Research Note The role of the West Spitsbergen Current in the formation of ice edge position ALEXANDER MAKSHTAS and IGOR PODGORNY Makshtas. A . P. & Podgorny, ]..A. 1991: The role of the West Spitsbcrgen Current in the formation of ice edge position. P d u r Research 9(2), 207-210. The intensity of sea-ice interaction in the Fram Strait is estimated from hydrographic data by use of three different methods. Preliminary results concerning fresh water formation rate are reported. A . P . M a k s h m and I . A . Podgorny, Arctic and Antarctic Research Insriture. 38 Bering St.. 199226 Sr. Petersburg. U.S.S. R. It is commonly accepted that advective heat fluxes in the atmos- phere and the hydrosphere along with sea ice transport from the Arctic Ocean control in general the spatial ice distribution in the Greenland Sea. The advection of warm Atlantic Water appears to be the main factor in the formation of the ice edge position and the upper layer of low salinity water in the north- east part of the Fram Strait, where the WesYSpitsbergen Current (WSC) intcracts very actively with the ice edge. The intensity of this interaction has been investigated by manyoceanographers. for example Quadfasel et al. (1987) who used temperature and salinity profiles from two oceano- graphical stations in the WSC (one at 77"30'N and one at 80"20'N). They estimated the influence of Atlantic Water on Fig. 1 . Location of all hydrographic stations used in the hori- zontal maps. The solid line indicates the ice-breaker drift during operation of the two-day ice station. The dotted line indicates the ice margin position (May 1989). the ice edge formation and found the difference in the heat content in the upper I00 m layer between the two profiles to be 5.03 x lo8 J/mL with an associated salt deficit of 3 g/cm'. This might have been caused by the daily retreat of the ice edge of about 5 km with a thickness of 1.5 m. More recently Unterstciner (1988) described the above pro- cesses on the basis of a steady heat-balance model in which the ice edge was interpreted as a wedgc with a thickness changing from zero on the ice margin to 2.5 m at 150 km to the north. The relevant heat flux from the upper ocean layer to melting ice was estimated as being about to 300 W/m'. Nevertheless, the data of Vinje & Finnekisa (1986) indicate considerablc ice Aoe thicknesses immediately near the ice edge as being up to 2 m. On thc basis of these data and our own observations i n the area under study, wc suggested the wedge description to be rather questionable, and we have therefore tried to estimate the role of the WSC in the ice edge and upper layer formation anew, paying particular attention to the localization of the intensive ice melting. For this purpose, during the R/v OTTO SCHMIDT cruise in May 1989 two oceanographical sections up and down the westcrn branch of the WSC were defined and a two-day ice station was established for collecting data. The study area and the station locations are shown in Fig. 1. Horizontal distributions of the main oceanographical charac- teristics in the upper 100 m in the castcrn part of the Fram Strait are shown in Fig. 2. One can see that T and S fields vary significantly from place to place. especially under the sea ice. The distribution of the heat content (Q) in layers & 3 0 m and 3&100 m is similar to that of T and S near thc surface. The main dccrease of Q takes place in the vicinity of the compact ice margin between 80"16'N and 80"03'N. This may he illus- trated with the aid of Table I . where "normalized" differences in Q ( A Q ) between several profiles of the first section (Fig. 2A) are determined. The origin of the horizontal coordinate x corresponds to XV16'N. Temperature and salinity profiles from two stations in the WSC (one at 80"16'N and one at XO"03'N) are plotted in Fig. 3. 208 Aiexnnder Mriksktns & lgnr Podgorriy The role of the West Spitsbergen Current in the formation of ice edge position 209 Tubk 1. "Normalized" differences A Q (J/mi) at the section along the WSC (see Fig 2) A z l A x G15.6 15 c 5 3 . 5 53.5-72.1 72.1-153 5 153 5-264.7 &30 - 370 -65 - 16 18 -2 3(!