest room was a table filled with plates of specially prepared foods and an assortment of drinks. Filling this room were about 35 visitors from the Palmer and 15 or so of the residents.  Our host described highlights of Argentina (which he called our “temptation”), the history of the station, and its mission today.  Then about 1300, there was an exchange of small gifts, the eating of the food, and the cutting of a cake made for the occasion.  It was a very good time and it was clear that they were very happy to have us accept their invitation.  Likewise, our group very much enjoyed their hospitality and the opportunity to visit their station.

            With the light fading fast, the group started moving down to shore, a boat load at a time, and by 1530 or so, all had returned to the Palmer and the anchor was hauled up.  By 1600, we were back to steaming for the next station and the final few of days of work before we head for Punta Arenas.

            We steamed about 40 nm over to just north of the Faure Island group and began a CTD/BIOMAPER-II section that was intended to sample the coastal current along the southern end of Adelaide Island.   First, a CTD cast was done at consecutive station #91 in about 200 m of water and then BIOMAPER-II was put into the water for towyoing to the next station about 8 nm away.  The bottom topography, as we steamed to station #92, was very shallow and much like a roller coaster.  Most of the shallow topography was not on the charts and it was slow going. We reached the second CTD site around midnight on 28 May.  After the transects were started, word was received from the University of Wisconsin, Antarctic Meteorological Research Center that the software in the AWS on Kirkwood Islands that logs the wind speed was faulty and needed to be fixed to enable wind speeds above 5 m s^-1 (10 kts) to be recorded.

           

30 and 31 May

            The series of short transects and CTD stations to define the coastal current off the southwestern portion continued during the early morning of 30 May with the weather a bit better than the previous day.  The winds were light, < 10 kts, and there was no snow, although it remained well below freezing, around -4.0EC.  High broken clouds allowed much better visibility. BIOMAPER-II was in the water and we were just finishing the CTD at station #94 when a facsimile transmission was received around 0900 that had the code to do the software fix for the ailing AWS wind speed logger.  The survey was discontinued and with the gear on board, we steamed for the Kirkwood Islands to repair the AWS.  By 1230, we had arrived at the Kirkwoods.  In spite of less than optimal conditions, a Zodiac with a party of six made its way through the choppy seas to the Island.  They had some difficulty finding a landing spot because the surf was up much more than 5 days earlier when the installation took place.  Eventually, a spot was found where they could get ashore.  Around 1415, as the light of the day was fading, the group finished the job and were picked up by the Zodiac, which had been waiting just offshore. While the island party was doing the AWS repair, a test of the track point system was being conducted on board the Palmer to see whether a repair of the cable was successful.  Jan Szelag had discovered at least one of the wires was broken in the base of one of the cable terminations and it may have been the reason for the poor performance of the track point system to date.  A final test was done when the Zodiac returned to the side of the Palmer.  A transducer was put at 3 m over the side of the inflatable and boat then drifted out away from the side of the ship while track point system tried to follow them.  This was done successfully.

            By 1600, the ship was back underway for the place where work had been stopped in the morning.  BIOMAPER-II was deployed when the break-off point was reached about 1930 and towyos were recommenced along the trackline to station #95 where the next CTD was done.  The last of the coastal current CTD stations (#99) in the section was completed about 1000 on the 31^st after BIOMAPER-II was brought on board.

            We then steamed northeast to the location of mooring A1 (-67E01.134S; -69E01.217 W) in calm seas and light winds (about 10 kts out of south-southwest at 215E) and with an air temperature (-5.0EC) that was fostering the development of sea ice.  Large patches of sea ice occurred along the trackline. Early in the transit, the coastline of Adelaide Island was visible with the cloud line nearly down to the water in some places and peaks showing in others.  The sun, low in the sky, was out and casting shadows on the mountains and on the sea ice chunks as they flowed by.  The bright rays created a rainbow where snow showers were coming down between us and Adelaide Island. 

            During the Seabeam survey of the bathymetry around mooring A1 in the early afternoon, the skies cleared over the mountains of Adelaide Island and a magnificent view appeared.  The tall white snow-cloaked mountains with almost no rock showing merged at their base with the Fuchs Ice Piedmont. This tremendous ice sheet is 610 meters or so high at the top and 30+ meters tall where the ice shelf meets the ocean.  Many crevasses were evident along the margin of the shelf and it became evident why there were so many icebergs present in the local waters.  From our work site, the ice shelf edge was about 9 nm away and the mountain peaks were about 20 nm. During the period of the survey (1300 to 1430), the sun, which had shown so brightly at the beginning, was setting by the time we left the site and deep shadows developed on the mountain sides that disappeared a short time later, leaving only the peak tops lit  for a few minutes before they too faded into dusk.

            Following the Seabeam survey, we steamed to the near shore end of broad-scale survey transect #2.  But instead of stopping at survey station #6, the ship was moved in towards the ice shelf edge to where the water was about 100 m deep to define the inner edge of the coastal current in this area.  This depth occurred about 1.5 miles from the ice cliffs.  There a CTD profile was made around 1700 and then the ROV was deployed to look for krill under the sea ice, which was 10/10 in the area.  BIOMAPER-II was put back into the water after the station work was completed and the towyos along survey transect line #2 were started about 2200 on the 31^st.  Along the trackline, XBTs were dropped at 10 nm intervals.

            The 31^st of May was also noteworthy because of a visit from King Neptune.  Pollywogs, those who had never crossed the Antarctic Circle, were sought to attend his appearance, which included his entourage of experienced circle crossers. The induction process that ensued was interesting, some might say fun, and largely enjoyed by all who participated in it.

 

1 June

            June 1 was the final day for collecting data in the Southern Ocean GLOBEC broad-scale survey of research site. A BIOMAPER-II towyo section along transect line two was completed about 1430.  This line was re-run because during the original survey, technical problems caused us to miss getting data with the towed body along this line.  After a short steam, two deep CTD stations were completed at locations beyond the continental shelf that were extensions of the lines of stations on survey transects one and two. The CTD came back aboard after the final cast about 2230.  The weather during the day was ideal for surveying birds and mammals. Skies had a high cloud overcast and visibility was very good. Winds remained under 10 kts out of the east-southeast for most of the day and the air temperature was -5.0EC.  Thus, the seas were calm and once clear of the inshore area, free of sea ice.

 

2-6 June

            The RVIB N.B. Palmer left the Western Antarctic Peninsula continental shelf research site of the U.S. Southern Ocean GLOBEC program and began the trek back to Punta Arenas, Chile around midnight on 1 June.   During the first day of steaming into the Drake Passage, we were met with near gale winds of 28 to 34 kts out of the east (090E).  Although the last of the station work was completed yesterday, XBTs were taken at a number of locations as we crossed the Polar Front and sonabuoys were deployed as needed to listen for marine mammal calls and sounds until reaching the 200-mile limit of Argentina. During the transit, the scientific party packed samples and gear either for storage in Chile awaiting the next cruise which begins in mid-July 2001 or for shipment to the U.S.  Leaders of the research parties wrote up sections for the cruise report. We arrived in Punta Arenas on the afternoon of 5 June, some 18 hours ahead of schedule.

 

INDIVIDUAL PROJECT REPORTS

 

1.0  Report for Hydrography and Circulation Component

 

Hydrography Group Personnel: Eileen Hofmann, Robert Beardsley, Susan Beardsley, Mark Christmas, Susan Howard, Baris Salihoglu, Rosario Sanay, Aparna Sreenivasan

 

1.1 Introduction

 

            The overall goal of the U.S. Southern Ocean GLOBEC program is to elucidate circulation processes and their effect on sea ice formation and Antarctic krill (Euphausia superba) distribution and to examine the factors that govern Antarctic krill survivorship and availability to higher trophic levels, including penguins, seals, and whales.  Consequently, a primary objective of this first U.S. SO GLOBEC broadscale survey cruise (NBP01-03) is to provide a description of the water mass distribution and circulation on the west Antarctic Peninsula (WAP) continental shelf in the region of Marguerite Bay.

            Historical hydrographic data for the region covered  during NBP01-03 are limited.  However, these data show that the water masses in the area consist of Antarctic Surface Water (AASW) in the upper 100 m to 120 m, a cold Winter Water layer at 80 m to 120 m, and a modified form of Upper Circumpolar Deep Water (UCDW) that covers the shelf below the permanent pycnocline at 150 m to 200 m.  The UCDW, which is the source for the modified water on the WAP shelf, is found at the outer edge of the continental shelf at depths of 200 to 600 m.  Thus, the first objective of the hydrography component is to fully describe the water mass distribution on the WAP continental shelf.  The hydrographic distribution from this cruise will also provide a base line for assessing changes observed in subsequent cruises. 

            Circulation in the study region, which has been inferred from the limited hydrographic observations, suggests a clockwise gyre on the continental shelf near Marguerite Bay, upwelling of UCDW at specific sites in the study region, and across-shelf flow of UCDW into Marguerite Bay at depth.  However, the details of the circulation and the spatial and temporal variability of the flow remain to be determined.  Thus, the second objective of the hydrography component is to provide a description of the large-scale circulation for the portion of the WAP continental shelf included in the study region.  The resulting circulation distribution can then be compared with drifter measurements, moored current measurements, and circulation distributions derived from theoretical models.

 

1.2 Data Collection and Methods

 

1.2.1 Data Distribution

            The hydrographic data set was collected from individual stations that were aligned in across-shelf transects that ran perpendicular to a baseline situated along the coast.  The base survey grid consisted of thirteen across-shelf transects and 84 stations.  However, as described below, 17 of the original survey stations were not occupied due to weather, and survey station #21 was dropped from the grid after deciding it was not necessary to extend the survey grid further beyond the shelf break.  The stations were run from north to south, starting with the outer shelf station on survey transect one.  Spacing between transects was 40 km; station spacing along individual transects varied from 10 to 40 km.  As the cruise progressed, 18 additional stations were added to the survey grid to provide environmental data for specialized studies and to provide coverage in regions not included in the original survey grid.  As a result, a total of 84 hydrographic stations were occupied during the cruise, giving the same number of stations as originally planned. 

 

 

1.2.2 CTD and Water Samples

            The primary instrument used in the hydrographic work was a SeaBird 911+ Niskin/Rosette conductivity-temperature-depth (CTD) sensor system.  The CTD included dual sensors for temperature and conductivity.  Other sensors mounted on the CTD-Rosette system were for measuring dissolved oxygen concentration, transmission (water clarity), fluorescence, and photosynthetically active radiation (PAR).  All CTD profiles were done to within 5 to 20 m of the bottom, depending on weather and sea state conditions.  At stations where the bottom depth was more than 500 m, a second CTD cast was made to 50 m with a Fast Repetition Rate Fluorometer (FRRF) mounted on the Rosette.  In all, 102 casts were made with the CTD (Appendix 2).

