monitoring volume transport through measurement of cable

APRIL 2004 671K I M E T A L .

q 2004 American Meteorological Society

Monitoring Volume Transport through Measurement of Cable Voltage across theKorea Strait*

KUH KIM AND SANG JIN LYU

OCEAN Laboratory/RIO, School of Earth and Environmental Sciences, Seoul National University, Seoul, Korea

YOUNG-GYU KIM

Naval Systems R&D Center, Agency for Defense Development, Chinhae, Korea

BYUNG HO CHOI

Department of Civil Engineering, Sung Kyun Kwan University, Suwon, Korea

KEISUKE TAIRA

Ocean Research Institute, University of Tokyo, Tokyo, Japan

HENRY T. PERKINS, WILLIAM J. TEAGUE, AND JEFFREY W. BOOK

Naval Research Laboratory, Stennis Space Center, Mississippi

(Manuscript received 16 August 2002, in final form 5 August 2003)

ABSTRACT

Voltage induced by the Tsushima Current on an abandoned submarine telephone cable between Pusan, Korea,and Hamada, Japan, has been measured since March 1998 in order to monitor the volume transport through theKorea Strait. Voltage has a good linear relationship with the transport measured by bottom-mounted acousticDoppler current profilers (ADCPs) along a section spanning the Korea Strait. The linear conversion factor isestimated to be L0 5 (8.06 6 0.63) 3 106 m3 s21 V21 with the reference voltage of V0 5 0.48 6 0.07 V. Thevoltage-derived transport reveals various temporal variations that have not been known previously. Measurementof the cable voltage provides a reliable means for continuous monitoring of the volume transport of the TsushimaCurrent, which determines the major surface circulation and hydrography in the East Sea.

1. Introduction

The East Sea (Sea of Japan) is a semienclosed mar-ginal sea connected to the Pacific Ocean through fourstraits shallower than 200 m, as shown in Fig. 1. Thehorizontal length scale of the East Sea is about 1000km and its mean depth is about 1700 m with a maximumdepth of 4000 m. The Tsushima Current, which flowsthrough the Korea Strait (also known as the Korea/Tsushima Strait), is a major contributor to the circulationof the East Sea and to the transport of heat and saltfrom the Pacific Ocean to the East Sea (Moriyasu 1972).

* OCEAN Laboratory Contribution Number 17.

Corresponding author address: Kuh Kim, OCEAN Lab./RIO,School of Earth and Environmental Sciences, Seoul National Uni-versity, Seoul 151-742, Korea.E-mail: [email protected]

Numerical experiments show that the variability of thecirculation in the East Sea depends largely on the trans-port variation through the Korea Strait (Holloway et al.1995; Kim and Yoon 1999). Therefore, continuous mon-itoring of the transport in the Korea Strait is essentialto understand the circulation of the East Sea.

Estimates of the transport through the Korea Straitvary widely depending upon methods. Ranges for theannual mean are 0.5 ; 4.2 3 106 m3 s21 with a seasonalvariation of 0.7 ; 4.6 3 106 m3 s21 (Table 1). Yi (1966)estimated the transport through the Korea Strait usinga dynamical calculation with a reference level at thebottom, which cannot give a good estimate because ofthe shallow bottom. The transport was also calculatedfrom the sea level difference across the strait (Mizunoet al. 1989) based on an assumption that the current isvertically barotropic. However, it was reported later thatthe baroclinic effect on the sea level difference is sig-nificant in the Korea Strait (Isobe 1994; Lyu and Kim

Unauthenticated | Downloaded 11/09/21 08:38 AM UTC

672 VOLUME 21J O U R N A L O F A T M O S P H E R I C A N D O C E A N I C T E C H N O L O G Y

FIG. 1. Arrangement of the submarine cable between Pusan, Korea, and Hamada, Japan, bottom-mounted ADCP mooring stations, vessel-mounted ADCP track, and geomagnetic stations. (right) Bold arrows denote surface currents in the East Sea suggested by Uda (1934). Lightlines are depths in m.

TABLE 1. Previous estimates of the volume transport through the Korea Strait. (10 6 m3 s21)

StudySummer and

autumnWinter and

spring Mean transport Method

Yi (1966)Byun and Seung (1984)Miita and Ogawa (1984)Tawara et al. (1984)

2.210.83

4.1

0.330.16

2.2

1.350.54.2

Dynamic calculationDynamic calculationCurrent meterCurrent meter

Mizuno et al. (1989)Isobe et al. (1994)Katoh et al. (1996)Takikawa et al. (1999)Teague et al. (2002)

5.62.33.43.5

0.81.0

1.61.7

2.3

2.62.7

Sea level difference (East Channel)Towing ADCPTowing ADCPVessel-mounted ADCPBottom-mounted ADCP

2003). Miita and Ogawa (1984) used 431 current ob-servations lasting at least one day. Isobe et al. (1994)and Katoh et al. (1996) estimated the transport usingthe towed ADCP. As the methods provide instantaneouscurrent fields, it is difficult to remove large short-termvariations, such as tidal currents. Long-term mooringsof current meters are very few because of high fishingactivities in the Korea Strait.

