I. Southern Ocean GLOBEC
Moored Array, Drifter, and Float Component
Preliminary Results from the 2001 Drifter Program
The International Southern Ocean Global Ocean Ecosystems Dynamics (SO GLOBEC) program is designed to investigate the effects of shelf circulation and sea ice on Antarctic krill (Euphausia superba) distribution, and examine the factors that govern krill survivorship and availability to higher trophic levels, including seals, penguins, and whales. Of particular interest to SO GLOBEC are the factors regulating growth and survival of Antarctic krill during the austral winter.
The SO GLOBEC study site is the central West Antarctic Peninsula (WAP) shelf and Marguerite Bay area (Fig. 1). The bathymetry of this area is quite variable, especially within Marguerite Bay where many shallow reefs and small islands occur. A deep channel starting near the shelfbreak runs southeastward across the shelf off Adelaide Island, crosses Marguerite Bay, and continues southward around Alexander Island as George VI Sound. Past observations suggest that this region is characterized by unusually high krill production, which may result from a unique combination of physical and biological factors that contribute to enhanced krill growth, reproduction, recruitment and survivorship throughout the year, persistent winter sea ice cover, and the presence of predators in winter.
The U.S. contribution to the SO GLOBEC field program includes studies of surface forcing, ocean circulation, sea ice, microbial communities, zooplankton, Antarctic krill, and top predators. A sequence of austral fall and winter cruises in 2001 and 2002 will support both broad-scale surveys and process studies. The program also features a large physical oceanography component that includes deployments of a array of instrumented moorings and satellite-tracked surface drifters and isobaric floats. The moored array will be deployed in austral summer 2001, serviced the next summer, and then recovered in 2003. These Eularian and Lagrangian measurements are designed to investigate seasonal changes in water properties and the regional circulation and complement the ship-board physical and biological measurements made during the broad- scale and process cruises. A detailed description of this component is given below.
The U.S. SO GLOBEC program is funded
by the National Science Foundation Office of Polar Programs. The
program central planning and data management office website is
which provides a complete set of program planning reports,
descriptions of individual components, planned cruises, and links to
related websites. An initial description [add link to the SSC report
by hofmann et al] of the U.S. SO GLOBEC field study and related
international participation in SO GLOBEC has been prepared by the
science steering committee. Additional information on the SO GLOBEC
program is available at http://cbl.umces.edu/fogarty/usglobec
Figure 1. Study region for the U.S. SO GLOBEC field
program. The mooring site locations are designated by A1, A2, A3 and
B1, B2 and B3. The bathymetry is based on ETOPO2-v8.2A.
Recent hydrographic data provide a clear picture of the seasonal evolution of stratification over the central WAP shelf (Hofmann et al, 1996; Smith et al, 1999). An on-shelf flow of Upper Circumpolar Deep Water (UCDW) maintains a relatively warm, salty, nutrient-rich lower layer over the shelf beneath a upper layer of relatively cooler, fresher Antarctic Surface Water (AASW) (Fig. 2). The two layers are separated by a permanent pyncocline centered between 150 to 200 m, which provides a vertical flux of heat and salt into the upper layer (Klinck, 1998). Over much of the shelf, net surface heating in austral spring through fall (~ 80 W/m^2) warms the upper water column to depths of 50 to 100 m, while increased wind mixing and surface cooling in fall through winter (~ -80 W/m^2) creates a 100-m deep surface mixed layer. This seasonal cycle of deep mixing and re-stratification helps transport nutrients upward.
Fig. 2. Vertical profiles of temperature and salinity obtained during 1993 austral summer (S), fall (F) and winter (W) just off the northwest corner of Adelaide Island. At these cold temperatures, density is determined primarily by salinity. Note the cooling and freshening of the surface 50 m from summer to fall, indicative of a southwestward buoyant coastal current; the deep winter surface mixed layer; the summer and fall temperature minimum layer at 100 m, the remnant of last winter's mixed layer (called Winter Water, WW); the permanent pycnocline (150 to 200 m); and the deeper UCDW. The crosses (+) shown between the two panels indicate the common measurement depths in the moored array. (Data from Smith et al, 1999.)