            The 24-place Rosette was equipped with 10-liter Niskin bottles.  The number of discrete water samples taken on each cast was variable (Appendix 3).  However, samples were generally taken at the surface and bottom, above and below the oxygen minimum layer, at the oxygen minimum layer, and at a series of standard depths between 50 m and the surface. Additional water samples were taken in order to better resolve specific features seen in the vertical profiles.

            On each cast, water samples were taken at several depths for salinity determinations to be used for calibration of the conductivity sensors on the CTD (Appendix 3). A total of 431 bottle samples were collected for salinity analysis during the 102 CTD stations made on NBP01-03. The bottle conductivities were measured during the cruise using the NBP Guildline AutoSal 8400B No. 2 laboratory salinometer, and the values converted to salinity using the MATLAB program read_bot_data. The CTD primary and secondary temperature and conductivity sensors were compared to look at internal consistency, and the salinity values computed using the primary sensor set were then compared with the bottle salinities.  Figure 2 shows these comparisons for the 431 samples, and a summary of the mean difference and 95% confidence limit of the mean difference are as follows:

 

Differences between the CTD primary (0) and secondary (1) temperature (T) and salinity (S)  sensors,

T0 - T1 = 0.0013 +/- 0.0047 °C,  for N=399

S0 - S1 = -0.009 +/- 0.0047 psu, for N= 413

 

Differences between the primary sensor (0) and secondary (1) CTD sensors and bottle (b) salinity values,

 

S0 - Sb = -0.0002 +/- 0.0053, for N = 388

S1 - Sb = -0.0007 +/- 0.0053 psu, for N = 390

 

Difference values greater than +/- 0.01 were not included in the computation of mean and confidence limits.

            Overall, the NBP CTD worked well. The differences between the primary and secondary temperature and conductivity values were small throughout the cruise, with no indication of change with time in temperature and only a slight suggestion of drift in conductivity. The resulting primary and secondary salinities agree well, with a mean difference less than -0.001 psu for the entire cruise.

            The comparisons of the primary and secondary salinities with the bottle salinities suggest: 1) a small drift over time by both primary and secondary salinities relative to the bottle salinities, and 2) a small jump between the primary and bottle salinities starting about sample 350. The overall drift between CTD and bottle salinities is small.

            Linear regression between the primary minus bottle and secondary minus bottle values versus sample number gives a mean drift of -0.0014 psu and -0.0027 psu, respectively, over the first 350 samples (these two drifts are statistically different from zero, but not each other at 95% confidence). The shift between primary and bottle salinities around sample 350 is believed due to contamination of the conductivity cell during the cast at station #84. The salinity profile at this station exhibited both positive and negative jumps not noticed at any other station. The primary sensor set was flushed with distilled water after the station and the next CTD cast at station 85 appeared normal. However, Figure 2 suggests that there was a small shift in the primary conductivity cell, resulting in a mean difference of -0.0032 psu from the bottle values. The apparent lack of a similar jump in the secondary cell suggests that the jump observed with the primary cell was not due to any subtle change in the AutoSal accuracy and stability.  Even with these small drifts and jump, the CTD produced data of very high quality during the entire cruise.

             

Figure 2.  Comparisons between bottle salinity data and the difference in the primary (0) and secondary (1) temperature and conductivity sensors on the CTD (top two panels), the difference in salinity calculated from CTD conductivity (middle panel), and the difference between CTD-derived salinity

and bottle salinity for each sensor (bottom two panels).

            On most CTD casts, water samples were taken for determination of dissolved oxygen concentration (Appendix 3).  A total of 366 oxygen samples were taken during the cruise.  The oxygen samples were analyzed on board the ship, usually within 48 hours of collection, using an automated amperometric oxygen titrator developed at Lamont-Doherty Earth Observatory.  Comparison of the titrated oxygen values with the corresponding values from the oxygen sensor on the CTD (Figure 3) showed excellent agreement.   Also, the comparison of the titrated and CTD-measured oxygen concentrations did not show any drift or trend over time.  Thus, no corrections to the dissolved oxygen concentrations obtained from the CTD are indicated.  However, a more detailed look at the comparison data will be made to determine if any corrections are needed. 

 

Figure 3.  Comparison of dissolved oxygen concentration obtained from titration and

the corresponding value obtained with the oxygen sensor on the CTD.

 

 

            Preliminary processing of the CTD data was done during the cruise using the procedures and algorithms given in UNESCO (1983).  The temperature and salinity values were plotted and compared with historical data sets to check the accuracy of the data.  Additional checking of data quality consisted of comparing the temperature and salinity values obtained from the dual sensors on the CTD.  These comparisons showed deviations that were at the level of instrument precision. However, additional checking and post-cruise calibration of the sensors on the CTD by SeaBird remains to be done. It is anticipated that the final hydrographic data set will not differ substantially from what is described in this report. 

            Water samples for nutrient determination were taken from each Niskin bottle on each cast.  The methods and techniques used for this are described in section 5.0 of the cruise report.  Similarly, water for chlorophyll determination was taken on each cast and the methods and techniques used for this are described in section 6.0 of the cruise report.  The discrete chlorophyll samples provide calibration for the fluorometer on the CTD. 

                       

1.2.3 Expendable Probes

            The intent was to make CTD measurements at each survey station.  However, at 17 of the stations, the weather conditions were such that it was not possible to deploy the CTD.  At these stations, either an expendable CTD (XCTD) or expendable bathythermograph (XBT) probe was deployed.  The XCTD probes provide data to 1000 m.  The XBT data were collected using either T-4 (nominal depth of 460 m), T-7 (nominal depth of 760 m), or T-5 (nominal depth of 1830 m) probes.  The XBT probes were also used to increase the resolution of temperature measurements in specific sections of the survey grid.  The XCTD and XBT probe drops are summarized in Appendices 4 and 5, respectively. 

            The XCTD and XBT probes were deployed using a hand-held launcher, either from the main deck or the 01 deck of the ship, depending on weather conditions.  The XCTD and XBT probes were manufactured by Sippican and had a high failure rate.  It was frequently necessary to use several probes to get a single profile.  This was especially true for the XCTD probes, which had about a 25% failure rate.  Comparisons between the vertical profiles obtained with the XCTD and XBT probes and those obtained with the CTD show no appreciable differences.  Thus, no calibrations are necessary in order to merge the two data sets. 

 

1.2.4 ADCP Measurements

            The RDI 150 kHz Acoustic Doppler Current Profiler (ADCP) system mounted in the hull of the RVIB N. B. Palmer was set to begin collecting data at 0300 GMT on 25 April 2001 at the start of the SO GLOBEC cruise.  The system continued to collect data until 5 June 2001, when it was turned off at the end of the cruise.  Thus, the ADCP system ran continuously throughout the cruise without any instrument or software problems.  The ADCP system was configured to acquire velocity measurements using fifty eight-meter depth bins and five-minute ensemble averages.  This configuration provided velocity measurements from the first bin, at 31 m, to 300 m and sometimes 400 m.  Depth bins two through ten were used as the reference layer.

            The ADCP was run in bottom tracking mode during times when the survey was taking place in water with depths less than 500 m.  Because much of the area included in the survey grid is less than 500 m, the majority of the ADCP data were collected in this mode.  Bottom tracking was disabled during times when the survey extended beyond the continental shelf edge and into deeper water for several hours. 

            Preliminary processing of the ADCP data was done during the cruise using an automated version of the  Common Oceanographic Data Access System (CODAS) developed by E.  Firing and  J.  Hummon from the University of Hawaii.  Maps of the ADCP-derived current vectors along the ship track were generated at daily intervals.  Overall, the automatic data processing provided good quality data.  However, some editing of the ADCP data was needed to remove large current vectors that occurred during times of intense storms when the ADCP had trouble maintaining bottom track mode due to ship motion.  These periods account for the gaps that appear in the current velocity distributions.  It is anticipated that the final ADCP data set will not differ substantially from what is provided in this report. 

 

1.3 Preliminary Results

 

1.3.1 Water Mass Distributions

            The potential temperature-salinity (2-S) diagram constructed using all of the CTD and XCTD data (Figure 4) allows the water masses in the study region to be identified.  Temperatures of -1.5°C to 1.0°C and salinities of 33.0 to 33.7 at F₂ values of less than 27.4 represent AASW.  The large scatter in these data indicates the temporal changes in the AASW that occurred as this water mass underwent seasonal heating and cooling.  The temperature minimum at about -1.5°C at salinities of 33.8 to 34.2 is associated with Winter Water.  The signature of Winter Water is eroded by mixing and seasonal heating and this is reflected in the 2-S diagram by the deviation of this water from -1.8°C and 34.0.

 

Figure 4.  Potential temperature-salinity diagram constructed using the CTD data collected

during NBP01-03. The contours represent lines of constant s_q and the small box indicates the region characteristic of Circumpolar Deep Water.  The dashed line indicates the freezing point of water

as a function of salinity.

 

 

 

            The cluster of points on the 2-S diagram at temperatures of 1.0°C to 2.0°C and salinities of 34.6 and 34.7 represents Circumpolar Deep Water.  This water is composed of two varieties: Upper and Lower Circumpolar Deep Water.  The Upper CDW is characterized by a temperature maximum at a density of 27.72.  Lower Circumpolar Deep Water is characterized by a salinity maximum of 34.72 at a potential density of 27.8. 

            The majority of the points on the 2-S diagram are associated with a modified form of Upper CDW.  This water is the result of mixing of the Upper CDW with AASW on the shelf.  The modified CDW water is characterized by temperatures of -1.5°C to 1.5°C and salinities of 34.3 to 34.6.

            An additional water type seen on the 2-S diagram is characterized by temperatures of -1.6°C to -1.7°C and salinities of 33.4 to 33.0.  This water was observed in the CTD casts from inshore waters near the ice shelves on Adelaide Island.  The cold, fresh water is found at the surface and is the result of melting from the ice shelves. 

 

1.3.2 Distribution of Temperature Maximum Below 200 m

            An approach for determining the circulation in the study region is to map the distribution of the temperature maximum below 200 m (Figure 5), which tracks the movement of UCDW and modified CDW on the WAP shelf.  This approach assumes that the isotherm patterns can be used to approximate the current flow.