-50 -272 25 - 15 40 - 16 50-1Ci) - Y 72 - 43 -12 - 10 100-200 -57 53 -102 - 37 7 Note: A Q is the average heat content change of a cubic metre seawater p e r one metre distance. A z ( m ) characterizes the upper and lower boundaries of a layer and Ax (km) is the distance between two successful stations. S 34 T 0 1 I 3 I I 0 1 I 3 I I I '\ . . f i g 3 Vertical profiles of temperature ("C) and salinity ( % a ) at stations 8732 (80"16'N, 4"05'E) and 8731 (80"03'N. K W E ) The iiioht intensive interaction between the WSC and sea ice appears to be in the upper 5 0 m layer. where AQ has its maximum change. We suggest that variations of AQ in the 50- 200 m layer arc mainly linked to the uncertainty in determining Q , and. perhaps, to eddy structures (Johannessen et al. 1987). I t should be mentioned also that the heat transfer from the open ocean to the atmosphere which was determined from the ship's meteorological data using the algorithm proposed by Makshtas et al. (1989) was quite close to zero (= 40 W/m') during the period of observations and has not been taken into consideration. Fig. 2.Horizontal distributions of main oceanographical charac- teristics in the upper 100 m layer in the east area of the Fram Strait (May 1989): A. B - surface temperature ("C) and salinity (YOU); C. D - the same at 100111; E, F - heat content (J/m2 X loR) in layers 0-30 and 3&100 m; ice concentrations: 0.1-0.2 (--), 0.9-1.0 (-,-); (+) - station locations used in Table 1; ( 0 ) -stations 8731 and 8732. Let us examine the heat balance equation to evaluate the heat flux from the upper ocean layer upwards near the margin of sea ice: J T Jt pwC, - = -div F ( 1 ) where pn and C, are the density and the specific heat of sea- water and F is the heat flux in the WSC. Because of the weak temporal variability of the seawater temperature T near the ice margin, (1) may be rewritten as: where ii, denotes the northward average current velocity. We found the term dF,/dx due to turbulence to be small in comparison with the term JFJJz and integrated (2) through the layer of the most intensive ice-sea interaction with the depth h: J h U, lo p,c,T dz = F, - Fh. ( 3 ) 210 Alexander Makshtas & 1gor Podgorny So. using (3). taking U, = [ ) . I m/s ( A a g a a r d et al. 1985) and h = 50 m , and considering a heat Rux through the stable pycno- cline as very weak. h e can estimate F,, as 1600 W / m ' o n the hasis of Table 1 This heat flux is suggested to be generally localized between WO3' and Xff 16". where ice concentration changed from 11.5 to 1.0. Direct measurements to the north of this region, where the two-day ice station wa\ located (Fig. I ) . gave a n exper- imental estimate of the ice bottom melting rate of about 2 . 5 cm! day, corresponding to F,, being only u p to YO Wim'. T o prove our result we also used another approach. having examined the salt balance equation: where S, and S, arc the seawater a n d the sea ice salinities. and p, and h, the density and the spatially averaged thickness of sea ice. The method. being similar to that used by Quadfascl c t al. (19x7). has been used again with S, = 2 7 ~ . h = 50 m and salinity profilcs. as plotted in Fig. 3 . W e found the value of dh,!dt to he of 0 . 3 m/daq and the associated heat flux of 7(X) Wim' as no! being remarkably different from the ahovc estimate. I t \hould he mentioned that o u r estimate of dh,,'dt d o e s not differ very much from that reported by Josberger (1987). who carried out direct measurements of the ice melting rate in the marginal ice zonc and found i t to be u p to 0 5 m/day. O u r results sugge\t that the greater part of fresh water and heat fluxes in the northeastern area of the Fram Strait u c r e formed under thc narrow icc hclt with width Ax = I5 k m run- ning zonally along the compact ice margin. However. thcre is some uncertainty regarding this conclusion. It is possible that the oceanographical characteristics of the analysed stations simply corresponded t o the different water masses of the West Spits- hergen and the East Greenland Currents. To examine the a h w e results once again let us use the ice mass balance equation: dh dt h, U, = Ax ( - 5 ) Here U, denotes the x-component of the drift. I f the advcction of warm Atlantic Water from the south. determining thc valuc of dh,/dt. and the advection of sea ice from the north keep ice edge stationary. ( 5 ) is true Assuming h, = ? m . Ax = I 5 km. we get U, = 5.6cm;s and 3.5 cm/s corresponding to F,, = 1600 Wim' and 700 W!m! respectively These estimates of U, d o not differ very much from the value U, = 6.5 cm/s reported by Vinje & Finnekasa (1986) Taking the northward transport in the WSC a t 79"N in the upper 5 0 m layer as cqual t o 0 . 