The CTD/rosette sampler on its way back from the water. Photo: Sofia Rickberg.

The CTD/rosette sampler on its way back from the water. Photo: Sofia Rickberg.

The oceanographic component of LOMROG focuses on constraining and understanding the pathways of the Atlantic water and deep water across the Lomonosov Ridge, between the Eurasian Basin and Canadian Basin and through the western Fram Strait. A deep water pathway may exist between the Lomonosov Ridge and the Northern Greenland shelf, and thus knowledge of the shape of the seafloor in this area is also of interest for the oceanography programme. The LOMROG oceanography programme is an extension of the 2005 scientific programme conducted during the Healy-Oden Trans-Arctic expedition (HOTRAX) in 2005 when oceanographic station work was carried out from the icebreaker Oden and multibeam mapping from USCGC Healy. The data from HOTRAX showed that water overflow from the Makarov Basin (part of the Canadian Basin) to the Amundsen Basin (part of the Eurasian Basin) takes place across a 1 870 m deep sill in the central Lomonosov Ridge at about 88°25’N, 150°E (Björk et al., 2007). This water appears to follow the Lomonosov Ridge slope southwards towards Greenland but uncertainty remains in regard of whether it can be traced on the southernmost tip of the ridge and how the circulation continues from this point and onwards.

Data acquisition

Conductivity Temperature Depth (CTD) and water sampling were carried out at 26 stations covering parts of the western Eurasian Basin and one station at the East Greenland continental margin. Water was collected using a 24-bottle rosette sampler equipped with 7.5 l Niskin type bottles and a CTD (Sea Bird 911+) (picture 1). When brought back onboard the rosette was moved into a heated double container as quickly as possible to avoid freezing the samples. Water samples were immediately drawn for the individual parameters to be determined in the following order: CFCs, oxygen, dissolved inorganic carbon/pH/total alkalinity, nutrients, oxygen-18, and salinity. Number of samples analysed for the different constituents are given in Table 1.LOMROG-07-001

Shipboard processing

The CTD data were processed through the standard Sea-Bird software routines (data conversion, cell thermal mass, filter, loop edit, derive and bin average). The final data is averaged in 1 dbar bins. Salinities were also determined in each Niskin bottle using an Autosal lab-salinometer. The temperature in the clean room inside the laboratory on the foredeck fluctuated significantly and therefore conditions were not ideal for the salinity analyses, but were nevertheless manageable. The final bottle salinity data should be of good quality with an accuracy of ±0.001 psu. The bottle salinities were then compared with the CTD bottle file data in order to check the accuracy of the CTD system. The comparison with bottle data showed an offset of about -0.0044 psu between the CTD and Autosal salinities, inferring that the CTD sensor values were too low. The final CTD data were corrected by determining an average salinity offset using bottles with salinity > 34.8 psu and excluding outliers outside 1 SDA. One sample was then identified (Station 20, 4 000 dbar) as having identical offset as the average offset. The Autosal conductivity for this sample was determined for the same temperature and pressure when the bottle was tripped (P=4 000 dbar, T=-0.6708°C). The Autosal conductivity divided by the CTD conductivity then gives a slope correction of 1.000119, which has been used to post-process the data according to Sea-Bird recommendations.

Sara Jutterström collecting water for the determination of DIC. Photo: Leif Anderson.

Sara Jutterström collecting water for the determination of DIC. Photo: Leif Anderson.

Water for chemical analysis was drawn from water samplers directly after the rosette was brought on board and analysed within hours of sampling. The precisions given below were computed as standard deviations of duplicate analyses. Samples for CFCs were drawn from the bottles on the rosette with glass syringes, which were kept under cold water until analysis (within a few hours). They were measured by purge-and-trap extraction and pre-concentration, gas chromatographic separation on a capillary column, and electron capture detection calibrated against a standard gas mixture. The precision was in the order of 1% and the accuracy was about 0.02 pmol/kg. Oxygen was determined using automatic Winkler titration system, precision ~1 μmol/kg. Total dissolved inorganic carbon (DIC) was determined by a coulometric titration method having a precision of ~1 μmol/kg, with the accuracy set by calibration against certified reference materials (CRM), supplied by A. Dickson, Scripps Institution of Oceanography (USA). Total alkalinity (TA) was determined by potentiometric titration, precision ~1 μmol/kg, (Haraldsson et al., 1997), with the accuracy set in the same way as for DIC. The determination of pH was performed by the use of a diode-array spectrophotometer using a sulphonephtalein dye, m-cresol purple, as indicator (Clayton and Byrne 1993, Lee and Millero 1995), and measured in a 1 cm flowcell thermostated to 15°C (pH15). The precision and accuracy for the pH15 measurements were ±0.0005 and ±0.002 pH units, respectively. pH in situ was calculated from TA, pH15 and in situ temperature by using the CO2-system program by Lewis and Wallace (1998). For these calculations the carbon dioxide constants of Roy et al. (1993 and 1994) were applied, and the pH was on the total hydrogen ion scale. The nutrients (phosphate, nitrate, and silicate) were determined using a SMARTCHEM autoanalyser applying standard analytical protocol giving a precision near 1% at full scale.

Preliminary results

The observations at the Amundsen Basin slope of the Lomonosov Ridge (Stations 4, 9 and 10) show a clear signal of Canadian Basin Deep Water (CBDW) at around 2 000 m with similar characteristics to a station (Station 41) in the vicinity of the 1 870 m deep channel in the central Lomonosov Ridge at about 88°25’N, 150°E found during the Beringia/HOTRAX 2005 expedition (Björk et al., 2007) (see also figure 1 for some selected station data). A vertically broad CBDW signal was also clearly visible at the flanks of the Morris Jesup Plateau (Stations 12, 17, 18 and 19). A more vertically narrow and weaker signal was observed over the western Amundsen Basin (Stations 3 and 21–25). The observations during LOMROG infer that the major inflow of CBDW to the Amundsen Basin indeed occurs at the central Lomonosov Ridge channel and that the flow continues along the slope of the ridge towards Greenland. The further circulation follows the Greenland continental slope towards the southeast to the Morris Jesup Plateau where eventually some of the flow leaves the continental slope, becomes interleaved, and can be identified as a weak salinity maximum  over the central Amundsen Basin.

Temperature and salinity data showing the structure of the CBDW signal seen as a temperature and salinity maximum at around 2 000 m depth. The CBDW water is also less ventilated (“older”) which is seen in the oxygen and CFC-11 signals. The Beringia/HOTRAX station 41 is from the central Lomonosov Ridge, LOMROG Station 4 is further towards Greenland at the Amundsen Basin flank of the ridge, and LOMROG Station 3 is from the central Amundsen Basin.

Temperature and salinity data showing the structure of the CBDW signal seen as a temperature and salinity maximum at around 2 000 m depth. The CBDW water is also less ventilated (“older”) which is seen in the oxygen and CFC-11 signals. The Beringia/HOTRAX station 41 is from the central Lomonosov Ridge, LOMROG Station 4 is further towards Greenland at the Amundsen Basin flank of the ridge, and LOMROG Station 3 is from the central Amundsen Basin.