ISSN 00014370, Oceanology, 2014, Vol. 54, No. 1, pp. 106–112. © Pleiades Publishing, Inc., 2014. Original Russian Text © S.V. Gladyshev, A.V. Sokov, 2014, published in Okeanologiya, 2014, Vol. 54, No. 1, pp. 117–123.

INFORMATION

Underway Current Measurements in the Drake Passage S. V. Gladyshev and A. V. Sokov Shirshov Institute of Oceanology, Russian Academy of Sciences, Moscow email: [email protected] Received January 21, 2013; in final form, June 20, 2013

DOI: 10.1134/S0001437014010056

The Antarctic Circumpolar Current (ACC) is a powerful system of currents in the World Ocean that are the main part of the Southern Ocean circulation. Because of the lack of continental barriers, the ACC connects three oceans and has a huge impact on the global circulation and climate changes. This current system divides the warm subtropics and cold polar regions. The ACC is a global regulator and provides a dynamic link between the cells of the ocean conveyor belt. In the absence of continental barriers and the influence of the β effect, the ACC organizes itself into a system of intense jets that meander, generate synop tic eddies of opposite sign of rotation, and merge with each other, thus forming superjets [12]. Being particu larly intense, these processes occur in the area of rapid changes of the bottom relief [13]. One of such key areas from the standpoint of the formation of the ACC structure is the Drake Passage. Underway observations in the Drake Passage have been carried out by the Shirshov Institute of Oceanol ogy (IO RAS) starting from January 2010 for the pur poses of studying the ACC. The Drake Passage was chosen as an area of ACC monitoring for the following reasons. First, as already mentioned, the Drake Pas sage is the key region for the ACC formation. Second, two IO RAS ships—the R/V Akademik Ioffe and the R/V Akademik Sergei Vavilov—each year have repeat edly been crossing this strait from November to March. Third, this is the narrowest place in the South ern Ocean, where the current is bounded by the conti nental slopes of South America and Antarctica. Its width here is determined by the width of the strait, which is about 800 km. For example, the width of the ACC to the south of Africa increases up to 1500–2000 km, and its study in this area of the ocean is becoming more finan cially costly. Fourth, the Drake Passage is the most studied place in the Southern Ocean; therefore, new results of ACC monitoring obtained by the program in conjunction with the archive of historical data will of course greatly improve our understanding of the ACC

structure, its variability, and its impact on the earth’s climate changes. Regular observations in the Drake Passage have been carried out by the IO RAS since 2003. To date, the observations include seven quasimeridional CTD/LADCP sections that cross the strait from the surface to the bottom and from shore to shore with a distance between stations of about 10 miles. A detailed survey of the ACC Polar Current (PC) was carried out in the area of the Shackleton Fracture Zone (SFZ) in 2007 [2]. Since 2010, these sections have being accomplished together with shipboard ADCP mea surements. The main results obtained from the analy sis of complex observations with high spatial resolu tion are published in a series of articles [1, 3, 4, 5, 6], and these results can be considered as a starting point or a baseline on which the interpretation of the under way data will be based. The first underway current observations in the Drake Passage were started by US researchers in 1999 [9] and continue to be carried out at the present time [8]. For these purposes, the ship Laurence M. Gould (LMG) is used, which regularly (about once a month) supplies the Antarctic Palmer Station on the Antarctic Peninsula, as well as conducts scientific studies in this area. Prior to 2004, the observations of currents were carried in the upper 300m layer of the ocean (153.6 kHz, RD Instruments ADCP). In late 2004, the ship was installed with a new system of transducers (38 kHz, TRD Instruments ADCP), which measures the currents to depths of about 1000 m in the Drake Passage. Unlike that of the United States, the IO RAS program of underway observations is carried out only in summer season of the Southern Hemisphere but the intensity of our observations is significantly higher (up to 8.5 strait crossings a month vs. 2 crossings per month by the LMG). Increasing the observational fre quency can confidently allow for the synoptic time scales, and thus offer prospects for obtaining reliable estimates of the meridional transfer of mass, heat, and

106

UNDERWAY CURRENT MEASUREMENTS IN THE DRAKE PASSAGE

107

Number of Drake Passage crossings by IO RAS ships in 2010–2012 Name of ship

2010

2010–2011

2011–2012

In total

R/V Akademik Ioffe

10 (2)*

14 (1)

19 (5)

43 (8)

R/V Akademik Sergei Vavilov

–

15 (10)

19 (1)

34 (11)

In total

10

29

38

77

* The number of crossings with missing data is shown in parenthesis.