Since February 1997 currents in the Korea Strait have

been measured by a vessel-mounted ADCP along a trackbetween Pusan, Korea, and Hakata, Japan (Fig. 1), sixtimes per week (Takikawa et al. 1999). The U.S. NavalResearch Laboratory (NRL) also deployed 12 trawl-resistant bottom-mounted (TRBM) ADCPs along twolines in the Korea Strait (Fig. 1) for the time period ofMay 1999 to March 2000 (Perkins et al. 2000). Theaverage spacing between ADCPs is 25 km and theirsampling interval is generally 30 min. However, it is



FIG. 2. Cable voltage measurement system at the Pusan SubmarineCable Relay Station and the vertical section across the cable groundsites, Pusan and Hamada.

difficult to maintain these measurements continuouslybecause of economic considerations. Thus the charac-teristics of temporal variations of the transport in theKorea Strait have been poorly understood until now.

According to Faraday’s law, when seawater possess-ing electric conductivity flows through a strait under thegeomagnetic field, an electric potential difference is in-duced that is related to the volume transport through thestrait (Sanford and Flick 1975). Longuet-Higgins (1949)originally introduced this possibility and Bowden(1956) applied it to the Dover Strait. Larsen (1992) andBaringer and Larsen (2001) have successfully monitoredthe volume transport through the Florida Strait usingthis technique.

The same technique was applied to the Korea Straitusing an in-service submarine telephone cable betweenPusan, Korea, and Hamada, Japan (Fig. 1). Kawatate etal. (1991) found the energy peaks at tidal frequenciesand high coherencies between the voltage and currentmeter data for the period from 1987 to 1996. Choi etal. (1992) and Choi et al. (1997) showed also high spec-tral energies at low and tidal frequencies, and estimatedthe conversion factor for voltage to transport as 30 ;35 3 106 m3 s21 V21 by comparing the amplitude ofthe M2 signal in the measured voltage at Pusan withthat of the M2 tidal transport from a tidal model. Re-cently Lyu et al. (2002b) analyzed the cable voltagemeasured simultaneously at Hamada, Japan, and Pusan,Korea, using the in-service telephone cable in 1990.Comparing the voltage difference in 1990 with that mea-sured using the now abandoned cable in 1998, theyshowed that both datasets have approximately the sameamplitude and phase for M2 and O1. They concludedthat the relationship between the voltage and the volumetransport through the Korea Strait can be consideredrobust and stable over time.

As a new optical fiber cable began service for com-munication between Korea and Japan in July 1997, theabandoned telephone cable was given to the ResearchInstitute of Oceanography, Seoul National University,and to the Ocean Research Institute, the University ofTokyo from Korea Telecom and Kokusai Denshin Den-wa for a cooperative research program of NorthEastAsian Region Global Ocean Observing System. Since1 March 1998 the voltage has been measured at thePusan Submarine Cable Relay Station, which is in factthe potential difference across the Korea Strait as thecable was grounded at Hamada and Pusan using thecopper electrodes (Fig. 2).

The voltage measurement system and data are de-scribed in section 2. Geomagnetic correction and errorestimates of voltage are discussed in section 3 and theconversion of voltage to transport using observed trans-ports is described in section 4. The voltage-derivedtransports are discussed for different timescales andcompared with the observed transport in the Korea Straitin section 5. A summary is provided in section 6.

2. Voltage measurement system and data

The voltage measurement system consists of a digitalvolt meter (Keithley DMM 2000), a personal computeras a data logger, a GPS (Global Positioning System),and a modem (Fig. 2). The computer clock is controlledby the GPS time. The sampling interval is 5 min andthe data are automatically transferred once per day tothe OCEAN Laboratory at Seoul National University,Seoul, Korea via modem.

Observed voltage is averaged hourly for analyses herefor the period from March 1998 to August 2001. TheUS NRL deployed 12 TRBM ADCPs (Fig. 1) from May1999 through March 2000 along two lines in the KoreaStrait (Perkins et al. 2000). The deployment was dividedinto two periods: the first half from May to October1999 and the second half from October 1999 to March2000. Transports are estimated from these observed cur-rent data (Jacobs et al. 2001; Teague et al. 2002). First,eight tidal constituents are removed from each velocitycomponent by harmonic analysis. Each velocity com-ponent is low-pass filtered with a cutoff period of 40 h.Transport values are then obtained by integrating thecurrents normal to the mooring section after optimalspatial interpolation of observed currents at 3-h inter-vals. Monthly mean transports from April to November1998, calculated using data from a vessel-mountedADCP (Takikawa et al. 1999) acquired along a trackbetween Pusan, Korea, and Hakata, Japan (Fig. 1) arealso compared with the voltage measurements.

The hourly raw voltage has short-period variationswith a range of about 1.5 V and a mean of 0.793 V.Ten-day segments of the time series of the voltage (darkline in Fig. 3a) and the depth-averaged along-strait cur-



FIG. 3. Ten-day time series of (a) raw voltage (dark line) and voltage corrected for geomagneticinduced voltage (light line) and (b) depth-averaged, along-strait current measured at N2 stationof the northern ADCP mooring line (Fig. 1). Here and elsewhere time is in days with day 0 being1 Jan 1998 in UTC.

rent (Fig. 3b) measured at N2 located on the northernADCP mooring line by the US NRL show prominentsemidiurnal and diurnal variations in both datasets. Ahigh coherence between datasets is visible with littlephase difference. However, there are occasionally abruptchanges in voltage that are not associated with the along-strait current. These changes are not related to transportvariations and will be discussed in the next section.

