1 - 15 April 2003





P.I.:   John P. Christensen             Bigelow Laboratory

                                                                        180 McKown Point

                                                                        W. Boothbay Harbor ME 04575


                                                                        jchristensen @


Co.P.I.  Humfrey Melling                Institute of Ocean Sciences

                                                                        P.O. Box 6000

                                                                        Sidney  BC   V8L 4B2  Canada


                                                                        mellingh @


SAMPLING.  Winter sampling was performed in the eastern area of the Shelf-Basin Interactions Project using aircraft.  Flights began on 1 April 2003 and finished on 15 April.  During this time, we were able to sample 42 sites (49 CTD casts) on a series of 5 transect lines which bracketed and were parallel to the SBI current meter mooring line located at 152oW longitude.  Stations were about 10 km apart along each transect line.  Transect lines B, C, and D were at the same spacing; while, lines A and E were 20 km from the nearest transect line.  The achieved sampling covered an area about 50 km across the slope and 80 km along it and represents a truly 3-dimensional view (Fig. 1).  Typically, 4-5 stations were sampled on each flying day.  At each site, a Seabird CTD and Satlantic ISUS nitrate analyzer with a water sampling bottle were deployed through an 8 inch hole augured in the pack-ice.  Continuous profiles of pressure, temperature, salinity, nitrate and bromide concentrations were made from the ice hole to either the sediment surface or to about 390 m.  Water samples (1.2 L total volume) were collected at key depths, as evaluated by the on-deck instrument readout.  Typically 5-8 water samples were taken on most casts.  Water was subdivided for samples of inorganic nutrients, organic N and P, bromide concentrations, salinity determinations, and O-18 isotopic content of the water. 


The ISUS nitrate analyzer measures nitrate concentrations optically at relatively high sampling rates (about 1 replicated measurement every 3 seconds).  As this is the first commercial instrument of its kind, we noticed variations possibly due to instrument drift (notice the possible scatter in individual nitrate readings in Fig. 2).  Also, we cannot at this time guarantee that the manufacturer's algorithms yield accurate nitrate concentrations; our own calibration will be conducted in the upcoming months and results will be correlated with autoanalyzer measurements.


RESULTS:  CTD and ISUS data are currently being processed and water samples await analysis.  However, the unprocessed results showed several important things.  Near the shelf break, intense mixing of shelf waters with slope waters were apparent by the occurrence of interleaving layers of waters of different temperature, salinity, and nitrate concentrations (Fig. 2).  The number of these interleaving layers and their water properties differed between the 5 transect lines suggesting that this mixing is associated with the mesoscale field in this area.  Secondly, in all transects, the data showed a slope boundary current centered at or near 150 m depth (Fig. 3c).  This current seemed to be associated with some of the highest nitrate concentrations seen in the area; although in this section, the maximum concentrations were found near the bottom of the current (Fig. 3c,d).  This association suggests this current may be extremely important in transporting and redistributing nutrients in the shelf-slope area.  Thirdly, in transect line C, a small eddy-like structure was seen.  Adjacent transects (i.e. line D, Fig. 3) showed little evidence of this small eddy.  From our sampling, its size would be about or less than 20 km in diameter and it did not seem connected to the slope current through meanders or other swirl-like distributions.  A similar small eddy was found in nearly the same location during the summer SBI mooring cruise.  Water properties of our eddy seemed to differ from shelf waters or the waters within the slope current.  Fourth, Atlantic water was found below about 200 m in all the deeper sites and distributions suggested upwelling along the slope and possible variations in depth across the sampled transects.  Graphs of temperature versus salinity from our cruise (Fig. 4) show the range in variability of the water masses relative to reference curves from sampling in the same area in 1981 and 2001.  The general freshening and the slight warming of the Fram Strait Branch of the Atlantic water in this region over the last two decades is clear.  These trends have accelerated noticeably since 1996 (viz. Melling JGR 103, C4, 7637-7645).  Water warmer than 0.7oC at this longitude has appeared only during the last year.  The recent change may indicate the arrival of the signal (much attenuated) that entered the Arctic through the Fram Strait more than a decade ago.


Overall, the entire data ensemble demonstrates the startling magnitude of variability in T-S correlation over the small (mesoscale) domain of our survey.  Near the peak in temperature, the variation over the study area encompasses the full range of change during the past twenty years.  Our three-dimensional survey resolves important scales of baroclinic and topographic change, but also encourages future caution in interpreting data acquired during broad surveys over the continental slope in the Arctic.


UPCOMING:  Over the next months, we will finalize the electronic data and analyze the returned water samples.  The resulting data will be posted in the SBI data repository.  We expect to have a strong set of data regarding the late-winter nutrient conditions in the outer shelf and slope of this area; this will be useful in predicting spring algal bloom intensities.  We will be looking at the influence of the mesoscale physical field in determining nutrient properties.  Our data, coupled with the results from the current meter mooring line, should allow estimates of the temporal and spatial variation in the shelf break region.  By examining the details of the water properties, it may be possible to evaluate possible water sources.


file:  srpt3fec.doc                                                                    submission date:  16 June 2003







Figure  1.  Map of  sampling locations on cruise 2003-14.  CTD cast numbers identify each site.















Figure  2.  Vertical profiles of temperature, salinity, density (sigma-t), and ISUS nitrate concentrations at CTD # 12 (left) and 13 (right).  The two stations are about 10 km apart on transect Line D.  Interleaving layers of different temperature and nitrate content dominate the profiles at CTD 12 but appear lacking in the upper waters at CTD 13.




Figure 3.  Sections of temperature in degrees (A top), salinity in practical salinity units (B middle), and density in kg/m-3 (C bottom) along transect line D on cruise 2003-14, 1-15 April 2003.  The darkened oval represents the likely core of upper slope current, flow toward the reader.






Figure 3D.  Section of nitrate concentration in micromoles/l based on raw ISUS profiles from Line D of cruise 2003-14.









Figure 4.  Temperature-salinity relationships from all data during cruise 2003-14 off the Beaufort continental shelf in April 1-15 2003.  Reference lines represent samplings taken during 1981 and 2001.  The peak temperature is the signature of the core of the Atlantic water entering from Fram Strait as found in the southern Beaufort Sea.  Note that the Atlantic water core has warmed and become less salty over the last 20 years.