5 S v (Hanzlick 1983). and assuming that three quarters of this current continues t o Row to the north and the remaining goes round Spitsbergen. we found from ( 4 ) that the fresh water formation rate in the region of ourohservation is 1 2 0 krn'/year. Moreover. some fresh water appears to be formed to the north. but according to o u r data it is not possible to estimate the complete a m o u n t . According t o Ouadfawl et al. (1987) and Untersteiner ( I Y X X ) . an unknown part of this water is thought to recirculate and join the East Greenland Current Recently we invcstigated the interaction between the eastern branch of the WSC and the sea ice to the north of Spitsbergen. Direct measurements. carried o u t during the K/V OTTO SCHMIDT cruise in April 19'91 to the east of the ice margin in this region. gave an estimated ice melting rate of about Ycm/day. T h e maximum melting rate was not over the warmest A!lantic Water. hut rather ovcr the continental slope, being more than that reported above. W e suggest this fact to be possibly con- nected with the recent results by Padman & Dillon (1991) who investigated water mass modification n e a r the Yermak plateau and found i t to be linked t o current-topography interactions. A detailed discussion of such a mechanism is beyond the scope of this study a n d will be presented in o u r next paper. We would like to mention. however. that the reaion north of Spitshergen probably corresponds bctter with Untersteinder's (1988) "ice wcdge" description than with the northeastern area of the Fram Strait. I n o u r opinion, the latter may be associated with the existence of a mechanism that appears to neutralize the suppression of the heat flux from the ocean upwards despite the rapid ice melting and when a strong vertical mixing by surface waves (evidently the main mechanism of vertical heat transfer in the northeastern part of the Fram Strait) is absent. As is well-known the storage of fresh water in the Arctic Ocean represents o n e of the major uncertainties in under- \tanding hoth Arctic and global climate (Aagaard & Carmack 1989). Thus. the importance of the problem demands further observational and modelling efforts to improve o u r knowledge of air-ice-sea interactions in the Frain Strait. Ackrio~,lengernerif.\. - We would like t o acknowledge two anonymous reviewers for their valuable commcnts on a n early draft of this p a p e r . We a r e also grateful for the help provided by the officers a n d research scientists aboard the ice-breaker R V OTTO S C H M I I Y I References Aagaard. K . A , . Darnall, C . , Foldvik, A , , & Terrcscn. T. 1985: Fram Strait Current measurements. 198619x5, Joint data report. Bergen. Norway. 85 p p . Aagaard. K . A . & Carmack, E . C. 1989: The role of sea ice and other fresh water in the Arctic circulation. J . Geophys. Res. Y4 (C/O). 14485-14498. Hanzlick. D . J . 1983: T h e West Spitsbergen Current: transport, forcing and variability. P h . D . thesis. Univcrsity of Wash- ington. Seattle. Johannesscn. J . A , . Johannessen. 0. M . , Svendsen. E.. Shuch- man. R . . Manley. T . . Campbell. W . J . , Josherger. E. G . . Sandvcn. S . . Gascard, J . - C . , Olaussen. T.. Davidsoii. K. & Van Leer. J . 19x7: Mesoscale eddies in the Fram Strait marginal ice zone during the 1983 and 1984 Marginal Ice Zone Experiment. J . Geophys. Res 92 (C7). 67546772, Josberger. E . G . 1987: B o t t o m ablation a n d heat transfer coef- ficients from the 1983 marginal ice zone expcrinicnts. J . Geuph,vs. Res. 92 (C7), 7012-7016. Makshtas. A . P.. Tikmachev, V. F. & Ivanov. B . V. 1989: Energy exchange between the atmosphere and the underlying surface in the marginal ice zone. Workshop proceedings 'Regional and Mezoscale modelling of Ice Cover Oceans'. N R S C Conference R e p o r f N 3 , IX6190. Padman, L . & Dillon. T . M . 1991: Turbulent mixing near the Yermak plateau during the Coordinated Eastern Arctic Experiment. J . G e o p h y s . Res. Y6(Cll). 476F4782. K.-P. 1987: Large-scale oceanography in Fram Strait during the 1984 Marginal Ice Z o n e Experiment. J . Geophys. Res. 92 (C7). 67194728. Untersteiner. N. 1988: On the ice and heat balance in Frdm Strait. 1. G e o p h y s . R e s . 93 (Cl), 521-533. Vinje. T. & Finnekisa, 0. 1986: T h e ice transport through the Fram Strait. Norsk Polarinsf. Skrifrer 186. 39 pp. Ouadfasel, D. J . , Gascard. J.-C. & Koltermann