momentum by the eddies, which are considered as one of the main sources of the meridional transfer through the ACC. In addition, the IO RAS program of the underway observations spans over a more western region in the northern part of the strait—the area of Cape Horn, which is characterized by a sharp change in the direction of the isobaths of the continental slope. Since the ACC jets propagate to the bottom and, hence, are topographically controlled [11], there is an abrupt change of the northern jet direction when it simultaneously interacts with the SFZ. Thus, the data in this area will help us to investigate the mecha nisms that influence the behavior of the current as it interacts with the continental slope of complex config uration. The main objectives of the project, which are derived from its primary goal—the monitoring of the ACC—are the following: (1) Studying the ACC frontal structure and its vari ation in a wide range of scales. (2) Estimation of the ACC volume transport in the upper layer and its shortterm variability. (3) Estimation of the meridional flux through the ACC. (4) To establish connections between the fluctua tions of the wind field and the shortterm variability of the ACC volume transport. (5) Calculation of the tidal harmonic constants in the Drake Passage. (6) Estimation of the impact of the topography on the position (meandering) of the ACC jets in the Drake Passage. To monitor the ACC in Drake Passage by the IO RAS ships while moving at a speed of about 10 knots, the surface temperature and conductivity of seawater (SBE 21), as well as the speed and direction of the cur rents in the upper 30–1000m layer (TRDI OS 38 kHz OCEANOLOGY

Vol. 54

No. 1

2014

and 75 kHz ADCP), are measured. The thermosalino graph SBE21 runs at a frequency of 0.33 Hz, which at the speed of 10 knots is on average of one measure ment every 15 m of travel. The current measurements are carried out using acoustic Doppler current profil ers (ADCPs) mounted in special chests of the vertical shafts located in ships central parts at a depth of 5.8 m. The frequency of the ADCP measurements is 0.33– 0.5 Hz, and the collected data are averaged by ADCP software over 120 s and include, generally, 40 ensem bles. Thus, there is one averaged current profile for every 600 m of travel. The software simultaneously assimilates the navigation information (including the heading) transmitted through a highprecision two antenna GPS. Data collection begins (and stops) at depths of 50–100 m on the shelves of South America and Antarctica (and its islands). The total number of crossings in the Drake Passage obtained to date by the IO RAS ships is shown in table. The table also gives the number of crossings for each ship per year. The crossings with missing data is indi cated in parentheses. The total number of crossings collected in Drake Passage during three years is 77. The quality of the shipboard ADCP observations is highly dependent on the state of the ocean. During the rough weather, which is often in the Drake Passage at any time of the year, the quality of the received data can greatly decrease. The table shows that 19 of the 77 crossings (almost 25%) are incomplete or have gaps in the data. The IO RAS ship tracks in the Drake Passage are shown in Fig. 1. The OS38 kHz measures the currents to a depth of 1000–1200 m with a vertical resolution of 24 m. The first OS38 kHz record (the central point of the first averaged bin) occurs at a depth of 46 m. The OS75 kHz range is 800–900 m (vertical resolution—16 m, first measurement at 30 m depth). For calculation of

108

GLADYSHEV, SOKOV S –54

–55

–56

–57

Latitude, deg

–58

–59

–60

–61

–62

–63

–64

–65 –70

–68

–66

–64

–62 –60 Longitude, deg

–58

–56

–54 W

Fig. 1. Position of ship tracks in the Drake Passage in 2010–2012 where the data were collected. Depths shallower than 3500 m are shown in different shades of gray.

the ACC transport, the upper 100m layer is usually excluded for several reasons. First, the data of the first bin are usually very noisy, especially in the southern part of the strait. Second, the upper mixed layer, which in the Drake Passage has a thickness of about 100 m, has well developed inertial and drift currents. These

currents are impossible to extract and to remove, and they cause significant uncertainty in the calculation of the transport. The ADCP data processing includes cleaning the data and removing outliers, vertical and horizontal smoothing, and removal of the barotropic tide accord

Fig. 2. Data statistics (number of crossings per 0.2° × 0.5° or 22 × 28 km squares) in the Drake Passage for 2010–2012. The legend is shown in the lower left corner of the figure. The continuous thin line shows the 200m isobath. OCEANOLOGY

