One-year records of current and bottom pressure in the strait between Nordaustlandet and Kvitoya, Svalbard, 1980-81 K . AAGAARD. A . FOLDVIK, T. GAMMELSRBD AND T. VINJE Aagaard, K . , Foldvik, A , . G a m m e l s r ~ d , T . & Vinje, T. 1983: One-year records of current and bottom pressure in the strait between Nordaustlandet and K v i t ~ y a . Svalbard, 198&81. Polar Research I n . s . . 107-113. We have obtained one year of measurements from a subsurface instrumented mooring carrying two current meters and one bottom pressure recorder in the strait between Nordaustlandet and Kvitcbya in the northeastern Svalbard archipelago. The observations show a mixed tide with typical amplitudes 0.4 d b and 1Ocm sec-I. The semidiurnal tide is characterized by a progressive wave propagating toward the south. together with a cross-channel baroclinic mode. The annual average (non-tidal) current is less than 2 cm secc’ toward the north-east, suggesting that the transport into the Arctic Ocean is approximately 0.2 x lo6 m3 SKI. K . Aagaard, University of Washington, 4057 Rooseuelt Way, N. E. Seattle, Washington 98105. U . S . A . ; A . Folduik and T . Gammelsr@d, Geofysisk Institutt, Aud. A , Uniuersitetet i Bergen, 5014 Bergen. Norway: T . Vinje, Norsk Polarinstirurt, Rolfsrangueien 12, 1330 Oslo Lufthaun, Norway; December 1982 (reorsed February 1983). Introduction In this paper we present the results from a sub- surface mooring between Kvitoya and Nordaust- landet in the northeastern Svalbard archipelago (Fig. 1). T h e rig was launched from the Swedish icebreaker YMER on 30 July 1980 and recovered from t h e Norwegian vessel L A N C E on 25 August 1981. I t consisted of two Aanderaa RCM-4 cur- rent meters at 75 m and 255 m, and o n e Aanderaa Fig. I . Map of Svalbard showing positions of mooring and CTD section 108 K . Aagaard et al. . . F 3 r . . . TG-2A pressure recorder at the bottom (260 m ) . T h e sampling interval was o n e hour. An apparent deterioration in the speed recording of the lower current meter over the last 20 weeks of deploy- ment caused the only data gap. Pressure measurements The bottom pressure record reflects the combined effect of variations in sea level ( e . g . that due to tides). varying atmospheric pressure. and baro- clinic currents. Fig. 2 shows the pressure record- ing for the two months immediately following the launching. A typical mixed tide prevailed. a i t h the semidiurnal components dominating. The fortnightly period is due to the interference of the semidiurnal periods M: and S:. T h e maximum amplitude is seen to be roughly 0 . 4 decibar. equiv- alent to 0 . 4 m in sea level. Also presented in Fig. 2 is the residual pressure . ~ ._ . . . - . ....._I__ .- 1 . - 1 1 5 : , - e , . o , i P r s ; F i g . .? T h e p o u c r spectrum for the entire one-year hortom pressure record. cumputcd using the maximum entrop) method. The lrequencies of the malor tidal constituents are also ahti\\n. .. . ._ . Fig. 2 . Hourly bottom pressure from August t o September 1980. Time in Julian days. Also shown is the low-passed series, using a filter with a 40 hr cut- off. after tidal variations have been filtered out using a low-pass filter which suppresses variations with periods shorter than 40 hours. T h e amplitude of the filtered curve is equivalent t o about 10 cm and represents the combined effects of low-frequency currents and atmospheric pressure variations. T h e power spectrum of the pressure record (Fig. 3) yields the expected tidal bands. T h e large subharmonic components which appear near 8 h r , 6 hr. etc.. are typical for shallow water tides, and arise from the non-harmonic tidal wave form in shallow water. T h e amplitude and phase of the major tidal constituents have been computed from the entire one-year record using Foreman’s (1977. 1978) method. T h e results a r e listed in Table 1 . Current and temperature measurements Fig. 4 shows a temperature, salinity, and density section across the strait two days after the recov- ery of the mooring. T h e lower current meter had been situated near the bottom in the warmest water, while the upper meter was a t the position of the cold core at 75 m d e p t h . Portions of t h e various temperature and veloc- ity records are shown in Fig. 5 . covering the same period as the pressure record in Fig. 2 . As with the pressure signal, the tides also dominate the currents. T h e amplitude and phase of the major tidal constituents have been computed using the full-length records, and the results are listed in Table 1. T h e V (north)-component of the upper current meter exhibits the largest amplitude, about 10 cm sec-’. Power spectra of the current components (not shown) are similar to the power spectrum for the pressure record (Fig. 3 ) . T h e temperature spectra (Fig. 6) show no tidal Current and bottom pressure records 109 Table 1. Major tidal constituents for the upper current meter (upper numbers). the lower current meter (lower numbers) and the pressure. Positive minor axis denotes counter-clockwise rotation. The angle of inclination denotes the orientation of the major axis and is measured in degrees counter-clockwise from east. Greenwich phase is referred to the northcrn major semiaxis for the current and to high water for the pressure. ~ ~~~ Current measurements Pressure measurements Major Minor Angle of Greenwich Greenuich axis axis inclination phase Amplitude phase Constituent cm 7ec-l cm sec-' degree5 degrees dectbars degrees 0 1 0.45 -0.18 147 _ _ 34 0.25 -0.01 158 26 0.01 320 PI 0.47 -0.00 95 0.24 0.04 109 1.29 0.88 - 0.09 104 0.16 117 68 67 67 68 0.02 0.07 '95 305 0 04 68 N2 1 6 5 -0 15 x9 230 0 70 0 39 104 237 265 3.61 2.37 102 267 337 1.39 0.73 39 268 0.21 47 M2 8.63 -2.18 96 0.06 152 s2 3.50 -1.32 105 influence other than a rather broad semidiurnal pressure, current and temperature records: the peak, probably due t o a semidiurnal internal tide results are shown in Fig. 7 . The filtered currents (see discussion below). T h e different spectral lev- are weak, about 2 cm sec-', and exceed els for the two instruments reflect the weaker 5 c m s e c - ' only for short periods of time. The temperature strat:clcation in the lower layer (cf. average flow is approximately northeast, i.e. into Fig. 4). the Arctic Ocean, as is also apparent from the We have applied the 40-hr filter t o the entire progressive vector diagrams (Fig. 8). 