            The southern boundary of the Antarctic Circumpolar Current (ACC) is denoted by the 1.6°C isotherm and the southern ACC Front is denoted by the 1.8°C isotherm.  Both isotherms were present along the outer edge of the continental shelf over the entire survey grid.  This indicates that the ACC was situated along the shelf edge for the duration of the study.  There is no evidence of a shelf-slope front, which is as expected for this region.  Temperatures greater than 1.6°C were found extending onto the WAP shelf in two places.  The first is at the northern end of the survey grid and is associated with a meander in the southern ACC boundary.  This meander resulted in a bottom intrusion of warm CDW onto the shelf.  The onshore movement of the warm water at depth is aligned with a deep depression that connects the outer shelf and the inner portion of Marguerite Bay.

            The second occurrence of 1.6°C water on the shelf is near survey transect seven.  Again, a meander in the southern boundary of the ACC occurs which produces a bottom intrusion of upper CDW on the shelf.  This second meander is associated with a topographic feature that extends seaward from the continental shelf edge. 

            A third bottom intrusion of upper CDW is suggested by the occurrence of 1.3°C water inside of Marguerite Bay.  The isotherm pattern suggests that this event is separated from the intrusion that is occurring at the northern end of the survey grid.  The reduced temperature associated with this feature suggests that it is an older intrusion in which the upper CDW has been modified by mixing with the shelf water and AASW. 

            The isotherm pattern on the inner shelf in the northern part of the survey grid suggests a southerly flow along the coast that turns around Adelaide Island and enters Marguerite Bay.  This isotherm pattern continues around the southern portion of the inner Bay, extending into the far southern corner of the Bay.  The isotherm pattern in the southern portion of the Bay along the coast of Alexander Island suggests flow out of the southern end of the Bay onto the continental shelf.  The isotherms extend outward across the shelf where they spread and mix with the shelf waters.

           

1.3.3 ADCP-derived Current Distributions

            The ADCP-derived current distribution between 31 and 75 m over the survey grid (Figure 6) shows patterns that are consistent with those suggested by the distribution of the maximum temperature below 200 m.  The largest currents are at the shelf edge and are associated with the ACC.  These currents are to the north-northeast and deviations from this direction are in the areas where the southern boundary of the ACC meanders onto the WAP continental shelf.

 

 

 

 

Figure 5.  Distribution of the temperature maximum below 200 m constructed from CTD

temperature observations taken during NBP01-03.  Station locations are indicated by dots.

 

 

           

            The currents along survey transects one and two show reversals that coincide with the flow in the meander that produces the bottom intrusion of UCDW.  Currents associated with the second meander on survey transect seven are larger, reflecting the stronger flows associated with the southern ACC boundary.

            Currents over the shelf tend to be small, on the order of 8 to 10 cm s^-1.  Shelf currents are predominately to the south-southwest.  The coastal current on the inner shelf is well resolved in the ADCP-derived current distribution.  This current flows south-southwest along the outer part of Adelaide Island and turns into Marguerite Bay around the southern tip of the Island.  Velocities associated with the current are on the order of 10 to 25 cm s^-1.

 

Figure 6.  ADCP-derived surface currents plotted along the cruise track.  The velocity vectors

represent averages in both space (31-75 m vertical depth averages) and time (2 hr averages).

The 3000 m, 2000 m, 1000 m, and 500 m isobaths are shown as solid black contours.  The

heavy black line represents the edge of the ice shelves.

                       

 

 

            The flow out of Marguerite Bay around Alexander Island is seen as a coherent current along the inner shelf region until the end of the area included in the survey grid.  Velocities associated with this current are similar to those observed for the current flowing into Marguerite Bay. 

            The currents within Marguerite Bay appear to form two clockwise gyres.  The presence of the two gyres may be the result of bathymetry control on the flow in the Bay.  If Marguerite Bay does have two circulation cells, then this has implications for exchanges within the Bay and with the adjacent continental shelf.  Finer scale mapping of the current distribution in Marguerite Bay should be incorporated into the survey design of future Southern Ocean GLOBEC cruises.

                                                                       

1.3.4 ADCP-derived Shear and Richardson Number Profiles

            Large shears in velocity profiles obtained from the ADCP data were present at a number of CTD stations occupied during NBP01-03. To investigate if these shears were strong enough to cause active mixing in the water column, gradient Richardson numbers were estimated using the ADCP-derived shear and CTD density data. Gradient Richardson numbers smaller than 0.25 indicate active mixing, and values less than 1.0 suggest that mixing is probable.

            The gradient Richardson number is defined as Ri = N²/Shear², where N is the buoyancy frequency and the Shear is the total vertical current shear. To calculate Ri, the x and y components of shear were first calculated between adjacent 8-m bins using the mean U and V profiles computed from the individual ADCP profiles collected during the CTD cast. The total shear was then:

 

and N was calculated from the density profile derived from the CTD temperature and salinity data collected at that station. The density data were then averaged into 8-m bins corresponding to the depths of the ADCP bins to give N at the center of the shear estimate.

            Plots of U, V, shear, density, buoyancy frequency, and gradient Ri numbers for CTD stations 3 and 15 are shown in Figures 7 and 8, respectively. These two figures illustrate different regimes found at stations along the survey grid. Station #3, having higher Ri numbers, appears to be much more stable than does station #15. To determine the cause(s) of the differences, we looked at both the stratification and shear. Although station #3 has a stronger peak in buoyancy frequency than station #15, the value of N is very similar below 100 m. The shear at station #15 is much greater below 100 m than the shear at station #3 and results in the lower values of Ri found at station #15. Station #3 was located at mid-shelf, while station #15 was located in the coastal current flowing along the coast of Adelaide Island. Whether there is a systematic pattern of lower Ri numbers in the fresher coastal currents as opposed to the more saline shelf water remains to be determined once similar analyses are done for all stations made during NBP01-03.

            One point of showing the shear and Ri profiles at these two stations is to illustrate the presence of both weak and strong shears in the SO GLOBEC study region. The cause of the strong shear found at station #15 is not clear at this time. Possible candidates include inertial motion and internal tidal motion. The inertial period is roughly 12.9 hours in this area, close to the M2 period of 12.42 hours, so differentiating inertial from tidal motion is difficult with short current records. A drifter deployed on this cruise exhibited inertial/tidal motion for about three days, with major and minor axes currents of 17 and 14 cm s^-1, respectively. Identifying the sources of strong current shear is an important program goal, since they may cause significant mixing at depth in this area and play an important role in water mass formation and evolution.

 

1.4 Acknowledgments

 

Much of the credit for the high quality hydrographic data set collected during NBP01-03 is due to the efforts of Raytheon marine technicians, Matthew Burke and David Green, and electronics technicians, Jan Szelag and Jeff Ottern.  Their willing and cheerful response to all requests made the collection of the hydrographic data set a pleasure.  Their efforts are most appreciated. 

 

 

Figure 7.  Top panels (from left to right): Plots of U, V, and shear at station #3.  The heavy black

line represents the average of U and V on station.  Lower panels (from left to right): plots of the

density, buoyancy frequency, and Richardson number at station #3.  The dotted line marks the

location of the Ri=0.25.  The dashed line marks the 1.0 value.

 

 

 

2.0 Drifter Measurements  (Bob Beardsley and Dick Limeburner)

 

2.1 Introduction

 

            Surface drifters are being deployed and tracked via satellite to study the near‑surface Lagrangian currents in the SO GLOBEC study area on the western Antarctic Peninsula Shelf.  Each drifter has a small (~ 30 cm diameter) surface float with ARGOS transmitter and batteries tethered to a holey sock drogue centered at 15 m below the surface.  The drogue, about 10 m tall and 1 m in diameter, is designed to "lock" itself to the water so that the surface float follows the mean water motion at 15 m depth with very little slippage even in high winds.  Thus measuring the drifter's position as a function of time provides a Lagrangian measurement of the 15-m ocean current.  The drifter's signal is picked up by satellite and its position determined roughly 10 times each day with an accuracy of better than 1 km.  The raw time/position/water temperature/drogue data are sent daily to Dick Limeburner at WHOI, who edits and filters the data to remove tidal and higher frequency motions and then sends the resulting 6‑hourly time and position data for all drifters to date to the ship as a MATLAB file attachment in an email.

 

Figure 8. Top panels (from left to right): Plots of U, V, and shear at station #15.  The heavy black

line represents the average of U and V on station.  Lower panels (from left to right): plots of the

density, buoyancy frequency, and Richardson number at station #15.  The dotted line marks the

location of the Ri=0.25.  The dashed line marks the 1.0 value.

 

 

 

2.2 Drifter Deployments on NBP01‑03

 

            Eight surface drifters were deployed during NBP01‑03.  These drifters augmented the six drifters launched during the first SO GLOBEC mooring cruise LMG01‑03.  The drifters were obtained from ClearWater and NOAA. Table 1 gives the launch time and position for each drifter and its status as of 1 June 2001, the last day that drifter data were sent to the ship. 

 

 

 

 

 

Table 1.  List of launch times and positions and status as of 1 June 2001 for the 14 surface

drifters deployed to date as part of the SO GLOBEC program.  The month, day, and time of the

last good data for the three drifters that ceased operation early is also given.

 

ID	Time (GMT)	Latitude	Longitude	Status
a1	3 26 23.01	‑68 20.71	‑71 26.74	Off (5 6 0:00)
a2	3 30  6.84	‑68 54.42	‑74 54.24	On
a3	3 30  7.78	‑69 09.82	‑75 21.40	On
a4	3 30 14.48	‑68 12.75	‑74 55.98	On
a5	3 31  1.63	‑66 30.99	‑71 08.87	On
a6	3 31 20.48	‑66 45.43	‑70 57.86	Off (4 5 18:00)
a7	5  3  6.65	‑68 23.71	‑67 18.94	On
a8	5  5 19.40	‑68 55.15	‑67 45.48	On
a9	5  7  6.82	‑69 00.27	‑68 50.45	On
a10	5  8  8.42	‑68 50.25	‑71 26.11	Off (5 22 0:00)
a11	5 16  1.00	‑69 01.29	‑75 42.43	On
a12	5 26 18.25	‑68 22.86	‑70 37.92	On

a13	5 27  0.23	‑68 04.92	‑69 13.38	On
a14	6  1  6.87	-66 36.67	‑69  6.45	On

 

 

2.3 Preliminary results

 

2.3.1 Low‑frequency flow

            The low‑pass filtered trajectories of 13 of the 14 drifters are plotted in Figure 9.  Drifter a6 ceased transmitting shortly after launch and is not included in this summary.   The tracks are plotted for four 15‑day time windows, starting on March 31 (YD 90).  The last window ends on June 1 (YD 152), when drifter a14 was launched.  An asterisk is plotted at the last position for each drifter, showing the direction of the drifter motion. The drifter identification number is plotted to the right of the asterisk.  A brief analysis of these tracks is given next.