Vol. 54

No. 1

2014

UNDERWAY CURRENT MEASUREMENTS IN THE DRAKE PASSAGE

109

–54 S

–55

–56

–57

Latitude, deg

–58

–59

–60

–61

–62

–63

3–5 6–10 11–15 16–20 21–25

–64

–65 –70

OCEANOLOGY

–69

Vol. 54

–68

No. 1

–67

2014

–66

–65 –64 –63 Longitude, deg

–62

–61

–60

–59

–58 W

110

GLADYSHEV, SOKOV

ing to the global model TPXO 7.2 [7]. The analysis uses the data that percent good exceeds 70%. The sta tistical errors of 2minute velocity ensembles range from 2 to 3 cm/s [10], and they are considerably less than the maximum velocities of the ACC jets (50–80 cm/s) and, most importantly, are several times less than the stan dard deviation of the currents at any segment of the path 0.6 km in length (at which the averaging takes place). In order to obtain an average picture of data irreg ularly distributed in space and time, it is necessary to use statistical methods of analysis. For this purpose, the data of each velocity track were interpolated into grid nodes with 0.2° of latitude horizontally and 50 m vertically. Segments of the tracks with missing data were excluded. Next, a map for the data statistics for 0.2° × 0.5° squares of latitude and longitude with hor izontal scales on the order of 22 × 28 km, respectively, was constructed (shown in Fig. 2). Figure 2 shows the squares where three or more crossings were made by different gray shading. For the construction of the mean current field at different lev els in the Drake Passage, only those squares were used in which the number of crossings exceeded five. The velocity vector components were interpolated into the central points of these squares followed by averaging. As an example of such a construction, Fig. 3 shows an average map of the current vectors at a depth of 200 m. For convenience, the figure also includes contours of the absolute dynamic topography (ADT) of the sea surface that correspond to the axes of individual ACC jets [13]. Figure 3 shows the average position of the “axial” ACC (ADT) jet contours for the entire period of the observations in the austral summer of 2010– 2012. The joint analysis of the distribution of the velocity maxima in 100–500m layer, which were obtained by the program of underway observations, and the distribution of ADT gradients in the Drake Passage for the entire observational period from 1992 to 2012 as a function of the ADT allowed to identify the ACC jets in this field. The distribution of the ADT in the Drake Passage is uploaded from the site www.aviso.oceanobs.com. This product (ADT DTUPD) is a synthesis of the altimetry observations of the satellites TOPEX/Poseidon, ERS, GFO, JASON, and Envisat, which are interpolated to the

nodes of the Mercator projection map with a resolu tion of 1/3°. The system of currents in the Drake Passage is quite complicated, though largely coherent in the upper 500m layer. The south Southern Current jet (sSC) is located on the slope of the Antarctic Pen insula. Over the Phoenix Fracture Zone (PFZ), the sSC accelerates and has obvious flow disturbances. The north SC (nSC) jet is located 20–50 miles to the north of the sSC and, as can be seen from Fig. 3, is better expressed in the velocity field. The nSC jet tends to cross the PFZ through the southernmost deep passage. Over the Phoenix Plateau (PP), intense eddies between the SC jets and to the south of the southern Polar Current (sPC) jet, that are also regularly observed in this area on the weekly maps of ADT and have lifetime of a few months, are revealed. The sPC jet forms a powerful stationary meander over the PFZ, shifting to the north by almost 2° of lat itude. According to the map, this jet is accelerated to the west of the PFZ. The middle PC (mPC) jet (as well shown by the “axial” contour) also shifts to the north, forming a meander over the PFZ. It is a much more powerful current as compared to the sPC, which flows almost parallel to the latter over the PFZ. In this region, these two jets are located sufficiently close to each other and they are often very difficult to separate, even on individual crossings. The northern PT (nPT) jet moves closer to the southern jets over the PFZ and SFZ. This jet is also very close to the southern Suban tarctic Current (sSAC) jet, which seems to abruptly change direction after passing the SFZ. In Fig. 3, one can notice acceleration of these jets over the fracture zones. According to the data of individual crossings, four jets (three jets of the PC and sSAC) periodically form a “superjet,” in which the maximum velocities reach 0.8 m/s. The middle and northern jets of the SAC (mSAC and nSAC) also often merge into one jet. After passing the SFZ, they sharply turn to the northeast, following the bottom topography in this area. The sSAC extends over the Yagan mountains (YaM) from the north and the mSAC turns them from the south. After the SFZ, these jets are clearly separated from the southern jets. This result agrees with the conclusions of [8, 9]. In the

Fig. 3. Objectively interpolated mean current map at 200 m in the Drake Passage based on the data averaged within 0.2° × 0.5° squares with the number of crossings being at least 5. The solid lines show the “axial” contours of absolute dynamic topography of the ACC jets. Abbreviations are as follows: the sSC and nSC denote the northern and southern Southern Current jets; the sPC, mPC, and nPC are the southern, middle, and northern Polar Current jets; the sSAC, mSAC, and nSAC are the southern, middle, and northern Subantarctic Current jets; the SC is the Slope Current, YaM is the Yagan mountains, SFZ is the Shackleton Fracture Zone, PFZ is the Phoenix Fracture Zone, WSR is the Western Scotia Ridge, PP is the Phoenix Plateau, ATF is the archipelago of Tierra del Fuego, AP is the Antarctic Peninsula, and SSI are the South Shetland Islands. Depths shallower than 3500 m are shown in different shades of gray. OCEANOLOGY