0' Sd D I S T . I N KH: TERPERRTURE I N O E G . C 0 25 5 0 0 SO 0157. 1N K f l : 0187. I N Kn: P R R C T I C R L S R L I N I T Y S I G V R - T H E T R . Fig. 4 . A temperature. salinity and density section across the strait betwecn Nordauatlandet and Kvitbya. The upper numbers identify the oceanographic stations. The deepest observations are marked with crosses and the positions of the current meters with black dots. 110 K . Aagaard et a l . The average velocities in the upper and lower layers are 1.8 cm sec-' and 0.5 cm sec-', respec- tively (the latter value was calculated u p until 4 April only). From this we estimate that the annual average flow through the strait probably does not exceed about 0.2 x 10bm3 sec-' (0.2 Sv) toward the northeast. In terms of mass balance for the Arctic O c e a n , this is a small contribution (SCOR Working G r o u p 58,1979); for example, it is prob- ably less than 5 % of t h e West Spitsbergen Current transport. T h e temperature records (Fig. 7) show that the structure revealed in Fig. 4, with a cold core above a warm bottom layer, is maintained throughout the year. Current and bottom pressure records 111 Discussion Neither the filtered series themselves (Fig. 7) nor their power spectra (not shown) indicate distinct long-term periodicities. There are, however, a few events of strong northerly currents at both instruments (e.g. 28 January and 23 February, cf. Fig. 7) which coincide with anomalously high bottom pressure. On both occasions, relatively intense cyclones passed over the northern Barents Sea, and the strong current events were probably forced by the atmospheric events. 20 10 8 7 6 5 1 P P e r , o d ( H l r ) The rotary spectra for the current records are shown in Fig. 9. A comparison of the energy Fig. 6 . Power spectra of temperature. -2.w I I 112 K . Aagaard et al. 04 P ~ . . . . , . . . , , . , . . , . . . . , . , , 1 L I00 0. ICO. ZDO . 3w. 400. D I S P L A C E M E N T E A S T W A R D S ' K r r 1 Fig. 8. Progressive lector diagram for the two current meters. Circles (uppcr meter 1 and crosses (lower meter) denote position every 30 days levels in the tidal bands shows that the counter- clockwise components are equal in the upper and lower layers. whereas the energy in the clockwise component is smaller at the lower instrument. particularly for the semidiurnal tide. T h e distri- bution of energy in the tidal bands also shows that the associated tidal currents rotate in opposite directions at the two instruments. clock- wise in the upper layer and counter-clockwise in the lower. This is clearly seen in Fig. 10. where the tidal current ellipses are plotted; the current vectors are shown at the time of high tide. The maximum southward flow in both layers is seen to coincide with high tide. Thus the north-south tidal currents have the characteristics of south- ward-propagating progressive waves. A quarter period after high tide the maximum westward flow in the upper layer coincides with maximum eastward flow- in the lower layer. T h e motion across the axis of the channel thus has the charac- teristics of a standing internal wave. A c k n o k l e d g e r n e n r , . - The measurements described in this 95 P E r j i F ' I l ~- ~ ~ . -.--I L 1 _ ~ l , , ~-~ --I T T ~ - - L ..~. - . -~~ , ~. .C? . ICR . -)1 'i' i t i i - l - t N C " : CPH Fig. Y. Rotary current spectra. The frequencies of the major tidal constituents are indicated report were the combined efforts of the Norwegian Polar Research Institute. the Geophysical Institute of the University of Bergen. and the University of Washington. We express our appreciation to the officers and crew of the YMER and the LANCE for their enthusiastic cooperation in carrying out the deployment and dredging operations: to Clark Darnall and Richard Tripp at the University of Washington for help with instrumentation and data handling; and to Tor Torresen and Stein Sandven at the Geophysical Institute. University of Ber- gen. for programming and helpful discussions. Knut Aagaard was supported financially by the Office of Naval Research through contract No. N00014-75-C-0893 and by the National Science Foundation through grant No. DPP 81-00153. References Foreman. M. G . G. 1977: Manual for Tidal Heights Analysis and Prediction. Pat@ Marine Science Report 77-10 (Corr. 1979). Institute of Ocean Sciences. Patricia Bay, Sidney. B.C. Foreman. M. G. G . 1978: Manual for Tidal Currents. Analysis and Prediction. Pacific Marine Science Reporr 78-6 (Corr. 1979). lnstitute of Ocean Sciences, Patricia Bay, Sidney, B.C. SCOR Working Group 58, 1979: The Arctic Ocean Heat Budget. Reporr Y o . 52. Geophysical Institute, University of Bergen. Norway. Current and bottom pressure records 113 i is indicated and the arrows show Fig. 10. Tidal ellipses for the major constituents according to Table 1. The shaded ellipses refer to the lower current meter. The direction of rotation the tidal current at the time of high water 2 w+E S Ocrn/s 1 2 - ‘L L 8 8 i upper inst 4888 l o w e r inst I t o n e d 1