In comparison, our present knowledge
of flow over this shelf is quite limited, being based primarily on
one FGGE drifter track from 1979, recent hydrographic data,
incomplete knowledge of regional wind forcing, and global tidal
models (Hofmann and Klinck, 1998). The FGGE drifter moved
northeastward along the shelf break during austral spring, then
crossed the shelf near 65.5 S and moved southwestward to Adelaide
Island (AdI) in summer (mean speed ~ 6 cm/s) before failing. The 1993
summer and fall salinity sections off AdI show a coastal band of
fresher (lighter) water, which increased in depth, width, and
salinity contrast into fall, when it extended to ~ 50 m depth and ~
70 km offshore with a difference (from adjacent water) of ~ 0.5 psu
(Fig. 2; Smith et al, 1999). These observations suggest the formation
of a buoyant coastal current during spring melt, which continued to
strengthen into fall towards the southwest. NSCAT-derived monthly
mean winds over the Marguerite Bay/WAP shelf region are southwestward
(down-welling- favorable) from spring into fall, which should help
maintain and strengthen a coherent coastal current. The drifter
track, hydrography, and dynamic topography support the idea of a
large but slow clockwise circulation gyre over the shelf to the north
and west of Marguerite Bay in the warmer seasons (Smith et al.,
1999). The lack of strong geostrophic shears suggests that this flow
may have a significant barotropic component, subject to topographic
steering. The existence of a clockwise gyre in winter, when the NSCAT
winds become northeastward (upwelling-favorable), is not known. Tidal
currents are predicted to be <10 cm/s over much of the shelf,
except along the shelfbreak where tidal resonances can occur and
perhaps over the shallow southern end of AdI due to tidal flow into
Marguerite Bay proper.
2. Scientific Goals and Approach:
The primary goals of this component are to: a) measure the temporal and spatial variability of currents and physical and biological water properties in the Marguerite Bay/WAP shelf area on time scales from hours to seasonal; b) improve our description and understanding of the regional general circulation; and c) identify and describe those physical processes which make this region well suited for krill production and survival.
Our approach uses a combination of
Eulerian and Lagrangian measurements. First, an array of six
subsurface moorings instrumented to measure currents, temperature,
conductivity (salinity), zoo-plankton concentration, and ice movement
will be deployed in the study area for a period of one year (Fig. 1,
Table 1). This array will monitor Eulerian currents and the seasonal
progression in water structure throughout the intensive austral
winter study (April-November 2001). A northern subset of this array
will be re-deployed to observe conditions during the second winter
study in 2002. Second, Lagrangian observations of the near-surface
(15 m) and subpycnocline (250 m) flow will be made using
satellite-tracked surface drifters and isobaric floats during
ice-free conditions. In addition, supporting data will be collected
to help document surface forcing during the 2001-2003 field work.
3. Proposed Field Program
A) Moored Array
The proposed moored array consists of
two cross-shelf transects radiating roughly west (A) and southwest
(B) from AdI (Fig. 1, Fig. 3). Transect A features three subsurface
moorings spanning the shelf from near AdI to the outer shelf. A1 is
located near the steep flank of AdI, in part to observe the
surface-intensified coastal current thought to exist during spring
through fall. A3 is located to observe possible episodic forcing by
the Antarctic Circumpolar Current (ACC) along the shelf break.
Transect B features three subsurface moorings spanning Adelaide
Basin, with B1 and B3 on the east and west flanks and B2 in the
center. This line is designed to look for clockwise flow around the
basin, the southwest continuation of the surface coastal current
towards the mouth of Marguerite Bay, and the filling and flushing of
the deepest part of the basin. The combined array will examine the
spatial and temporal variability of the large clockwise gyres thought
to exist over the shelf plus tidal and transient flows driven by
storms and southern meanders of the ACC.
Figure 3. Schematics of the A and B mooring transects. Bottom topography based on ETOPO2-v8.2A.
All moorings will measure the following physical variables at common depths: a) temperature at 50, 100, 150, 200, 250, 400 m, and bottom using SeaBird (SBE) temperature sensors; b) conductivity (salinity) at 50, 100, 250, and 400 m using SBE conductivity cells; and c) currents at 250 and 400 m using Vector Averaging Current Meters (VACMs). To obtain vertical profiles of currents and acoustic back-scatter intensity in the upper water column where krill are commonly found (Lascara et al, 1999), 300 kHz broadband RDI ADCPs would be deployed at 100 m on all moorings except B2, which would feature a 150 kHz broadband RDI ADCP moored at 250 m. Additional temperature and conductivity sensors would be deployed to improve vertical resolution through the pycnocline on the A3 and B2 moorings. These additional sensors will provide more detailed information about internal tides and waves that might be generated near the shelf break and propagate on-shelf and/or are generated more locally in Adelaide Basin.