 

Panel A: YD 90‑105

            Drifter a1, deployed west of Adelaide Island (AdI), moved southwest along the island for two days (mean speed ~ 23 cm s^-1) during strong southwest winds, then slowed and stopped transmitting and reappeared deep in Marguerite Bay (MB). Whether a1 lost its drogue and was blown into MB is not known, although its subsequent behavior suggests it still had its drogue. Drifter a2 moved into MB, turned clockwise and moved rapidly (2‑ day mean speed ~ 38 cm s^-1) northwest near Alexander Island for about 3 days before slowing.  Drifters a3, a4, and a5 showed little net movement during this period, with typical speeds of less than 10 cm s^-1.

 

 

Figure 9.  Satellite-tracked drogue movements on the Western Antarctic Peninsula

continental shelf region, including Marguerite Bay (April/May 2001).

Panel B: YD 105‑120

            Drifter a1, apparently still with drogue, began to leave MB, moving along the northeast side of Alexander Island, following the path taken by drifter a2. Drifter a3 moved in a clockwise loop towards the mouth of MB.  The other drifters showed little net movement, especially a5.

 

Panel C: YD 120‑135

            Drifters a7, a8, and a10 (deployed on NBP01‑03) moved into MB, while a9 moved deeper into MB.  Winds during much of this period were towards the south and quite strong starting around YD 130 and 133, when the wind speed jumped from near 0 to over 40 kts in 6 hours.  These four drifters accelerated towards YD 133 and had speeds of ~ 40(a8), 30(a7), 22(a9), and 22(a10) cm s^-1 during the first half of YD 133.

            During this period, a1 continued northwestward out of MB to mid‑shelf, near where a1 and a4 were located.  Drifter a3 continued to move towards the southwest across the mouth of MB to the northwest coast of Alexander Island. This drifter passed very close to CTD station 53, located close to the coast of Alexander Island.  This station was made deep within brash and pancake ice, with many grounded icebergs around.  While drifter a3 did not get trapped in the ice and continued southwest along the coast of Alexander Island, drifter a10 later did move to this area and was stopped in the sea ice.

 

Panel D: YD 135 - 153

            Drifters a7, a8, and a9 continued to move deeper into MB.  Drifter a7 appears to have moved to the coast, while the other two drifters turned towards the south and continued along the coast.  Drifter a13 moved around the southern tip of Adelaide Island into MB, while drifter a12 moved southward towards Alexander Island.

            Drifter a10, which had entered MB earlier, moved quickly out of MB along the northeast side of Alexander Island with a mean speed of ~ 25 cm s^-1, before turning around the northern tip of Alexander Island and moving along the coast, following closely the path of drifter a3.  Near CTD station #53, drifter a10 apparently became trapped in the ice and stopped moving.   Drifter a3 left station #53 and moved very rapidly (with maximum speeds 40‑50 cm/s) towards the southwest along the coast of Alexander Island, arriving off Lazarev Bay on YD 140. Then drifter a3 turned offshore and moved towards the northwest to mid‑shelf, where it joined drifters a11 and a2.  These three drifters continued to move towards the west along the shelf, with speeds varying from ~ 10 cm s^-1 for a11 to 20 - 40 cm s^-1 for a2.  Drifter a4 left this group and moved towards the shelfbreak at speeds around ~ 10 cm s^-1.

            The lack of motion of drifter a5 is notable. Over the 61.4‑day period shown in Figure 9, drifter a5 moved a net distance of 62.9 km towards the northwest, with a mean vector speed of 1.2 cm s^-1 and a mean scalar speed of 2.9 cm s^-1.  While this drifter did reach a peak speed of 18 cm s^-1 briefly, its overall movement (or lack of it) is quite different in character than the other drifters.  Is there a simple physical explanation for this?  The broad‑scale hydrographic survey suggests that drifter a5 was deployed in the center of a deep bubble of cooler shelf water surrounded on three sides by warmer slope water.  If this interpretation is correct, it would mean that the feature evolves slowly in time, keeping the drifter inside the feature for over two months. Further analysis is needed to check this interpretation and investigate its consequences for the shelf circulation.

 

2.3.2 High‑frequency flow

            The above summary is based on the low‑pass filtered drifter motion; however, the high frequency of ARGOS fixes per day (approximately 20 per day) allow some investigation of the higher frequency drifter motions. As an example, drifter a8 deployed near station #26 on 5 May moved in counterclockwise loops for the next 3 days while slowing moving towards the northeast.  This looping motion appears to be inertial.  The inertial period at the drifter latitude is 12.99 hours, close to the M2 period of 12.42 hours, so differentiating inertial from tidal motion is difficult with short current records. To quantify this motion, a simple model consisting of a mean current plus inertial component was fit in a least‑squares sense to the drifter position data.  During this 2.9‑day period, drifter a8 moved in a counterclockwise elliptical path towards the northeast with a mean speed of 3.5 cm s^-1.  The elliptical motion had a major axis of 16.8 cm s^-1 and a minor axis of 10.5 cm s^-1, with the major axis oriented toward 28EN.

            While the looping motion of a8 could be a combination of inertial and tidal or tidal, this example illustrates the usefulness of the original unfiltered drifter position data to look for high frequency motions. A detailed analysis of both raw and filtered drifter data in combination with surface wind data collected by the Automated Weather Stations (AWSs) on Kirkwood and Faure Islands and by the ship will be made after the cruise, as additional drifter and meteorological data are obtained.

 

2.4 Summary

 

            The SO GLOBEC drifter data collected to date suggests that there is a strong surface current into Marguerite Bay around the southern end of Adelaide Island, with a return flow out of the Bay along the northeastern tip of Alexander Island.  The inflow appears to be both broad and surface intensified, with drifters crossing significant topographic variability as they pass around the southern tip of Adelaide Island. It is not clear how persistent this inflow is; however, all four drifters deployed upstream (a1, a7, a8, and a13) entered the Bay. Drifter 14 was deployed further upstream to help determine the origin of this inflow.  Hydrographic data collected west of Adelaide Island does suggest a surface layer of fresher water along the west and south coast of Adelaide Island; it is not clear if this alone could support the observed inflow.

            All three drifters (a1, a2, and a10) that exited the Bay did so along the northeastern tip of Alexander Island, suggesting the presence of a strong exit current centered there. Surface salinity data support the idea of a relatively fresh coastal current flowing out of the Bay initially trapped to the topography along Alexander Island. While two of the three drifters (a1 and a2) moved northwest towards mid‑shelf after leaving the tip of Alexander Island, drifter a10 followed the coast around the northern tip of Alexander Island and moved along the northwest coast.  Again, surface salinity data plus the path of drifter a3 suggests the presence of a coastal current towards the southwest along this side of Alexander Island. Whether this flow is a continuation or branch of the exit flow along the northeastern side of Alexander Island is not known, a purely geostrophic coastal current would follow this path.

            While the very slow movement of drifter a5 is unusual in a shelf region with such strong surface forcing, it seems likely that this drifter was deployed in a slowly evolving slope water intrusion that are known to occur on the west Antarctic Peninsula shelf.  The other drifters exhibit a range of speeds, especially within Marguerite Bay, that suggest an energetic surface current field in the SO GLOBEC study region.

           

3.0 Meteorological Measurements  (Bob Beardsley and Jeff Otten)

 

3.1 Introduction

 

            Underway meteorological data were collected during NBP01-03 to help document the surface weather conditions encountered during the cruise and to characterize the surface forcing fields in the SO GLOBEC study area during austral fall.  The N.B.Palmer (NBP) arrived near the start of the large‑scale physical/biological survey on April 27 (YD 117) and left the area to return to Punta Arenas on June 2 (YD 153). A full suite of meteorological data were collected during this 37‑day period.  This report provides a preliminary description of the meteorological data collected on NBP01-03 and some initial results concerning the surface forcing during fall.

 

 

3.2  Instrumentation

 

            The NBP was equipped with the following set of meteorological and surface oceanographic instrumentation to collect continuous underway data during NBP01‑03 (Table 2).  A pair of Belfort propeller/vane anemometers and sensors to measure incident short‑ and long‑wave radiation (SW, LW) and PAR were mounted on the top of the NBP's main "science" mast (Figure 10). The air temperature (AT) and relative humidity (RH) sensors and precision barometer (BP) were mounted near the base of the main mast on the 04 deck, aft of the bridge.  The heights of the anemometers and the air temperature and relative humidity sensors above sea level were estimated to be 33.5 and 17 m, respectively.  Sea surface temperature (SST) was normally measured using a remote sensor and intake in the stern thruster housing when the thrusters were not in use or on standby. Sea surface salinity (SSS) and raw fluorescence (FL) were measured using a thermosalinograph and fluorometer placed in the aft chemistry laboratory. Water for both instruments came from the intake in the stern thruster housing when it was not in use.  A second intake, from the ship's sea chest, was used when the thrusters were on standby or in use.

 

 

 

Figure 10.  The NBP “science” mast, with port and starboard anemometers and shortwave, longwave, and PAR sensors mounted to the railing, during pre-cruise service in Punta Arenas.

 

Table 2.  NBP01‑03 meteorological sensors, their calibration history, time of installation,

and conversion factors used to convert raw voltage output to scientific units.

 

Sensor	Model	Serial Number	Last Calibration	Installed	Conversion
Starboard Anemometer thru 5/20/01	Belfort Model 5-122AHD	7957	04/18	04/18
Starboard Anemometer from 5/20/01	Belfort Model 5-122AHD	7956	04/18	05/20
Port Anemometer	Belfort Model 5-122AHD	92-2133	04/12	04/18
Anemometer north relative speed vector voltage					m s-1 = 7.553 x voltage
Anemometer east relative speed vector voltage					m s-1 = 7.553 x voltage
Air temperature	R. M. Young 41342C	2267	01/23	04/18	°C = 10 x voltage - 50
PIR Eppley Pyrgeometer	Eppley PIR	33023F3	06/23	01/28	W m-2 = 923.87 x voltage
PSP Eppley Pyranometer	Eppley PSP	33090F3	11/07	01/28	W m-2 = 194.53 x voltage
Relative Humidity Sensor	Rotonics MP-101A-C4	R45618	06/20	10/24
Temperature at the Relative Humidity Sensor					°C = 10 x voltage – 40
Relative Humidity					%RH = 10 x voltage
PAR Irradiance	BSI QSR-240	6356	02/15	04/18	mEi/m2s = 1662.24 x voltage

Barometer

AIR-DB-3A

7G3095

07/21

01/26

(none)

 

 

3.3 Data Acquisition and Processing

 

            The raw NBP shipboard meteorological data were collected using the ship's data acquisition system (RVDAS).  A 1‑minute subsample 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:\NBP0103\geopdata\JGOF\ g099.dat and jg100.dat, respectively). This 1‑minute time series was produced using a JGOFS program 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.