The observed voltage (Fig. 4a) and the along-straitcurrent (Fig. 4b) have prominent peaks in their powerspectra at diurnal and semidiurnal tidal frequencies. Atlow frequencies spectral energy density is high, espe-cially at subinertial and monthly timescales in the volt-age. One can notice in the voltage spectra, however, thatthere are significant peaks at frequencies of 3, 4, and 5cpd, which are not significant in the spectra of the along-strait current. These variations may be related to thesolar diurnal variations of geomagnetic fields at 1, 2, 3,4, and 5 cpd as reported by Larsen (1992). Since thesegeomagnetic effects on the observed voltage are sig-nificant at timescales shorter than 5 days according toLarsen (1992), they must be removed from the observedvoltage to investigate oceanographic tidal and subiner-tial variations.

3. Geomagnetic correction and error estimates ofvoltage

a. Geomagnetic correction

The method of Larsen et al. (1996) is applied for thegeomagnetic correction. This method involves construc-tion of a transfer function, Z, between voltage and geo-magnetic fields, and removal of the voltage variationsthat are coherent with geomagnetic ones. The horizontalgeomagnetic data at Kanoya and Memambetsu, Japan

(Fig. 1), are used to remove the geomagnetic effect fromthe observed voltage. The dominant geomagnetic com-ponent is located 1258 clockwise from the north, thatis, about 358 from the cable line. The magnitude of thetransfer function, Z, increases with frequency and itsphase has values from 908 to 1808 (Fig. 5). At lowfrequencies the magnitude is very small, which suggeststhat the geomagnetic effect is negligible for long peri-ods.

The geomagnetic-induced voltage, VG, is derivedfrom the transfer function and the corrected voltage,Vcor, is obtained by subtracting VG from the raw voltage,Vraw (Fig. 6). Tide-induced voltage, VT, is calculated byharmonic analyses of Vcor. The voltage related to thecurrent, Vres, becomes available by subtracting VT fromVcor. Most of the abrupt changes in Vraw are clearly elim-inated by this procedure as shown in Fig. 3a, where Vcor

is designated by a light line.Spectral analysis of VG (Fig. 4c) shows that VG con-

tains peaks at frequencies of 1, 2, 3, 4, and 5 cpd. Itsenergy density is very low for periods longer than 2days. In the spectrum of Vcor (Fig. 4d) geomagnetic sig-nals are reduced at 3, 4, and 5 cpd. The relatively highspectrum at 3 and 4 cpd seems to be partly related tononlinear effects of tidal currents in the shallow strait,as peaks also occur in the along-strait current (Fig. 4b).

b. Error estimates of voltage in relation with thevolume transport

After geomagnetic correction the cross-stream volt-age (DFV) is given by

DF (t) 5 DF (t) 1 DF (t),V L I (1)



where DFL is the local motion-induced voltage near thecable and DFI is caused by the horizontal electric cur-rent (Sanford 1971). According to Sanford (1971) thelocal motion-induced voltage is

L 1DF (t) 5 FL z E5 s(x, t)H(x) 1 t9(x)0

0

3 s(x, z, t)y(x, z, t) dz dx ,E 6[ ]2H(x)

(2)

where s is the electrical conductivity of the ocean, t9the conductance of the sediment and conducting crustunder the strait, H the water depth, L the width of the

strait, FZ the vertical geomagnetic component, a( )depth-averaged value, and y the downstream velocity.The local motion-induced voltage, DFL, is proportionalto the conductivity-weighted, vertically averaged ve-locity. Therefore, DFL may vary even with a fixed vol-ume transport if there are changes in the water con-ductivity and the spatial structure of the flow.

In this study we simply estimate the deviation of DFL

from the linear relation between DFL and the volumetransport by splitting s and y in Eq. (2) into two parts,average and deviation, as follows:

s(x, z, t) 5 s(x, t) 1 s9(x, z, t), (3)

y(x, z, t) 5 y (x, t) 1 y9(x, z, t). (4)

Then DFL can be written as

L 01DF (t) 5 F s(x, t)y (x, t)H(x) 1 s9(x, z, t)y9(x, z, t) dz dx (5a)L z E E[ ]s(x, t)H(x) 1 t9(x)0 2H(x)

LF sz 05 y (x, t)H(x) dx 1 « (5b)Es H 1 t90 0 0 0

F sz 05 T(t) 1 «, (5c)s H 1 t90 0 0

where subscript ‘‘0’’ is a cross-strait and time-averagedvalue, T the volume transport through the strait, and «the deviation from the linear relation between DFL andtransport. Thus, DFL is decomposed into two parts inEq. (5c); the first term is linearly related to the volumetransport and the second term is the deviation that iscaused by change of the flow structure over the varyingbottom depth and water conductivity.