Vol. 54

No. 1

2014

UNDERWAY CURRENT MEASUREMENTS IN THE DRAKE PASSAGE –54 S

111

Velocity scale m/s

ATF

0.5

0

–55 AC nS

YaM

–56

AC mS

SC –57 C sSA

–58 R WS

Z

Latitude, deg

SF

C nP

–59

–60

PFZ

PP sSC

–61 C mP sPC

–62

I SS

–63 nSC

AP

–64

–65 –70

OCEANOLOGY

–69

Vol. 54

–68

No. 1

–67

2014

–66

–65 –64 –63 Longitude, deg

–62

–61

–60

–59

–58 W

112

GLADYSHEV, SOKOV

area of the SFZ and PFZ, it is often very difficult to separate all the jets of the SAC and sPC even on indi vidual crossings because of their convergence into a sin gle “superjet” with a peak velocity of more than 0.7 m/s. In the area of 56° S and 66° W, the meander of the nSAC flows over the slope of Tierra del Fuego, feeding the Slope Current (SC), which in [3–5] is called the Cape Horn Current. Thus, the analysis of the Drake Passage current field, discovering some new circulation features, at the same time demonstrates current patterns that are con sistent with the previous studies provided by IO RAS and foreign scientific centers. ACKNOWLEDGMENTS This work was supported by the federal program World Ocean, the program of the Presidium of the Russian Academy of Sciences (project no. 23P), and the Russian Foundation for Basic Research (project no. 100500029).

5.

6.

7.

8.

9.

REFERENCES 1. S. V. Gladyshev, M. N. Koshlyakov, and R. Yu. Tara kanov, “Currents in the Drake Passage based on obser vations in 2007,” Oceanology (Engl. Transl.) 48 (6), 759–770 (2008). 2. M. N. Koshlyakov, S. V. Gladyshev, R. Yu. Tarakanov, and N. I. Ryzhikov, “Deep currents in the central part of the Drake Passage based on the data of the 2008 hydrographic survey,” Oceanology (Engl. Transl.) 50 (6), 821–828 (2010). 3. M. N. Koshlyakov, S. V. Gladyshev, R. Yu. Tarakanov, and D. A. Fedorov, “Currents in the Western Drake Passage according to the observations in January of 2010,” Oceanology (Engl. Transl.) 51 (2), 187–198 (2011). 4. M. N. Koshlyakov, S. V. Gladyshev, R. Yu. Tarakanov, and D. A. Fedorov, “Currents in the Drake Passage

10. 11.

12.

13.

according to the observations in November of 2010,” Oceanology (Engl. Transl.) 52 (3), 325–334 (2012). M. N. Koshlyakov, S. V. Gladyshev, R. Yu. Tarakanov, and D. A. Fedorov, “Currents in the Drake Passage based on the observations in October–November of 2011,” Oceanology (Engl. Transl.) 53 (1), 299–308 (2013). M. N. Koshlyakov, I. I. Lisina, E. G. Morozov, and R. Yu. Tarakanov, “Absolute geostrophic currents in the Drake Passage based on observations in 2003 and 2005,” Oceanology (Engl. Transl.) 47 (4), 451–463 (2007). G. D. Egbert, A. F. Bennett, and M. Foreman, “TOPEX/Poseidon tides estimates using a global inverse model,” J. Geophys. Res., C: Oceans Atmos. 99 (12), 24821–24852 (1994). Y. L. Firing, T. K. Chereskin, and M. R. Mazloff, “Ver tical structure and transport of the Antarctic Circum polar Current in Drake Passage from direct velocity observations,” J. Geophys. Res., C: Oceans Atmos. 116, C08015 (2011). doi 10.1029/2011JC006999. Y. D. Lenn, T. K. Chereskin, J. Sprintall, and E. Fir ing, “Mean jets, mesoscale variability and eddy momentum fluxes in the surface layer of the Antarctic Circumpolar Current in Drake Passage,” J. Mar. Res. 65 (1), 27–58 (2007). “Ocean surveyor installation guide,” in Teledyne RD Instruments, 2007. P/N 95A601900. S. R. Rintoul, C. W. Hughes, and D. Olbers, “The Ant arctic circumpolar system,” in Ocean Circulation and Climate, Ed. by G. Siedler, J. Church, and J. Gould (Academic, San Diego, CA, 2001), pp. 271–302. S. Sokolov and S. R. Rintoul, “Multiple jets of the Ant arctic Circumpolar Current south of Australia,” J. Phys. Oceanogr. 37, 1394–1412 (2007). doi 10.1175/JPO3111.1. S. Sokolov and S. R. Rintoul, “Circulation structure and distribution of the Antarctic Circumpolar Current fronts: 1. Mean circumpolar paths,” J. Geophys. Res., C: 114, 11019 (2009). doi Oceans Atmos. 10.1029/2008JC005108. Translated by K. Lazarev

OCEANOLOGY

Vol. 54

No. 1

2014