            The daily data were obtained from drive Q and converted into standard variables using the MATLAB m‑file read_palmer_met1m(yd).  This program also removed pad values (produced when the DAS recorded no data), edited several variables, and stored the new data set in a MATLAB mat‑file for each day (e.g., jg100.mat for YD=100). The SW signal and night‑time bias were comparable during most of the cruise, so the SW record was hand‑edited to remove the bias and make a positive‑only SW series.  Both the SST and SSS data included large spikes associated with the change in intake when the ship's stern thruster was placed on standby or being used.  For much of the cruise, these spikes were removed and the gaps filled by linear interpolation.

            The m‑file merge_palmer_met1m(first_yd,last_yd) was 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 palmer_met1m.mat.

            For further analysis, the 1‑minute data in palmer_met1m were low‑pass filtered and subsampled using make_palmer_met5m into 5‑minute time series. The filter used is the pl66tn set with a half‑amplitude period of 12 minutes.  The 5‑minute data were then used to estimate the surface wind stress and heat flux components using bulk methods called by compute_palmer_wshf5m.  The surface wind stress and heat flux data were then added to the palmer_met5m, so that this 5‑minute time series contains best versions of the surface meteorological conditions and forcing for the cruise.  Both the 1‑minute palmer_met1m and 5‑minute palmer_met5m data are included on the cruise data CD-ROM.

 

3.4 Problems and Solutions

 

            Several problems with the meteorological and underway instrumentation or data logging became clear during the cruise.  These problems and suggested solutions are summarized next.

 

3.4.1  RVDAS recording format

            Some of the NBP met data were recorded using limited precision.  Whether this is a hardware limitation of the digitizers used or simply setting the record format with too few significant places is not clear.  Table 3 gives the record most significant place for each variable:

 

Table 3. Record increment for different variables.

 

Parameter	Resolution
Air Temperature	0.1 oC
Sea Surface Temperature	0.01oC
Sea Surface Salinity	0.01 PSU
Short‑wave Radiation	2.0 W m-2
Long‑wave Radiation	8.0 W m-2
PAR	16.666 mE (m2s) -1
Fluorescence	0.01 V

 

            The present meteorological measurement system on the NBP will be replaced in July 2001, so that some variables will be recorded with greater precision. In particular, it would be good to record SW and LW with 0.1 W m^-2 resolution and PAR with 1 FE m^-2. Comparison of the SST and SSS with near‑surface CTD data collected at stations with a deep surface mixed layer suggested that SST and SSS should be recorded with 0.001^oC and 0.001 PSU resolution.  The air temperature sensor is normally calibrated to +/‑ 0.1 ^oC, so that it should be recorded with 0.01^oC resolution.

 

 

3.4.2.  PIR battery failure

            The Eppley longwave pyrgeometer (PIR) uses an internal mercury battery to supply a precise and stable reference voltage in the measurement circuit.  A brief inter‑ship comparison of the L.M. Gould (LMG) and NBP PIR values while at Palmer Station on 28 April 2001 suggested that the NBP PIR was reading low (Table 4).  A second inter‑ship comparison made in northern Marguerite Bay on 6 May 2001 also showed the NBP longwave values were significantly lower than the LMG values.  Over the next several days, the NBP longwave values started to drop rapidly.  On 12 May 2001, Jeff Otten and Jim Dolan replaced the PIR mercury battery and the PIR readings jumped back up to values more consistent with those measured on the R/V L.M. Gould.

 

Table 4.  Comparison of PIR incident longwave radiation recorded on the LMG and NBP

during two periods when the ships were collocated, first in the morning of 28 April 2001

for about 3.23 hrs at Palmer Station, and second, during the evening of 6 May 2001 for

9.85 hrs in northern Marguerite Bay.

 

 

            The PIR mercury battery has a very flat discharge curve, so that failure should occur rapidly.  The old battery measured about one‑half its rated voltage, indicating it was past its useful life. A visual inspection of the NBP longwave data suggests that data taken up to 7 May 2001 may be useable.  A detailed inter‑ship comparison of the LMG and NBP PIR data will be made after the cruise, in part to help determine how much of the early NBP longwave data are good (pre‑battery failure).

            The PIR battery can only be checked when the housing is opened. This is not normally done at sea due to the location of the PIR on the top of the mast.  One solution is to check the battery voltage prior to each cruise and develop a battery life history and replacement schedule. A second solution is to install new batteries prior to next year's SO GLOBEC cruises.  This will be requested.

            One unexpected benefit of making the inter‑ship comparisons of PIR data during NBP01‑03 and LMG01‑04 was the discovery that the LMG PIR values were being computed using the wrong calibration coefficients (actually the coefficients for the PIR used on LMG01‑03).  In hindsight, this was the reason for the initial offset in the inter‑ship longwave comparison. The subsequent differences were due to the battery failure. The LMG longwave radiation series was recomputed using the correct coefficients and these new data will be used in the post‑cruise analysis of the inter‑ ship comparison data.

 

3.4.3 Icing and anemometer failures

            The meteorological sensors on the mast collected ice during parts of this cruise.  For example, the starboard anemometer started to report near 0 wind speeds on 16 May 2001 while the port anemometer continued to report acceptable wind speeds and directions. Visual inspection of the anemometers from the bridge showed no difference, so the exact nature of this problem (e.g., icing, connections) was never determined. Whenever possible during or shortly after icing conditions, Jeff Otten (Electronics Technician) climbed the mast to "de‑ice" the sensors and check connections and wiring.

            Despite icing problems, the two anemometers appeared to give similar data for much of the cruise.  One failure mode of the port anemometer was for one of the two output voltage signals to be zero (presumably due to a connection problem), thus making both the wind speed and direction wrong.  Post‑cruise analysis of the raw anemometer data should allow a detailed comparison of the data from both units, which in turn should allow periods of poor performance to be identified and eliminated from the computation of "true" wind.

 

3.4.4  "True" wind computation

            There were several periods during the cruise when the JGOFS "true" winds were weak (under 5 m s^-1) and exhibited jumps in wind speed caused by the motion of the ship.  This was especially obvious as the ship towed BIOMAPER-II at 4 kts between CTD stations.  The JGOFS format includes only "true" wind and direction (thought to be computed using the port anemometer data only), so it is not possible to re-compute true wind using just the JGOFS data.

            The raw anemometer and other underway data are archived on the cruise data CD‑ROM.  After the cruise, we plan to use the raw data to produce a final edited meteorological data set for this cruise.  This will include assessing the wind data from both anemometers and using both to compute "true" wind speed and direction.  This final data set will be posted on the SO GLOBEC website.

 

3.4.5 Thermosalinograph contamination

            The NBP SeaBird thermosalinograph (TSG) produced high quality data for most of the cruise; however, the data taken just before and on station were corrupted and should be used only with great care.

            When approaching a station, the thruster generator was first started and a servo system activated so that the thrusters could be used on demand. The TSG intake is located in the stern thruster housing.  During NBP01-03, the system to automatically switch the TSG intake to another location was broken, so that the switch was done manually, usually just after the generator was running smoothly.  This change in intake caused a pulse of warm water to pass thought the TSG, creating a large spike in temperature and salinity lasting 5‑10 minutes.  A similar spike occurred when the ship left station and the thruster generator was turned off.  While these two spikes had a characteristic shape, using the thrusters on station also caused jumps in the TSG temperature and salinity data. These jumps were irregular in shape and duration, making identifying them difficult.  The SST and SSS data included in the palmer_met1m and palmer_met5m data sets have been edited to remove the most obvious of these fluctuations; however, we suggest not using the SST and SSS data when the ship was nearing or was on station.

            During the cruise, two comparisons of TSG data were made with other data taken when the ship was steaming between stations.  The first comparison was between TSG and bottle salinities where the bottle samples were taken from the TSG exit flow and the TSG trace indicated steady conditions.  The second comparison was between the SST and TSG salinity and the T and S measurements made with BIOMAPER-II nearest the surface as it was toyoed along transect 6. The mean depth of BIOMAPER-II during the comparison values was 33 m.  

            The results from both comparisons are shown in Figure 11.  While preliminary, these results suggest that the TSG salinity was accurate to within +/‑ 0.002 psu on average with no clear bias.  The NBP SST and BIOMAPER-II T data have a clear offset of about 0.06EC, with the ship's sensor reading higher than BIOMAPER-II. Comparisons between the CTD surface temperature and SST show close agreement, to within 0.01EC, suggesting that the BIOMAPER-II temperature sensor may read low by order 0.06EC.  Additional comparisons need to be made to test these preliminary results.

            In view of the inherent high accuracy of the ship's SST and TSG data when the ship is underway, thought should be given to relocate the intake so that the problems associated with the thruster generator and intake switching could be eliminated.  If this is not possible, then perhaps the actual switch in intake could be postponed until the ship is actually on station. This would allow the TSG to collect good data for the 10‑15 minutes prior to each station.  At a minimum, the bridge could keep a log of times when the thruster generator was turned on and off.

 

 

 

 

            Figure 11.  Comparison of the NBP TSG salinity and sea surface temperature data with

bottle salinity and BIOMAPER-II temperature and salinity data collected along transect 6.

           

 

3.5. Description of Cruise Weather and Surface Forcing

 

            Time series of the 5‑minute surface meteorological data and surface forcing collected during NBP01‑03 are shown in Figures 12 and 13.  Figure 12 shows the wind speed and direction, air and sea surface temperatures, relative humidity, barometric pressure, and the incident short and longwave radiation. Figure 13 shows a vector plot of the surface wind stress plus stress amplitude and direction, the net surface heat flux (Q_net), and its four components, the shortwave (Q_sw), longwave (Q_lw), sensible (Q_sen), and latent (Q_lat) fluxes.  Note the period of very low longwave radiation data recorded from YD 127 to YD 132 when the battery was replaced.  These data were judged wrong and not used in the computation of the longwave heat flux component shown in Figure 13.

            Figures 12 and 13 cover the period 27 April to 3 June, when the NBP was working in the SO GLOBEC study area.  The large‑scale survey occurred during 27 April‑21 May (YD 117‑141), followed by visits to Charcot Bay and Lazarev Bay to 24 May (YD144), then work on the AWSs and additional survey work within Marguerite Bay and around the southern and western side of Adelaide Island to 1 June (YD 151), followed by a final cross‑ shelf transect, deep CTD stations, and an XBT survey across the Drake Passage, ending 3 June (YD 154).