It is possible to estimate this deviation, «, by usingcurrent data measured along the northern ADCP line inthe Korea Strait by the U.S. NRL (Jacobs et al. 2001)and water conductivity data calculated from hydro-graphic data of line 208 of the Korean OceanographicData Center (KODC) in 1998, which is nearly parallelto the northern ADCP line (Fig. 1). The current dataare low-pass filtered with a half-power period of 36 hand subsampled every 3 h. Figure 7 shows the time-averaged structure of the along-strait current in thenorthern section for the second half of the ADCP re-cords. Two main northeastward flows appear in thewestern and eastern side of the Korea Strait, while theflow is very weak and variable at the center due to theTsushima Island-induced wake (Perkins et al. 1999).Hydrographic data are measured at standard depths ev-ery two months. The magnitude of « is then estimatedas 0.011 V from these current and conductivity data.

As seen in Eq. (1), the cross-stream voltage (DFV)can be changed from the local-induced voltage (DFL)by DFI, which is caused by the electric current throughthe horizontal section when there are downstreamchanges of velocity and conductivity (Sanford 1971). Ifthe characteristic downstream length scale of the flowis Y, the horizontal electric current effect can be esti-mated as follows (Larsen 1992):

DF 5 2kDFI L

1k 5 , (6)

pY s H 1 t90 0 01 14L t*

where t * is the conductance of the land. The widthL of the Korea Strait, the water conductivity , ands 0

the water depth H 0 are about 200 km, 4 S m 21 , and100 m, respectively. Here and t * are assumed tot 90be 1000 S and 400 S using the values reported byUtada et al. (1986), and k can be estimated for variouscharacteristic downstream length scales (Table 2). Thecross-stream voltage (DFV ) generated by a fixedtransport will change if Y varies. If the length scaleY becomes larger, the horizontal electric current effectbecomes smaller from Eq. (6) and k changes less with



FIG. 4. Power spectrum with 95% confidence interval of (a) raw voltage, Vraw; (b) depth-averaged, along-strait current measured at N2station (Fig. 1) for the first (dark line) and second (light line) half of the ADCP records by the US NRL; (c) geomagnetic-induced voltage,VG; and (d) corrected voltage, Vcor.

Y. If downstream flows are assumed to have charac-teristic length scales between 200 and 600 km alongthe Korea Strait, k ranges from 0.11 to 0.27 and thestandard error in k is estimated to be 0.14. If thetransport through the Korea Strait is 3.0 3 10 6 m 3

s 21 and the conversion factor from voltage to transportis 8 3 10 6 m 3 s 21 V 21 , this error corresponds to avoltage error of 0.053 V. Since the horizontal electriccurrent effect (DF I ) increases with the width of thestrait as described in Eq. (6), the error of DF I ismuch larger than that of DFL in the Korea Strait withthe width of 200 km. This result is different from thatin the Florida Strait with the width of 60 km in Larsen(1992), where the horizontal electric current effect isnot dominant. The total rms error of voltage inestimating transport in the Korea Strait comes to0.054 V.

4. Conversion from voltage to transport

a. Conversion at subtidal frequencies

The residual voltage, Vres, is compared with the ob-served transport from current measurements by the U.S.NRL (Jacobs et al. 2001) after low-pass filtering witha half-power period of 36 h and subsampling every 3h. The cross spectra between transport and voltage areshown in Fig. 8. Dark and light lines denote the crossspectra of the transport of the southern section (TS) andthe northern section (TN) with voltage.

Over the first half of the ADCP transport observation(spring and summer), TS is highly correlated with volt-age at all periods (Fig. 8a), while coherency betweenTN and voltage is lower by 0.2. It is suspected that thelower coherency results from the missing current dataat N1 station (Fig. 1) in the first half of the deployment



FIG. 5. (a) Magnitude and (b) phase of smoothed (solid line) andband-averaged (dots with vertical bars for 95% confidence limits)transfer function, Z, calculated by the geomagnetic correction methodof Larsen et al. (1996).

FIG. 7. Time-averaged along-strait current (cm s21) in the northernvertical section for the second half of the ADCP records by the NRL,where northeastward currents are positive. The x axis is the distancein km from the Korean coast and y-axis depth in m.

FIG. 6. Time series of (a) raw voltage, Vraw; (b) geomagnetic-in-duced voltage, VG; (c) corrected voltage, Vcor; (d) tide-induced volt-age, VT; and (e) residual voltage, Vres.

TABLE 2. The values of k, which denotes the amount of the hor-izontal electric current effect on the local motion-induced voltageDFL in Eq. (6), for various downstream length scales (Y ).

Y (km) 100 200 300 400 500 600

k 0.42 0.27 0.20 0.15 0.13 0.11

period. The transport for the northern section could notresolve adequately the current intensification by the EastKorea Warm Current (EKWC) close to the Korean coast(Isobe et al. 1994). Additionally, there may also be moresurface intensification of current in the northern sectionthan in the southern section, which could not be mea-sured by the bottom-mounted ADCP. This effect is morepronounced during the first half of the record, sincecurrents are more baroclinic in spring and summer thanin fall and winter (Cho and Kim 1998; Lyu and Kim2003). Therefore TN may be underestimated in the firsthalf of the deployment period and this underestimationis clearly shown from May to July in 1999 (Jacobs etal. 2001). Phase differences are near 08 in both caseswith the variation range less than 108 (Fig. 8b). Thisimplies that there is little time lag between transport andvoltage and that the current is almost horizontally non-divergent in the Korea Strait at timescales longer than2 days.