Most moorings will feature SBE Paroscientific bottom pressure gauges to obtain both across- and along-shelf measurements of the pressure field over the shelf. We anticipate that this array will capture tidal signals, which propagate through this area, plus any shelf-scale wind- driven response. Co-located bottom pressure and water column density measurements also allow estimation of the internal pressure field, which will help understand the dynamics of the observed flows.
Two ASL Environmental Sciences ice-profiling sonars will be mounted at the top of the A2 and B2 moorings, to monitor ice thickness and velocity.
As mentioned above, the RDI broadband ADCPs also measure acoustic back- scatter intensity (Dienes, 1999). The broadband ADCP normally records the Received Signal Strength Indicator (RSSI) for each beam during each ping. To convert this to echo intensity, a unique (temperature- independent) RSSI scale factor can be determined for each transducer. These scale factors will be used with the RSSI data and moored measurements of water structure above the ADCP can be used to compute the absolute back-scatter coefficient as a function of range. This data will provide unique one- year time series records of the vertical distribution of krill and other zooplankton. Based on limited acoustic and net data (Lascara et al, 1999), we anticipate a cross-shelf and vertical change in the krill distribution from summer to winter. Detailed information on zoo- plankton distribution and acoustic signatures at ADCP frequencies will be obtained from shipboard sampling with a towed multi-frequency sonar system and net tows collected by other investigators, who will help with the interpretation of the moored ADCP back-scatter data.
We note that this mooring plan is
based on the assumption that only shallow icebergs (< 50m deep)
traverse the mooring sites. Preliminary information suggests that
this is correct. Addition information will be collected and, if
necessary, the moorings shortened so that the uppermost instruments
are not endangered. In addition, we plan to use two primary
floatation spheres on each mooring, one just above the 250-m
instrument and one at the top of the mooring, plus backup floatation.
Both floats will be equipped with subsurface ARGOS beacons, so that
we should be able to locate and recover the mooring parts if either
float surfaces prematurely.
B) Lagrangian Measurements
A primary objective of this component is to investigate the Lagrangian circulation (fluid particle motion over time) within the study area using satellite-tracked surface drifters and isobaric floats. These instruments provide unique information on the horizontal movement of fluid and passive biological particles that can not be obtained through moored measurements alone. Both Eulerian (fixed) and Lagrangian flow measurements are needed to describe the general circulation. Both drifter and float data will be processed in near-real time and quickly posted on this web site for use by all program investigators.
B1) Surface Drifter Measurements
We plan to deploy a total of 20 SVP surface drifters each year during both field years. The initial plan is to deploy 6 on our April 2001 mooring deployment cruise (one at each mooring site), 4 on the April German cruise in Marguerite Bay, 10 on the May 2001 broad-scale survey, another 10 during the November 2001 British cruise, and then five each during the April and May 2002 cruises. Depending on the movement of this initial set, release positions for later deployments may be modified. We anticipate that many of these initial drifters will remain within the study area for several months before exiting towards the northeast and/or becoming frozen in the ice. We plan not to release any drifters during ice-covered conditions, but want to use them to study the near- surface flows during the austral spring through fall ice-free conditions, when the large-scale surface-intensified clockwise gyre is thought to exist.
The SVP drifter is designed to measure the Lagrangian current using a holey sock drogue centered at 15 m plus sea surface temperature and a proxy to indicate if the drogue is attached. The drifter is located at least 6 times per day in the study area, and the position and other data provided daily via Internet by Service ARGOS. Half of the drifters will feature an ice-hardened surface float to resist being crushed in the ice, and the other half will be supplied by the Global Drifter Program with barometric pressure sensors. All SO GLOBEC drifter data will be submitted to the Global Drifter Program data archive.
B2) Isobaric Float Measurements
We also plan to deploy six isobaric floats each year during both field years. An initial deployment of 2-3 floats will be made in Marguerite Bay during the German SO GLOBEC April 2001 cruise. This will serve as an engineering test of this new float design and a first look at the Lagrangian flow within Marguerite Bay. Data from this initial deployment will be used to design subsequent deployment patterns and timing.