Figure 12.  Meteorological data collected during NBP01-03

 

 

 

 

Figure 13.  Surface flux data for NBP01-03.

 

           

            Weather conditions experienced during the cruise ranged from severe gale (with peak winds above 50 kts and heavy seas) to one very clear sunny day with glassy seas. Winds were predominantly from the north, especially the stronger winds associated with the passage of lows from west to east over the WAP shelf. Late in the cruise, a large low pressure system remained centered east of the Peninsula, causing a westward flow of very cold and drier continental air over the WAP and generally clearer skies.  In general, however, the skies were cloudy or overcast, with relatively moist air from the Pacific flowing over the WAP.  Daylight decreased with time and more southern latitude, until the incident shortwave radiation became almost less than the resolution of the shortwave sensor.

            As the cruise progressed and more was learned about the surface weather and forcing conditions on the WAP, it became possible to prepare preliminary notes on several aspects of the surface heat flux occurring during this cruise.  Three of these notes follow here.

 

3.5.1 Surface Cooling ‑ Part 1

            The NBP left Palmer Station about noon (local time) on 28 April 2001 (YD 118) and headed west onto the shelf and south to the SO GLOBEC study area to begin the large‑scale physical/biological survey.  Over the next six days, the NBP sampled the shelf region between Adelaide Island and the upper slope until 6 May 2001 (YD 127), when the survey track line took the NBP into Marguerite Bay.  Here is a brief summary of the mean surface forcing conditions over the shelf during the six days (yd 119‑125) when the NBP made five cross‑shelf transects.

            During this period, the winds were mostly southward and eastward, with speeds varying from near zero to a maximum of 22 m s^-1 (~44 kts).  The mean wind was 7 m s^-1 towards 128T (SE), with an average scalar wind speed of 10 m s^-1 (~20 kts).  The mean air temperature was ‑0.9EC, mean SST was ‑0.6EC, and the mean relative humidity was 92%. The skies were mostly overcast, with periods of fog and snow flurries and little direct sunlight.

            The shipboard met measurements were used to estimate the surface wind stress and heat flux components. The mean wind stress was 0.11 N m^-2 directed towards 132T (SE), closely aligned with the mean vector wind. The net surface heat flux (Q_net) is composed of four components, the net shortwave radiation flux (Q_sw) and net longwave radiation flux (Q_lw) and the two air‑sea flux components, sensible heat flux (Q_sen) and latent heat flux (Q_lat).  The mean and standard deviations of Q_net and the four components for the six‑day period are given in units of N m^-2 in Table 5:

 

Table 5. Q_net statistics.

 

Variable	Mean	STD	MIN	MAX
Qnet	‑108	38	-226	-32
Qsw	4	11	0	76
Qlw	-103	22	-165	-79
Qsen	-0	10	-32	29
Qlat	-9	13	-59	17

 

            Despite the relatively strong wind speeds, the air‑sea temperature difference is only 0.3EC, so that the mean sensible heat flux, Q_sen, is essentially zero. The latent heat flux, Q_lat, is also small due to the low air temperature and high relative humidity.  As austral winter proceeds, the incident shortwave radiation gets smaller, such that for this period the mean, Q_sw, is also essentially zero.  Thus, the net heat flux, Q_net, is due primarily to the net longwave flux, Q_lw.  For this six‑day period, 95% of the net heat loss was due to the longwave loss.

            While these estimates of the heat flux components include significant measurement uncertainty, the basic picture of persistent surface cooling over the shelf driven by net longwave radiation loss seems robust. During the pre‑ice fall, insolation essentially vanishes and the diminishing difference between air and ocean surface temperatures makes the sensible and latent heat loss relatively small.  Despite the high relative humidity, the sky re‑radiates little outgoing longwave radiation back into the ocean, causing the dominant heat flux to be longwave radiation.   To put a steady loss of 100 N m^-2 into perspective, a 50‑m deep surface mixed layer would cool about 0.25EC over a 5‑day period.  Thus, it would only take ~ 20 days for the surface mixed layer to lose ~ 1.0EC.

 

3.5.2 Surface Cooling ‑ Part 2

            During the end of the large‑scale survey on 14-18 May 2001, the NBP made four cross‑shelf transects off Alexander Island after leaving Marguerite Bay. The first transect was made on 14 May 2001 during a gale, with winds towards the south and peak winds for several hours above 40 kts. This storm bought relatively warm air over the shelf, such that during the strongest winds, the sensible and latent heat fluxes were positive into the ocean, and the net surface flux, Q_net, was slightly positive for three hours (above 20 W m^-2 for almost 2 hrs).  This was the only time of surface warming during this five‑day period.  Winds during the remaining transects were mostly south and southwestward and more moderate.  The air temperature dropped during this second period but remained above the ocean surface temperature, so that there were weak sensible and latent heat fluxes into the ocean.  These fluxes were too weak, however, to balance the large longwave heat loss, so that Q_net was negative for the remaining transects. The shortwave contribution to the surface heat flux was very small, due primarily to the more southern location of these transects made later in the month and the persistent overcast. The mean atmospheric conditions and surface forcing during the gale (14 May 2001) and the next four days (15‑18 May 2001) are summarized in Tables 6 and 7.

 

 

Table 6A.  Mean atmospheric conditions during 14 May 2001 (gale).     

 

Vector wind speed	16.0 m s-1
Vector wind dir	‑95.1E (counterclockwise from E)
Scalar wind speed	16.5 m s-1
Air temperature	0.16EC
Sea surface temp	‑1.15EC
Relative humidity	96%

   

Table 6B.  Surface forcing during 14 May 2001 (gale).

 

                      

 

Table 7A.  Mean atmospheric conditions during 15‑18 May 2001.

 

Vector wind speed	10.1 m s-1
Vector wind dir	‑130.5E
Scalar wind speed	11.4 m s-1

Air temperature	-0.78EC
Sea surface temp	‑1.24EC
Relative humidity	98%

 

 

Table 7B.  Surface forcing during 15‑18 May 2001.

 

           

 

            These tables suggest the following conclusions about the surface forcing during the cross‑shelf transects made 14‑18 May 2001:

 

a) Winds were generally strong and almost always southward, generating southward wind stresses that varied from near zero to about 0.8 N m^-2 (8 dynes cm^-2).

 

b) The combination of very strong winds and warm air from the north (during the gale) can produce large enough sensible and latent heat fluxes into the ocean that the net surface heat flux can be positive into the ocean for a short time.

 

c) Excluding the gale, the net surface heat was negative, cooling the ocean. The small sensible and latent fluxes into the ocean (due to the air being warmer than the ocean surface by ~ 0.5EC and mean winds near 20 kts) are more than offset by the roughly constant longwave heat loss (Q_lw ~ ‑80 W m^-2), resulting in a mean Q_net ~ ‑70 W m^-2

 

            These results support the basic picture of persistent surface cooling over the shelf driven by net longwave radiation loss.  The shortwave flux becomes negligible with time and moving further southward, and the sensible and latent fluxes are generally small except during high wind/warm air events (e.g., the 14 May 2001 gale). The persistent overcast reduces the net longwave cooling, but it remains the dominant process in the surface heat flux.

 

3.5.3  Charcot Bay

            After completion of the large‑scale survey, the NBP made a transit around the western end of Charcot Island and into a large bay just south of the Island on 21 May 2001.  This area seemed a good place to look for whales and birds and was thick enough ice for ice sampling and the deployment of a ROV to sample the zooplankton under the ice.  The ship steamed slowly eastward towards the ice shelf, collecting some samples along the way, and eventually launched the Zodiacs to look for whales and birds.  After sunset, the ROV was tested, and the ship then left the bay and returned to the last CTD station (84) location and headed eastward towards Lazarev Bay where the LMG was working.  During the roughly 12 hours spent deep in the bay, the surface forcing conditions showed some of the strongest surface cooling during the entire cruise.  This note summarizes that period.

            The NBP was deep into the bay southeast of Charcot Island during the period YD 141.4‑ 142.05. Winds during this period were relatively weak, dropping from about 20 to 6 kts, primarily from the north.  The air temperature dropped as we entered the bay, from roughly ‑2EC to a minimum of ‑5EC.  The sea surface temperature also decreased as the ship got further into the ice and closer to the Wilkins Ice Shelf, from ‑1.40EC to a minimum of ‑1.73EC at our closest approach. The surface salinity decreased into the bay, with the minimum salinities being below 33 PSU.  The ice was thickest there, and the ship stopped to take ice samples and deploy the ROV.

            This bay appears to be protected from the strong southward winds that carry relatively warm and moist air over the shelf.   Instead, the air reaching the ship was quite cold and drier, suggesting it came from Charcot Island and the ice shelf.  The mean wind speed was 6.1 m s^-1 and the mean air‑sea temperature difference was Ta ‑ SST = ‑2.2EC.  This negative air‑sea temperature difference, lower mean relative humidity (81%), and modest winds combined to produce sensible and latent heat losses of on average 19‑24 W m^-2.  The sky was clear as the NBP entered the bay, resulting in a relatively large mean longwave heat loss of 111 W m^-2.  With essentially zero shortwave heating, the mean net heat flux was Q_net = ‑153 W m^-2 during the period the NBP was in the bay.

            A simple interpretation follows.  Charcot Island and the Wilkins Ice Shelf seem to shelter the Bay from the southward flow of warm, moist air that generally occurs over the continental shelf.  The air over the bay is colder and drier (from over land), leading to surface cooling by the sensible and latent components in addition to the longwave cooling. These three terms contribute to make the surface heat loss a regional maximum, causing more ice formation here. Whether this leads to a feedback process, whereby more ice means colder air leading to increased sensible and latent cooling and thus more ice, is not clear.

            After returning to the western side of Charcot Island, the NBP moved eastward to Rothschild Island and Lazarev Bay.  During this run, as the ship got into continuous ice, the air temperature continued to drop with larger sensible and latent cooling. The pattern of larger net heat loss within these more protected bays where ice is being formed seems robust, but needs to be examined with additional data collected during the rest of this cruise and data from LMG01‑04.