In the second half of the ADCP records (fall andwinter) coherencies for transports from both ADCP sec-tions with voltage are high at all periods (Fig. 8c), re-flecting a weak baroclinicity and the complete data re-turn from the northern ADCP section for this period.Phase differences are also near 08 in both cases withthe variation range less than 108 (Fig. 8d). These resultsimply that voltage is closely related to the real transportthrough the Korea Strait. Here TS is used for calculatingthe relationship between transport and voltage since TN

data coverage is incomplete near Korea.The maximum likelihood method (Macdonald and

Thompson 1992) is used to obtain the linear relation



FIG. 8. Cross-spectra between transport and voltage after low-pass filtering with a half-power period of 36 h, where dark solid linesdesignate the cross-spectra between the transport of the southern section (TS) and voltage and light solid lines designate the cross-spectrabetween the transport of the northern section (TN) and voltage. (left) Coherency with 95% significance level (dashed line). (right) Cross-phase spectra with 95% confidence intervals (dark and light dashed lines for TS and TN, respectively). Upper panels are for the first transportmeasurement period and lower panels are for the second period.

between TS and voltage. To satisfy the basic assumptionin linear regression that each sample is not correlated,both time series are subsampled every two days, whichcorresponds to the critical time determined from auto-correlations. Standard errors of each coefficient of thelinear relation are estimated on the assumption that bothdatasets have known errors. The transport standard errorwas estimated to be 0.5 3 106 m3 s21 by Jacobs et al.(2001). The voltage standard error is estimated to be0.054 Volt in section 3.

For the above errors, the conversion factor from volt-age to the transport is given as L0 5 (8.06 6 0.63) 3106 m3 s21 V21 and the voltage bias corresponding tozero net volume transport in the vertical section in-cluding the cable ground points is V0 5 0.48 6 0.07 V(Fig. 9). The correlation coefficient is 0.85 and the stan-

dard error in L0 is 67.8%, which amounts to a voltage-derived transport error of 60.23 3 106 m3 s21 for atransport of 3 3 106 m3 s21. Considering the error inthe estimated voltage bias, the total standard error be-comes 60.6 3 106 m3 s21 for a transport of 3 3 106

m3 s21.As the cable is grounded at Pusan and Hamada using

the same electrode of copper, the bias voltage V0 of 0.48V in the observed voltage across the Korea Strait maybe caused by the differences in the temperature, salinity,and electrochemical state of the ground sites or DCcurrent in the vicinity of the cable (Larsen 1992). How-ever, it is still not clear what causes this bias voltageand how large its temporal changes may be. Since agood correlation between this voltage-derived transportand the sea level difference (SLD) across the Korea



FIG. 9. Linear regression between TS and voltage using 144 dataafter low-pass filtering with a half-power period of 36 h and sub-sampling every 2 days. The thick solid line is the linear regressionline and dashed lines are standard error bounds.

FIG. 10. Linear regression between monthly mean voltage andtransport measured by vessel-mounted ADCP between Pusan, Korea,and Hakata, Japan (Takikawa et al. 1999).

TABLE 3. Amplitudes of tidal transport variations from the tidemodel of Book et al. (2001) and tidal voltage variations and con-version factors from voltage to transport for four major tidal con-stituents.

ConstituentTransport

(106 m3 s21) Voltage (V)Conversion factor(106 m3 s21 V21)

M2

S2

K1

O1

4.451.942.261.78

0.3760.1650.1380.120

11.811.816.414.8

Strait was reported by Lyu and Kim (2003), this SLDmay be used as an independent data to continuouslyinspect steadiness of the conversion factor from cablevoltage to transport and any changes of the bias voltagein order to investigate long-term transport variationsfrom cable voltage.

Takikawa et al. (1999) has measured the velocity fieldbetween Pusan, Korea, and Hakata, Japan, using a ves-sel-mounted ADCP since February 1997 (Fig. 1). Alinear relationship is also calculated using monthly meantransport from these measurements and voltage fromApril to November in 1998 (Fig. 10), after removingthe mean transport for August 1998 that deviates twotimes more than the rms deviation from the regressionline. The correlation coefficient is as high as 0.98 witha bias of V1 5 0.45 V and conversion factor of L1 58.0 3 106 m3 s21 V21, which are within the error boundsof V0 and L0 estimated above, in spite of the differencein measurement methods of transport and sampling in-tervals. This consistency indicates clearly that the linearrelation between voltage and transport is reliable at thesesubtidal frequencies.

b. Conversion at tidal frequencies

Tidal transport in the Korea Strait has been estimatedfrom tide models by Kang et al. (1991) and Choi et al.(1994) and from limited current meter data by Odamaki(1989). They reported that the amplitude of M2 tidaltransport is about 5.0 ; 5.6 3 106 m3 s21 across thenortheastern part of the Korea Strait, which is largerthan the known mean transport (2.6 ; 2.7 3 106 m3

s21) in the Korea Strait (Takikawa et al. 1999; Teagueet al. 2002).