These PALACE-type floats are designed to first sink to a preset pressure (depth), then drift at that pressure for a preset period, then surface to transmit its GPS position and other in-situ data via satellite (ORBComm) to WHOI before repeating this cycle. Each float will carry SBE temperature and conductivity sensors to measure temperature and salinity since both are needed to identify water masses in the study area. The average temperature, salinity, and pressure during the float's constant pressure drift plus rapidly sampled temperature, conductivity and pressure data collected during the float's ascent will be transmitted to WHOI within a few minutes of the float reaching the ocean surface. This allows the float to return to its preset pressure quickly, thus improving the ability of the float to follow the same water parcel.
Since little is known about the deeper flow and vertical shear in the study area, the floats used in the initial deployment will be set todrift at a depth of 250 m with a cycle time of 50 hrs, chosen to obtain high temporal resolution with a minimum of tidal aliasing. The floats require about 15 min to descend to depth and the same to surface, only 1% of the 50-hr cycle time. This, coupled with GPS positioning, should reduce the inherent error due to vertical shear and position error to <0.25 cm/s.
C) Supporting Data
C1) Surface forcing
The U.S. research vessels NBPalmer and LMGould are equipted to collect a full suite of standard meteorological measurements (winds, air temperature, relativel humidity, barometric pressure, downward short- and long-wave radiation, and water temperature) on each SO GLOBEC cruise. We plan to use this shipboard data to estimate the surface wind stress and heat flux during each cruise. This data will be used with satellite scatterometer wind data and available surface weather analysis products to construct a detailed description of the spatial and temporal structure of the wind forcing during the two-year field study.
A preliminary look at the NSCAT monthly mean winds for the study area shows a seasonal difference from ECMWF climatology. To replace NSCAT (which failed June 1997), NASA launched SeaWinds on QuikSCAT on June 17, 1999. A second SeaWinds scatterometer will launch aboard ADEOS-II in 2001. We plan to capture data from both SeaWinds scatterometers as soon as it becomes available, and develop a climatology for the WAP shelf region. In particular, we hope to use this scatterometer data to develop monthly mean maps and explore the spatial structure of the wind field over the shelf during specific wind events. This effort has two main objectives, to improve our understanding of the regional atmospheric forcing and develop a time series of wind stress maps for the study area to help interpretation of the observed currents and water property evolution.
The best digital bathymetry for the SO GLOBEC is ETOPO2-v8.2A. This database is a combination of the most recent version of the Sandwell and Smith 2-min global bathymetry (ETOPO2) and additional depth data from older regional navigational charts. One objective of the U.S. SO GLOBEC program is to collect trackline bathymetry data and multibeam data that will be used to improve our knowledge of the topographic setting of this program and update ETOPO2- v8.2A. One important application of an improved bathymetry is in numerical circulation and tidal models.
Hofmann, E. E. and J. M. Klinck, 1998. Hydrography and circulation of the Antarctic continental shelf: 150E to the Greenwich Meridian. Ch 35, in The Sea, 11, 997-1042.
Klinck, J.M., 1998. Heat and salt changes on the continental shelf west of the Antarctic Peninsula between January 1993 and January 1994. JGR,103(4),7617-7636.
Hofmann, E. E., J. M. Klinck, C. M. Lascara, and D. A. Smith, 1996. Water mass distribution and circulation west of the Antarctic Peninsula and including Bransfield Strait. Foundations for Ecological Research West of the Antarctic Peninsula, AGU Antarctic Research Series, 70, 61-80.
Smith, D. A., Hofmann, E. E., J. M. Klinck, and C. M. Lascara, 1999. Hydrography and circulation of the West Antarctic Peninsula Continental Shelf. Deep-Sea Res., 46(6), 925-950.
Lascara, C. M., E. E. Hofmann, R. M. Ross, and L. B. Quetin, 1999. Seasonal variability in the distribution of Antarctic krill, Euphausia superba, west of the Antarctic Peninsula. Deep-Sea Res., 46(6), 951-984.
Deines, K. L., 1999. Backscatter Estimation Using Broadband Acoustic Doppler Current Profilers. Proceedings of the IEEE 6th Working Conference on Current Measurement, San Diego, CA, March.