            Near the end of the cruise, the NBP made CTD station #100 in freshly formed ice within 5 nm of the ice cliffs of northwest Adelaide Island. The air was very cold and dry, gently flowing westward from the island over the near‑shore waters, and the sky was very clear.  These factors all contributed to a net surface cooling rate of Q_net = ‑ 197 W m^-2 (with 83% due to longwave cooling and the rest equally to sensible and latent cooling). A moderate wind could increase this loss to over ‑300 W m^-2. To put these cooling rates into perspective, surface cooling at 200 W m^-2 over a 5‑day period will cause the temperature of a 25‑m deep mixed layer to drop by 0.84EC.  Considering that the nearshore surface temperatures were typically below ‑1.0 to ‑1.4EC, it is not surprising that active ice formation occurred near the coast during NBP01‑03, driven primarily by continuous radiative cooling combined with episodic sensible and latent cooling.

 

4.0 Automated Weather Station Installation Report  (Bob Beardsley and Jeff Otten)

 

            Two Automated Weather Stations (AWSs) were deployed within Marguerite Bay during NBP01‑03.  Each AWS measures wind speed and direction, air temperature and pressure, and relative humidity, using sensors mounted on a 10-foot mast.  The propeller anemometer is centered at an approximate height of 3.4 m above ground, the air temperature and relative humidity sensors at 3.1 m, and the barometer at 1.5 m.   A data logger collects data from the various sensors and sends reformed data to an ARGOS satellite transmitter. The AWS is powered by lead‑acid batteries that are recharged using a solar panel mounted on the mast oriented north. The AWS units were supplied by Dr. Charles Sterns and George Weidner at the University of  Wisconsin Antarctic Meteorological Research Center (AMRC), who receive the ARGOS AWS data and place edited data on the AMRC website (www.uwamrc.ssec.wisc.edu/aws) for public use. A summary of the two stations is given next.  See Appendix 6 for a more complete description of the AWS deployment and repair operations.

            AWS #8930 was installed on the main island in the Kirkwood Islands group on 25 May  2001. The AWS site is on a slab rock shoulder on a ridge heading approximately northwest on the northwestern tip of the island.  The site has open exposure from west through northeast; winds from the south may be distorted by the main snow cap. On 25 May 2001, the station was revisited and the AWS data logger was reprogrammed to fix a software error found in the initial wind speed data received at the AMRC.  Subsequent data received at AMRC indicate that the wind speed problem was fixed and the station was reporting good data for all variables.

            AWS #8932 was installed on a small rocky island just east of Dismal Island in the Faure Island group in Marguerite Bay on 27 May 2001.  The island is relatively low, with snow covered ridges and exposed rocky patches on top.  The AWS was installed on a small smooth rocky plateau on the north end of the island, with open exposure to the west through southeast.  Data received at AMRC indicate that the AWS is working properly.

            Background information about these islands was supplied by Dr. Colin Harris, Environmental Research and Assessment, British Antarctic Survey, Cambridge, England.

                        AWS # 8930 (Kirkwood Island)

                        Latitude: ‑68E 20.397 S

                        Longitude:  ‑69E 00.444 W

                        Height of site above sea level: ~ 75 ft (crude estimate) (25 m)

                        Station orientation: 77EN

                        Installation: 25 May 2001; reprogrammed 27 May 2001

 

                        AWS # 8932 (Dismal Island)

                        Latitude: ‑68E 05.243 S

                        Longitude:  ‑68E 49.480 W

                        Height of site above sea level: ~ 35 ft (crude estimate)  (12 m)

                        Station orientation: 124EN

                        Installation: 27 May 2001          

 

5.0 Nutrients  (Kent A. Fanning (project PI, not on cruise), Rebecca Conroy, E. Howard Rutherford)

 

5.1 Introduction

 

            It is reasonable to state that, after temperature and salinity, dissolved inorganic nutrients (nitrate, nitrite, phosphate, ammonia, and silica) are central to understanding the circulation of waters in and around Marguerite Bay.  Deeper water upwelling to shallower regions close to the Peninsula should be traceable by higher nutrient signatures.  Nutrient concentrations nearer to the sea surface are important to physical/chemical modeling of the fate of plankton in the region that sustain krill, both as "targets" to be explained by nowcasting and as starting points for forecasting.

 

5.2 Methods

           

            Analytical methods used for silica, phosphate, nitrite, and nitrate follow the recommendations of Gordon et al. (1993) for the WOCE WHP project.  The analytical system we employ is a five-channel Technicon Autoanalyzer II upgraded with new heating baths, proportional pumps, colorimeters, improved optics, and a computer-controlsystem (New Analyzer Program v. 2.40 by Labtronics, Inc.)  This Technicon is designed for shipboard, as well as laboratory, use.  Silica is determined by forming the heteropoly acid of dissolved orthosilicic acid and ammonium molybdate, reducing it with stannous chloride, and then measuring its optical transmittance.  Phosphate is determined by creating the phosphomolybdate heteropoly acid in much the same way as with the silica method.  However, its reducing agent is dihydrazine sulfate, after which its transmittance is also measured.  A heating bath is required to maximize the color yield.  Nitrite is determined essentially by the Bendschneider and Robinson (1952) technique, in which nitrite is reacted with sulfanilamide (SAN) to form a diazotized derivative that is then reacted with a substituted ethylenediamine compound (NED) to form a rose pink azo dye which is measured colorimetrically. Nitrate is determined by difference after a separate aliquot of a sample is passed through a Cd reduction column to covert its nitrate to nitrite, followed by the measurement of the "augmented" nitrite concentration using the same method as in the nitrite analysis.

            In the analytical ammonia method, ammonium reacts with alkaline phenol and hypochlorite to form indophenolblue. Sodium nitroferricyanide intensifies the blue color formed, which is then measured in a colorimeter of our nutrient analyzer.  Precipitation of calcium and magnesium hydroxides is eliminated by the addition of sodium citrate complexing reagent.  A heating bath is required.  Our version of this technique is based is based on modifications of published methods such as the article by F. Koroleff in Grasshoff (1976).  These modifications were made at Alpkem (now Astoria-Pacific International, Inc.) and at L.Gordon's nutrient laboratory at Oregon State University.

 

5.3 Data

 

            Nitrate, nitrite, phosphate, ammonia, and silica were measured in all hydrocasts on this cruise (Hydrography and Circulation Component, Appendix 3).

 

5.4 Preliminary Results

 

            Nutrient data show considerable structure along and across the Western Antarctic Peninsula continental shelf within the SO GLOBEC study region.  Regions of upwelling and downwelling are clearly evident in the nitrate and silicate distributions.  The ratio of silicate to nitrate was used to track the upwelling of Upper Circumpolar Deep Water within the study area. 

            Silica draw-down was also a feature observed along the shelf region.  One possible explanation may be due in part to nutrient-rich deep water upwelling into a region with a high relative abundance of diatoms.  The diatoms should start to consume nutrients faster than other, non-siliceous phytoplankton.  Therefore, there would be a greater consumption of silica than if diatoms were in low abundance.  If the ship passed through after the silica consumption started, lower silica:nitrate ratio would be observed than at depths from which the water originally upwelled.  Support for this explanation is dependent on determining phytoplankton species composition within the regions where silica draw-down was observed.

            High ammonia concentrations were observed at stations closer to land and further south along the shelf region.  Highest ammonia values, greater than 4.5 Fmol, were measured from stations within Marguerite Bay.  Reduced NO₃ and NO₂ values were also associated with the stations mentioned above.  Since all of these components are in dynamic balance moderated by microorganisms, the answer to why ammonia is so abundant could relate to its production rate being enhanced, possibly by large Antarctic krill populations or its consumption rate slowing down.  The two most likely ways that ammonia is consumed are uptake by primary producers (maybe low now that it's almost winter) and nitrification, in which bacteria oxidize ammonia to nitrite and nitrate.  Final analysis of Antarctic krill distribution patterns, along with nitrifying bacteria studies, during NBP01-03 are essential in determining processes contributing to the high ammonia measured.

 

5.5 References

 

Gordon, L.I., J.C. Jennings, Jr., A.A. Ross, and J.M. Krest, A Suggested Protocol For Continuous Flow Automated Analysis of Seawater Nutrients, in WOCE Operation Manual, WHP Office Report 90-1, WOCE Report 77, No. 68/91, 1-52, 1993.

Grasshoff, K., Methods of Seawater Analysis, Verlag Chemie, Weinheim, Germany, and New York, NY, 317 pp, 1976.

 

6.0 Primary Production  (Maria Vernet (project PI, not on cruise), Wendy Kozlowski, and Michael Thimgan)

 

6.1 Introduction

 

            The estimation of primary production has three main objectives: (1) estimation of primary productivity rates during fall and winter in the area of study as a possible source of food for Antarctic krill and other zooplanktors;  (2) understanding the mesoscale patterns of phytoplankton distribution with respect to physical, chemical and biological processes; and  (3) obtaining insight into the over-wintering dynamics of phytoplankton, including their interaction with sea ice communities.  For this purpose, primary production was measured with three methods during this cruise: Photosynthesis versus Irradiance (PI) curves to estimate potential primary production and information on the dynamics of light adaptation; simulated in situ (SIS) experiments to estimate daily primary production; and finally, profiles with a Fast Repetition Rate Fluorometer (FRRF), with the aim to increase resolution in the sampling of phytoplankton activity and the expectation of modeling primary production with this method using ¹⁴C experiments as comparison.  Additionally, measurements of chlorophyll and particulate carbon (POC) were taken for estimates of phytoplankton biomass and data collected from surface and profiling Photosynthetically Available Radiation (PAR) sensors.

 

6.2 Methods

 

            For production and POC sampling, stations were broken up into three different categories.  At Priority 1 stations, both PI and SIS experiments were done, and POCs were collected at all the primary depths (see below).  At Priority 2 stations, PI experiments were done, and POCs were collected at two other depths; and at Priority 3 stations, PI experiments were again done, and POCs were sampled only at one depth.   Priority 1 stations were wherever SIS experiments were done (see below).  Priority 2 stations were those that ran along the outermost line, the innermost line along the coast, and the 460.xxx, the 340.xxx, the 260.xxx and the 140.xxx lines.  All remaining stations were Priority 3.

 

6.2.1 Location

            PI experiments were done at all stations, except consecutive stations #58, 59, 85, 86, and 100.  When the weather did not allow deployment of the Rosette or if it was too rough to collect a surface sample with the Rosette, water was collected from the surface with a bucket and processed.  SIS experiments were done approximately once per day, with an attempt to sample evenly both inshore and off, and in the northern and southern parts of the grid and within the Marguerite Bay.  The FRRF was deployed at all stations where the Rosette was used, through consecutive station #49.  Chlorophylls were sampled from all stations where the Rosette was deployed, as well as from the bucket samples when necessary.