Recently tidal currents have been assimilated in abarotropic model for the Korea Strait using coastal sealevel data, TOPEX/Poseidon altimetry data and currentdata (Book et al. 2001). Amplitudes of tidal transportsacross the cable section are calculated as 4.45, 1.94,2.26, and 1.78 3 106 m3 s21 for M2, S2, K1, and O1,respectively, from this assimilation (Table 3). Conver-sion factors derived by comparing tidal amplitudes ofmodel-derived transport and voltage are 11.8, 11.8,16.4, and 14.8 3 106 m3 s21 V21 for M2, S2, K1, andO1, respectively, as shown in Table 3.

The frequency of K1 is close to 1 cpd, where thegeomagnetic effect is dominant and this effect cannotreadily be removed if the geomagnetic data are not mea-sured near the cable ground site (Larsen 1992). Sincethe geomagnetic data used in this study are measuredin the region farther than 400 km from the cable groundsite (Fig. 1), geomagnetic effects may remain in thevoltage-derived transport at the frequency of K1. How-ever derived conversion factors from other tidal con-



FIG. 11. Time series of voltage-derived transport TV (dark line) and TS (light line), whichis observed by the bottom-mounted ADCP, after low-pass filtering with a half-power periodof 36 h.

stituents are still larger than L0 derived from subtidaltransport variations (Table 3).

The difference in conversion factors can be explainedby the effect of horizontal electric current on the ob-served voltage in Eq. (6). The characteristic length scaleof tidal currents in the Korea Strait can be estimated tobe about 200 km (Book et al. 2001; Odamaki 1989). Inthis case the cross-stream electric potential difference,DFV may be reduced by about 27% at most from thelocal motion-induced potential difference, DFL (Table2). Since the amplitude of DFV is 0.376 V for M2, DFL

estimated from Eqs. (1) and (6) is 0.52 V. Comparingthis DFL and the tidal transport for M2, the conversionfactor changes to 8.6 3 106 m3 s21 V21. For O1, DFV

is 0.120 V and DFL is estimated to be 0.17 V, whichgives the conversion factor of 10 3 106 m3 s21 V21.

The difference in conversion factors for M2 and O1

could be explained by considering three factors. First,the amplitude and the phase of O1 signal in cable voltagehave relatively larger variances than those of M2 signal(Lyu et al. 2002b), that is, the former is less stable thanthe latter. Second, the amphidromic point of O1 tide isreproduced near Pusan from the tidal model (Book etal. 2001). This implies that O1 tidal currents are morecomplex than M2 currents near the cable section. Third,tidal flow in the deep ocean will have length scales as

large as 1000 km and this will give rise to horizontalelectric currents that will combine with the more locallyinduced horizontal electric currents, especially in thedirection of the strait. This will further confuse the tidalinterpretation since the cable is not perpendicular to thecoastline in the case of the Korea Strait (J. C. Larsen2001, personal communication).

The horizontal electric current effects must be con-sidered to analyze tidal transports from the observedvoltage across the Korea Strait. However, these effectsare less than about 10% of the local motion-inducedvoltage for subtidal downstream currents, assumingcharacteristic length scales longer than 500 km alongthe strait.

5. Transport variations in the Korea Strait

The voltage-derived transport, TV (dark line in Fig.11), from the conversion factor L0 and the bias V0 es-timated in section 4 and observed transport TS (light linein Fig. 11) from the bottom-mounted ADCP (Teague etal. 2002) have similar mean values and standard devi-ations: (2.69 6 0.83) 3 106 m3 s21 and (2.64 6 0.82)3 106 m3 s21 for TV and TS, respectively. Their cor-relation coefficient is as high as 0.85. The differencesin their time series are mostly less than the estimated



FIG. 12. Seasonal variations of TV (solid dark line) after low-pass filtering with a half-powerperiod of 90 days, where dashed lines are 95% confidence intervals. Open and closed circlesdesignate monthly mean transport observed by the vessel-mounted ADCP and the bottom-mountedADCP, respectively. The solid light line is the mean transport of TV.

standard error 0.6 3 106 m3 s21 in voltage-derived trans-port.

There are transport variations as large as 2 ; 3 (3106 m3 s21) on a timescale of 3 ; 7 days and about amonth in TV as shown in Fig. 11. It should be pointedout that these subinertial variations are larger than theknown seasonal variations in the Korea Strait (Takikawaet al. 1999; Teague et al. 2002). These transport fluc-tuations in the Korea Strait have not been known pre-viously. Although there are some differences in ampli-tude and phase between TS and TV, volume transportvariations clearly occur with timescales of 3 ; 7 daysand variation ranges of 2 ; 3 (3 106 m3 s21) in theKorea Strait. Lyu et al. (2002a) reported that these var-iations are related to nonisostatic responses of the EastSea to the atmospheric pressure forcing. These varia-tions as well as tidal variations must be separated fromthe instantaneous transport observations to investigatelong-term variations in the Korea Strait.

