Observations of the mass and flow field at Porcupine Bank

Christian Mohn 0 Joachim Bartsch 0 Jens Meincke 0 0 C. Mohn: Alfred Wegener Institute for Polar and Marine Research , Bremerhaven, Germany. J. Bartsch: HYDROMOD - SEAMAR Project Office, Mittelfeldweg 18e, 27607 Langen , Germany. J. Meincke: Institute of Oceanography, Center for Marine and Alfred Wegener Institute for Polar and Marine Research , Postfach 120161, 27515 Bremerhaven , Germany. Tel.: During spring 1994 and summer 1995 hydrographic transects across the Porcupine Bank, an elliptical topographic structure adjoining the shelf-edge west of Ireland, were carried out to investigate the thermohaline properties and flow field characteristics in the vicinity of the bank. The CTD observations show a dome-like deformation of the temperature and density field together with an intrusion of cold, dense water above the bank summit. Additional acoustic current measurements in summer 1995 indicate that the dome-like perturbation of the mass field is accompanied by an anti-cyclonic, bottom-intensified circulation along the flanks of the bank. The doming of the near-bank temperature and density field and the eddy-like pattern of the flow field in summer 1995 may result from a Taylor column formation. It is suggested that a persistent Taylor column circulation around Porcupine Bank provides an important mechanism for the retention of pelagic eggs and larvae of the various marine species spawning in the area. 1054-3139/02/040380+13 $35.00/0 - The investigation of the flow regime over seamounts has received growing attention during the past decade. Field observations on a number of isolated seamounts, including the Great Meteor Seamount in the North Atlantic (Meincke, 1971) and the Fieberling Guyot in the North Pacific (Kunze and Toole, 1997), identified dome-like deformations of the temperature and density field as well as substantial perturbations of the flow field in the vicinity of the seamount summit. From theoretical considerations the dynamics involved were interpreted as mechanisms based on the theory of Taylor column formation and the generation of freely-propagating trapped waves at isolated topographic features. Numerical studies at isolated seamounts with both idealized (Brink, 1990; Chapman and Haidvogel, 1992) and realistic topography (Beckmann and Haidvogel, 1997) showed that these processes lead to closed, anticyclonic re-circulations around the seamount. A number of geological and biological studies (e.g. Genin et al., 1986; Boehlert, 1988) suggest that seamount-related currents are an important factor in the distribution of fauna and sediments (Genin, 1989). An overview of comprehensively investigated seamounts is given by Rogers (1994). Near-shore banks, including Georges Bank off the east coast of North America (Loder et al., 1988) and Rockall Bank in the north-eastern North Atlantic (Dooley, 1984) have received special attention due to their importance for regional fisheries. The present study focuses on the hydrographic investigation of the area surrounding Porcupine Bank based on CTD and acoustic current measurements. The observations were carried out in the framework of the EU/AIR funded project SEFOS (Shelf-Edge Fisheries and Oceanography Study). During different seasons between March 1994 and July 1995 the data were obtained along zonal transects through the Porcupine Bank area. The observations are supplemented by sea surface temperature (SST) satellite measurements. In this study the properties of the stratification and flow field in the central Porcupine Bank relative to its surroundings in summer 1995, are described and compared with hydrographic measurements in spring 1994. Dynamical aspects of the observed flow pattern are discussed with respect to results of recent numerical and field studies of flow in the vicinity of seamounts. Materials and methods The area of investigation covers the central Porcupine Bank, an elliptically-shaped topographic feature, situated at the north-west European shelf-edge west of Ireland (Figure 1). Its main axis is orientated essentially March/April 1994 April/May 1994 June/July 1995 49N 16W 15W 14W 13W 12W 11W 10W 9W in the northsouth direction and rises to depths between 170 m and 500 m below the sea surface. The bank profile is marked by a strong east-west asymmetry. The western part of the bank falls into the northern Porcupine Abyssal Plain with a steep continental slope down to depths greater than 4000 m. To the east of the central bank plateau at 53.5 N there is no clear separation from the Irish shelf break; the bank slope is comparatively shallow with water depths not exceeding 300 m. Three transects with almost identical station grids were surveyed across the Porcupine Bank area (see Figure 1(a)) by RV Valdivia (May 1994, July 1995) and RV Heincke (March 1994). At each station CTD casts were performed from the surface to near the bottom except for technical limitations. The CTD used in March 1994 allowed a maximum deployment depth of 1000 m, in July 1995 the maximum sampling depth was limited to 3000 m because of problems with the winch/ wire system. Additional water samples were collected at up to 12 discrete depth levels using a rosette system. The bottle samples were analyzed for salinity, which was used for the calibration of the CTD data. WOCE standard accuracy was achieved (Mohn, 1999). Continuous current measurements were performed during the RV Valdivia cruises using a ship-mounted Acoustic Doppler Current Profiler (ADCP) with an operating frequency of 153.6 kHz and a maximum depth of 480 m. The water velocity relative to the ship was recorded in depth intervals of 16 m during sampling intervals of 6 minutes. The May 1994 data were rejected due to malfunction of three of the four transducers. In July 1995 the data acquisition was split into two legs due to a 5-day gap in the acquisition procedure stemming from technical problems. The resulting data set was post-processed following the strategy and recommendations of the Common Oceanographic Data Access System (CODAS) of the University of Hawaii (Firing et al., 1995). The cruise track is shown in Figure 2(b). The ADCP data were corrected for bad GPS fixes, misalignment of the transducer with the ships keel and the effects of the Schuler-oscillation of the ships gyro (Mohn, 1999). The misalignment of the transducer with the ships keel leads to a bias of the Doppler shift determination and significantly limits the accuracy of absolute water velocities. The misalignment angle ( ) and the scaling factor (A) were estimated using the water track calibration method (Firing et al., 1995), where all changes of ship speed and course during the cruise were selected. The selection criteria are based on reference values for the course change and the acceleration phase of the ship. A total of 150 estimates were used for the calibration. The estimations were performed separately for measurements before and after the interruption of data acquisition. The r (...truncated)