 

6.2.2 Depths

            For the PI curves, water was collected from the Rosette bottle that corresponded most closely to a depth of 5 m.  For the SIS experiments, water was collected at what was called the primary depths: surface, and at 5, 10, 15, 20, and 30 m.  While it was working, the FRRF was deployed as part of the CTD Rosette, to a depth of 50 m, with a descent rate of only 10-15 m per minute, somewhat slower than the standard CTD casts.  At stations where only PI curves were done, POC samples were taken from at least the 5 m depth and along.  At stations where SIS experiments were done, POC samples were collected from at least the surface, 5, 10, 15, 20, and 30 m.  Chlorophylls were collected at the same depths as those for the SIS experiments, plus an additional standard depth of 50 m.  Occasionally,  when the CTD fluorometer trace showed deep water fluorescence, additional samples were taken at depths between 50 and 210 m (see Section 1.2, Appendix 3).

 

6.2.3 Sea Ice Sampling

            Sea ice sampling took place whenever it was in the vicinity of the ship and reasonably accessible.  This was done by one of the following methods:  1)  when open water was available, off the side of a deployed zodiac;  2) from personnel carrier placed on large pancake type ice; or 3) using a weighted bucket lowered directly off the side of the ship.

 

6.2.4 Equipment

            Chlorophylls were measured using a Turner Designs Digital  10-AU-05 Fluorometer, serial number 5333-FXXX, calibrated using a chlorophyll a standard from Sigma Chemicals, dissolved in 90% acetone.  The “Fast Tracka” Fast Repetition Rate Fluorometer, serial number 182037,  is made by Chelsea Instruments and was outfitted with independent depth and PAR sensors.  All data were recorded internally to the instrument and data were downloaded directly to a computer after ever few casts.  Incubations for the SIS experiments were done in Plexiglas tubes, shaded to simulate collection light levels with window screening, incubated in an on-deck Plexiglas tank, which was outfitted with running seawater in order to maintain in situ temperatures.  PI curves were done in custom built incubators, designed to hold 7ml vials, irradiate at light levels between zero and 460 µE (m²s)^-1, and were attached to water baths to maintain in situ collections temperatures.  POC samples will be analyzed upon return to the United States.  Light data were collected using a Biospherical Instruments GUV Radiometer, serial number 9250, mounted on the science mast and configured with a PAR channel, as well as channels for 305, 320, 340, and 380 nm wavelengths.  Additional PAR data were collected on six days using a Biospherical Instruments QSR-240 sensor, serial number 6357, which was also mounted on the science mast.

 

6.3 Data Collected

 

            Over the course of the 35 science days of this trip, we have carried out a total of 139 PI experiments at 96 of the 101 stations sampled.  Thirteen of those experiments were done on samples taken from buckets at 12 different locations.  An additional nine PI curves were done on seven different sea ice samples (Table 8).  A total of 23 SIS experiments were completed (Table 9), and the FRRF was deployed at a total of 40 stations before an electrical failure occurred, causing damage to at least one internal card in the instrument.  Though the RPSC  Electronics Technicians worked diligently fix it, it proved to be unrepairable and the instrument will be returned to Chelsea for repair and return before the July cruise.

            For estimations of biomass (standing carbon stocks), both POC and chlorophyll samples were taken.  A total of 101 POC samples were taken (plus blanks), of which nine were sampled from buckets and eight were from sea ice samples.  Seven hundred chlorophyll samples were taken from 82 different CTD stations,  17 bucket stations, and eight ice stations.  For a summary of the bucket stations, see Table 10.

            Surface PAR data were collected on all days that primary production experiments were done (Table 11).  Due to a combination of a loose connection and high winds, no GUV data were collected on 14 May 2001.  GUV data were collected at one minute intervals and logged directly to a computer.  QSR data were collected two ways:  1) as part of the meteorological data set, logged as raw voltage; and 2) onto a LICOR LI-1000 data logger, also logged as raw voltage, but with four additional decimal points for increased resolution.  A comparison of the two instruments was done to determine collection differences between the two types (scalar vs. cosine) of sensor.  PAR data were also collected during each daylight CTD cast using a profiling PAR sensor and will be used in conjunction with surface PAR data for the analysis of water column production.

 

6.4 Preliminary Results

 

            Final analysis is yet to be completed on the majority of the data collected on this cruise.  However, there appears to be a stronger North-South trend in the chlorophyll data than an onshore-offshore trend, with noticeably higher chlorophyll levels in the northern, offshore part of the grid.  There also was a high chlorophyll spike on the northern, inshore portion of the grid, as well as slightly higher levels in the northern part of Marguerite Bay.  Water column primary production levels seem to mirror this, with the highest production matching the location of the highest chlorophyll values, in the areas of consecutive stations #23 through 25.

 

Table 8.   Summary of stations where sea ice was sampled.

 

Sample #	Latitude (ES)	Longitude (EW)	Sample Description	Collection Method	Samples Taken
Ice 1	-68.738	-70.983	slush around early pancakes	zodiac	chl, POC, CHN, PI, nutrients
Ice 2	-68.738	-70.983	early pancakes	zodiac	chl, POC, CHN, nutrients
Ice 3	-70.325	-75.155	slush between large pancakes	personnel carrier	chl, POC, CHN, nutrients
Ice 4	-70.295	-75.299	slush between large pancakes	personnel carrier	chl, POC, CHN, PI, nutrients
Ice 5	-70.295	-75.299	large pancake	personnel carrier	nutrients
Ice 6	-70.302	-75.618	formed pancake	zodiac	chl, POC, CHN, PI, nutrients
Ice 7	-79.587	-74.607	early pancakes	bucket	chl, POC, CHN, PI, nutrients
Ice 8	-69.257	-72.492	early pancakes	bucket	chl, POC, CHN, PI, nutrients
Ice 9	-68.763	-71.408	crystalline slush	zodiac	chl, POC, CHN, PI, nutrients

 

 

Table 9.  Summary of SIS stations, and preliminary primary production estimates, with

values integrated to 30 m.  PAR is as measured by the GUV, over the duration of each

experiment (except consecutive station #99, where PAR numbers are estimates based on the

QSR240 sensor). Note that these values are only preliminary (*) estimates and further refinement

and QC will be done upon return to the United States.

 

Consecutive Station Number	Location	Primary Production* (mgC/m2/day)	PAR (µE/cm2/expt)
3	500.180	15.0	49.3
9	460.220	36.7	193.9
14	420.145	19.1	66.2
23	340.253	171.9	171.1
28	335.060	9.5	53.7
33	300.-020	4.5	46.1
41	260.295	5.5	47.4
51	215.-015	1.6	26.3
53	220.075	2.1	40.9
55	220.140	3.3	not available
57	220.220	4.0	not available
68	180.100	0.2	7.6
69	140.100	1.4	10.7
72	140.220	1.0	11.7
77	100.140	0.9	12.5
81	060.255	0.6	11.4
84	020.180	1.2	10.7
87	062.122	0.6	7.6
89	239.057	1.2	17.4
90	367.036	0.3	6.0
91	338.044	1.4	14.4
95	344.052	4.2	44.0
99	353.099	3.2	44.8
101	372.110	5.6	49.2

 

 

Table 10.  Summary of stations where samples were taken from a bucket, either in addition to,

or instead of, sampling with the Rosette on the CTD. 

 

Sample #	Grid Location	Reason	Samples Taken
Bucket 1	300.140	No CTD	chl, POC, PI
Bucket 2	300.180	No CTD	chl, POC, PI
Bucket 3	300.220	No CTD	chl, POC, PI
Bucket 4	300.265	No CTD	chl, POC, PI
Bucket 5	260.180	No Surface Bottle	chl
Bucket 6	260.140	No CTD	chl, POC, PI
Bucket 7	260.100	No CTD	chl, POC, PI
Bucket 8	255.080	No CTD	chl, POC, PI
Bucket 9	267.057	No CTD	chl, POC, PI
Bucket 10	220.100	No CTD	chl, POC, PI
Bucket 11	220.220	No Surface Bottle	chl , POC, SIS
Bucket 12	220.250	No CTD	chl
Bucket 13	220.265	No CTD	chl
Bucket 14	220.280	No CTD	chl
Bucket 15	220.295	No CTD	chl, POC, PI
Bucket 16	140.255	No CTD	chl, POC, PI
Bucket 17	100.255	No CTD	chl, POC, PI

 

 

Table 11.    PAR (Photosynthetically Available Radiation, 400 – 700 nm) data, from

BSI GUV500 mounted on Science Mast.  Day lengths and daily irradiance values are calculated

using PAR values above 0.0 µE/cm²*sec.  Values for 14 May (*) are missing due to instrument

failure, and values for 30 May (**) are estimated from the BSI QSR240 sensor.  A correction

factor was applied to the QSR data to accommodate the difference between the cosine and

scalar sensor types.

 

Date	Sunrise	Sunset	Day Length	µE/cm2
29 Apr	12:21	20:57	8.60	295.06
30 Apr	12:26	20:35	8.15	49.26
1 May	12:25	20:55	8.50	193.91
2 May	12:38	20:38	8.00	171.82
3 May	12:46	20:29	7.72	61.61
4 May	12:54	20:38	7.73	89.87
5 May	12:47	20:28	7.68	171.10
6 May	13:12	20:05	6.88	53.71
7 May	12:51	19:56	7.08	57.47
8 May	13:12	20:10	6.97	47.93
9 May	13:16	20:26	7.17	69.49
10 May	13:13	19:58	6.75	47.40
11 May	13:23	19:46	6.38	49.63
12 May	14:18	19:29	5.18	26.27
13 May	13:37	19:52	6.25	40.93
14 May*	13:54	not available	not available	not available
15 May	14:01	19:52	5.85	23.88
16 May	14:08	19:19	5.18	7.80
17 May	14:19	19:40	5.35	10.73
18 May	14:23	19:12	4.82	7.75
19 May	14:27	19:28	5.02	12.50
20 May	14:34	19:18	4.73	11.42
21 May	14:41	19:10	4.48	6.94
22 May	14:35	19:18	4.72	10.78
23 May	14:35	19:06	4.52	7.63
24 May	14:22	19:00	4.63	8.46
25 May	14:02	19:11	5.15	17.43
26 May	14:06	19:15	5.15	17.58
27 May	14:16	18:49	4.55	6.53
28 May	14:14	18:43	4.48	9.43
29 May	14:15	18:42	4.45	5.83
30 May	14:14	18:59	4.75	14.42
31 May**	13:59	19:17	5.30	44.05
1 Jun	13:55	19:27	5.53	28.74
2 Jun	13:31	19:32	6.02	49.17

 

 

 

 

7.0 Microplankton studies  (Scott Gallager, Karen Fisher, Susan Beardsley)

 

7.1 Objectives