To investigate seasonal variations, a low-pass filterwith a half-power period of 90 days is applied to TV.The result shown in Fig. 12 has a mean of 2.5 3 106

m3 s21 with a maximum of 3.4 3 106 m3 s21 in October1999 and a minimum of 1.6 3 106 m3 s21 in January2000. The monthly mean transports from the vessel-mounted ADCP (Takikawa et al. 1999) and the bottom-mounted ADCP (Teague et al. 2002), indicated in Fig.12, agree well with TV within a 95% confidence interval.There seems to be a tendency that the transport becomeslarge in summer and autumn and small in winter andspring. However, interannual variations are prominent.The seasonal variation is very weak from spring 2000to spring 2001 and annual mean transport reduces by0.4 3 106 m3 s21 from 1999 to 2000.

6. Summary

Cable voltage measured from the abandoned sub-marine telephone cable between Pusan, Korea, and Ha-

mada, Japan, since March 1998 is used to monitor thevolume transport through the Korea Strait. The geo-magnetic-induced voltage, which is substantially largeat periods shorter than two days, is removed from themeasured voltage by the method of Larsen et al. (1996).The observed voltage in the Korea Strait is found to becontaminated by the horizontal electric current due toa relatively large width of the strait. This horizontalelectric current effect can cause time/length scale de-pendence of the relationship between voltage and trans-port in the Korea Strait. It is found that there is a verygood linear relationship between voltage and the ob-served transports from the bottom-mounted ADCP(Teague et al. 2002) and the vessel-mounted ADCP(Takikawa et al. 1999). The conversion factor from volt-age to transport is estimated to be L0 5 (8.06 6 0.63)3 106 m3 s21 V21 with the bias voltage of V0 5 0.486 0.07 V.

In the voltage-derived transport, tidal variations aredominant with amplitudes larger than the mean transportknown in the Korea Strait. However, the horizontal elec-tric current effects must be considered to estimate tidaltransports from the observed voltage across the KoreaStrait because the characteristic downstream lengthscale of tidal currents is comparable to the width of thestrait. It is fortunate, however, that these effects are smallfor subtidal downstream currents, which have long char-acteristic length scales.

There are transport variations as large as 2 ; 3 (3106 m3 s21) on timescales of 3 ; 7 days. There are alsolarge variations in the transport with a range of 2 3 106

m3 s21 on timescales of about a month. Three-year datashow a relatively larger transport in summer and autumnthan in winter and spring but there is a large interannualvariation. The voltage measured between Pusan and Ha-mada provides a good means to monitor the transportthrough the Korea Strait continuously and enables us toinvestigate various temporal transport variations and



their causes, which affect the circulation and the hy-drography in the East Sea decisively.

Acknowledgments. This work was supported byOCEAN Laboratory and the BK21 project of the KoreanGovernment in 1999–2003. H. T. Perkins, W. J. Teague,and J. W. Book were supported by the U.S. Office ofNaval Research as part of the Basic Research Project‘‘Linkages of Asian Marginal Seas’’ and ‘‘Japan EastSea DRI’’ under Program Element 0601153N.

REFERENCES

Baringer, M., and J. C. Larsen, 2001: Sixteen years of Florida currenttransport at 278N. Geophys. Res. Lett., 28, 3179–3182.

Book, J. W., and Coauthors, 2001: Tides in the Korea/Tsushima Strait:Observation and model predictions. Proc. 11th PAMS/JECSSWorkshop, Cheju, Korea, Korean Science and Engineering Foun-dation and the Korean Foundation of Science and TechonologySocieties, 207–210.

Bowden, K. F., 1956: The flow of water through the Strait of Doverrelated to wind and differences in sea level. Philos. Trans. Roy.Soc. London, A248, 517–551.

Byun, S. K., and Y. H. Seung, 1984: Description of current structureand coastal upwelling in the southwest Japan Sea—Summer1981 and spring 1982. Ocean Hydrodynamics of the Japan andEast China Seas, T. Ichiye, Ed., Elsevier, 83–93.

Cho, Y. K., and K. Kim, 1998: Structure of the Korea Strait BottomCold Water and its seasonal variation in 1991. Cont. Shelf Res.,18, 791–804.

Choi, B.-H., K. Kim, Y.-G. Kim, K. S. Bahk, J. O. Choi, and K.Kawatate, 1992: Submarine cable measurements between Pusanand Hamada. La Mer, 30, 157–167.

——, I. K. Bang, and K. H. Kim, 1994: Vertical distribution of tidalcurrent in the Korea Strait. J. Korean Soc. Coastal Ocean Eng.,6, 421–438.

——, K. Kim, Y.-G. Kim, and K. Kawatate, 1997: Submarine cablevoltage measurement between Pusan and Hamada for the years1987–1996. Acta Oceanogr. Taiwan., 36, 33–46.

Holloway, G., T. Sou, and M. Eby, 1995: Dynamics of circulation ofthe Japan Sea. J. Mar. Res., 53, 539–569.

Isobe, A., 1994: Seasonal variability of the barotropic and baroclinicmotion in the Tsushima–Korea Strait. J. Oceanogr., 50, 223–238.

——, S. Tawara, A. Kaneko, and M. Kawano, 1994: Seasonal var-iability in the Tsushima Warm Current, Tsushima–Korea Strait.Cont. Shelf Res., 14, 23–35.

Jacobs, G. A., H. T. Perkins, W. J. Teague, and P. J. Hogan, 2001:Summer transport through the Korea–Tsushima Strait. J. Geo-phys. Res., 106, 6917–6929.

Kang, S. K., S.-R. Lee, and K.-D. Yum, 1991: Tidal computation ofthe East China Sea, the Yellow Sea and the East Sea. Ocean-ography of Asian Marginal Seas, K. Takano, Ed., ElsevierOceanography Series, Vol. 54, Elsevier, 24–48.

Katoh, O., K. Teshima, K. Kubota, and K. Tsukiyama, 1996: Down-stream transition of the Tsushima Current west of Kyushu insummer. J. Oceanogr., 52, 93–108.

Kawatate, K., and Coauthors, 1991: A cross-spectral analysis of smallvoltage variation in a submarine cable between Hamada andPusan with speed variation of the Tsushima Warm Current.

Oceanography of Asian Marginal Seas, K. Takano, Ed., ElsevierOceanography Series, Vol. 54, Elsevier, 207–222.

Kim, C.-H., and J.-H. Yoon, 1999: A numerical modeling of the upperand the intermediate layer circulation in the East Sea. J. Ocean-ogr., 55, 327–345.

Larsen, J. C., 1992: Transport and heat flux of the Florida Currentat 278N derived from cross-stream voltages and profiling data:Theory and observations. Philos. Trans. Roy. Soc. London,A338, 169–236.

——, R. L. Mackie, A. Manzella, A. Fiordelisi, and S. Rieven, 1996:Robust smooth magnetotelluric transfer functions. Geophys. J.Int., 124, 801–819.

Longuet-Higgins, M. S., 1949: The electrical and magnetic effect oftidal streams. Mon. Not. Roy. Astron. Soc., 16, 285–307.

Lyu, S. J., and K. Kim, 2003: Absolute transport from the sea leveldifference across the Korea Strait. Geophys. Res. Lett., 30, 1285,doi:10.1029/2002GL016233.

——, ——, and H. T. Perkins, 2002a: Atmospheric pressure-forcedsubinertial variations in the transport through the Korea Strait.Geophys. Res. Lett., 29, 1294, doi:10.1029/2001GL014366.

——, Y.-G. Kim, K. Kim, J. W. Book, and B. H. Choi, 2002b: Tidalvariations in the cable voltage across the Korea Strait. J. KoreanSoc. Oceanogr., 37, 1–9.

Macdonald, J. R., and W. J. Thompson, 1992: Least-squares fittingwhen both variables contain errors: Pitfalls and possibilities.Amer. J. Phys., 60, 66–73.

Miita, T., and Y. Ogawa, 1984: Tsushima currents measured withcurrent meters and drifters. Ocean Hydrodynamics of the Japanand East China Seas, T. Ichiye, Ed., Elsevier OceanographySeries, Vol. 39, Elsevier, 67–76.

Mizuno, S., K. Kawatate, T. Nagahama, and T. Miita, 1989: Mea-surements of east Tsushima current in winter and estimation ofits seasonal variability. J. Oceanogr. Soc. Japan, 45, 375–384.

Moriyasu, S., 1972: The Tsushima Current. Kuroshio, H. Stommeland K. Yoshida, Eds., University of Tokyo Press, 353–369.

Odamaki, M., 1989: Tides and tidal currents in the Tsushima Strait.J. Oceanogr. Soc. Japan, 45, 65–82.

Perkins, H. T., W. J. Teague, G. A. Jacobs, K.-I. Chang, and M.-S.Suk, 2000: Currents in Korea–Tsushima Strait during summer1999. Geophys. Res. Lett., 27, 3033–3036.

Sanford, T. B., 1971: Motionally induced electric and magnetic fieldsin the sea. J. Geophys. Res., 76, 3476–3492.

——, and R. E. Flick, 1975: On the relationship between transportand motional electric potentials in broad, shallow currents. J.Mar. Res., 33, 123–139.

Takikawa, T., J.-H. Yoon, H. Hase, and K.-D. Cho, 1999: Monitoringof the Tsushima Current at the Tsushima/Korea Straits. Proc.3rd CREAMS Int. Symp., Fukuoka, Japan, Japanese Ministry ofEducation, Science, Sports and Culture, 15–18.

Tawara, S., T. Miita, and T. Fujiwara, 1984: The hydrography andvariability in the Tsushima Straits (in Japanese). Bull. CoastalOceanogr., 22, 50–58.

Teague, W. J., G. A. Jacobs, H. T. Perkins, J. W. Book, K.-I. Chang,and M.-S. Suk, 2002: Low-frequency current observations in theKorea/Tsushima Strait. J. Phys. Oceanogr., 32, 1621–1641.

Uda, M., 1934: The results of simultaneous oceanographical inves-tigations in the Japan Sea and its adjacent waters in May andJune, 1932 (in Japanese). J. Imp. Fish. Exp. Sta., 5, 57–190.

Utada, H., Y. Hamano, and T. Yukutake, 1986: A two-dimensionalconductivity model across central Japan. J. Geomagn. Geo-electr., 38, 447–473.

Yi, S.-U., 1966: Seasonal and secular variations of the water volumetransport across the Korea Strait. J. Oceanol. Soc. Korea, 1, 7–13.


monitoring volume transport through measurement of cable

Documents