Naokazu Taniguchi

Mapping Tidal Current Structures in Zhitouyang Bay, China, Using Coastal Acoustic Tomography

The first Chinese coastal acoustic tomography (CAT) experiment for mapping the tidal currents in Zhitouyang Bay near Zhoushan Island was successfully performed with seven acoustic stations from July 12 to 13, 2009. Using CAT, the horizontal distributions of the tidal currents in the tomography domain were calculated by the inverse analysis, in which the travel time differences for sound traveling reciprocally between the station pairs are used as data. The specified tidal current structures, such as the strong east-west oscillation of the tidal current, the branched current, and the tidal vortices, were reconstructed as snapshots at the successive tidal phases. The relative vorticity calculated from the inverted current fields served to specify the current structures, such as tidal vortices. The inversion-estimated uncertainty of (0.02-0.08) m s-1 narrowed the root-mean-square difference (RMSD) of (0.00-0.11) m s-1 between the 3-min interval original data and the hourly mean data for all the sampled data, which may be regarded as a measure of error. Throughout the tidal phases, the divergence from the inverted current showed a positive (negative) distribution in the shallow (deep) region as an overall view. However, the divergence for the entire tomography domain was nearly equal to zero, corresponding to no net transport. This result implies that the observational errors are quite small for the present experiment. This experiment reaffirms that coastal acoustic tomography is an accurate and efficient observational method for continuously mapping tidal current structures in coastal regions that are characterized by heavy shipping traffic and active fishing.

Measuring the Kuroshio Current with ocean acoustic tomography

Ocean current profiling using ocean acoustic tomography (OAT) was conducted in the Kuroshio Current southeast of Taiwan from August 20 to September 15, 2009. Sound pulses were transmitted reciprocally between two acoustic stations placed near the underwater sound channel axis and separated by 48 km. Based on the result of ray simulation, the received signals are divided into multiple ray groups because it is difficult to resolve the ray arrivals for individual rays. The average differential travel times from these ray groups are used to reconstruct the vertical profiles of currents. The currents are estimated with respect to the deepest water layer via two methods: An explicit solution and an inversion with regularization. The strong currents were confined to the upper 200 m and rapidly weakened toward 500 m in depth. Both methods give similar results and are consistent with shipboard acoustic Doppler current profiler results in the upper 150 m. The observed temporal variation demonstrates a similar trend to the prediction from the Hybrid Coordinate Ocean Model.

The first Chinese coastal acoustic tomography experiment

The first Chinese coastal acoustic tomography (CAT) experiment was successfully carried out during July 12 to 13, 2009 with seven acoustic stations for mapping the tidal currents in the Zhitouyang Bay near the Zhoushan Island. The CAT results showed the strong eastward tidal current which entered into the tomography site from the Loutou Channel at the western part of the bay, and separated into two branches, flowing into the Laoshumen Channel, Qingzimeng Channel and Fodu Channel during the ebb tide. On the other hand, the strong westward tidal current, joined with the currents from Laoshumen Channel, Qingzimeng Channel and Fodu Channel, went back to the Loutou Channel during the flood tide. A clockwise tidal vortex of diameter about 5 km was generated in the eastern part of the bay in the transition phase from the ebb to flood. No tidal vortices were generated in the weak tidal currents in the transition phase from the flood to ebb, changing the current direction from westward to eastward. The range-averaged current obtained along the transmission lines after the inverse analysis was in agreement with the acoustic Doppler current profiler (ADCP) results, producing the root-mean-square difference (RMSD) of 1.77 cm/s. The horizontal distributions of tidal current calculated by the inverse analysis were in rough agreement with that of the shipboard ADCP velocity. The RMSD is 11.70 cm/s for the eastward current and 16.50 cm/s for the northward current. It is suggested that the CAT is a powerful instrument for continuously mapping the horizontal tidal current structures in coastal regions in China with heavy shipping traffic and active fishing.

Acoustic current measurement using travel-time method in Bachimen Harbor, Taiwan

A reciprocal acoustic transmission experiment was conducted in Bachimen Harbor, Keelung on December 4, 2009. Sound pulses were transmitted reciprocally between two Coastal Acoustic Tomography Systems (CATs) placed on the North bank and separated by 490 m. The received pulse responses show two dominant ray groups. Using horizontal ray simulation from the BELLHOP ray tracing program, the two arrival groups were identified as the direct rays and the horizontally reflected rays reflected from the North bank. Two methods were used to obtain the differential travel times: peak-picking and waveform-matching methods. In general, the results from the reciprocal transmission data were in good agreement with the measurement of Horizontal Acoustic Doppler Current Profiler (HADCP). The results suggest that reciprocal sound transmission is applicable to current reconstruction in the harbor sea.

Autonomous underwater vehicle localization using ocean tomography sensor nodes

The work is a preliminary study of autonomous underwater vehicle (AUV) localization using ocean acoustic tomography (OAT) sensor nodes. OAT is a method to measure the current velocity and temperature fields using the time-of-flight of acoustic signals. These acoustic sensor nodes broadcast signals periodically in the area of interest. Each ocean tomography sensor node consists of a surface buoy, an acoustic transceiver, GPS module and the processing unit. An acoustic tomography sensor was installed on an AUV, acting as a moving node in addition to the seafloor moored nodes. Since the GPS signal is not always available on AUV, the chip scale atomic clock is operated as the system clock to reduce the clock drift. With the scheduled transmission times, the one-way travel time is measured between the moored nodes and the AUV. The high accuracy of travel time between moored nodes and AUV is used to improve the AUV localization. Based on dead-reckoning and travel-time measurement, Extended Kalman Filter is employed to improve the vehicle localization. An experiment was conducted using an AUV and one moored node deployed in a shallow water environment southwest of Taiwan. The AUV was operated near the surface to obtain the GPS position as the ground truth for the performance evaluation. Preliminary results show the RMS position error between EKF prediction and the ground truth is about 80 m by incorporating the tomographic signals into a low-cost AUV navigation system.

Simulated tomographic reconstruction of ocean current profiles in a bottom‐limited sound channel

Tomographic reconstruction of the vertical current profile in a bottom-limited sound channel requires solving a difficult ray identification problem. An approach to deal with this problem is a ray group method in which received arrival pulses are divided into several ray groups according to the characteristics of the arrival patterns. The method is validated using numerically simulated reciprocal acoustic transmission in a synthetic ocean in the Luzon Strait, where the Kuroshio Current has speeds as high as 1.2 m/s, for both narrowband and broadband signals. Four ray groups are found for the synthetic data; these are chosen based on arrival time. The differential travel time is determined by pairing up the reciprocal arrival peaks and then averaging the differential travel times within the selected time windows. Compared with the narrowband case, the estimated broadband differential travel time is more consistent with that computed from the current magnitude in the synthetic ocean. The vertical current profile is reconstructed from the broadband differential travel times by a generalized Tikhonov regularization approach. The data weighting matrix includes observation error in picking and pairing travel times and model parameter error due to path length uncertainty. The time series of the reconstructed current agrees with the synthetic ocean current; the fractional residual variance is 0.013 for the surface layer and 0.01 for the entire water column. The ray group method mitigates the ray identification problem in the bottom-limited environment and could offer valuable data regarding the range-integrated current velocity.

Acoustic mapping of ocean currents using moving vehicles

With the increased availability of highly maneuverable unmanned surface and underwater vehicles, abundant ocean environmental data can now be collected. Most environmental surveys by unmanned vehicles conduct point measurements of the ocean properties along survey lines. This study uses tomographic techniques to extend the survey area covered by the autonomous vehicles and obtain a synoptic ocean current distribution. An acoustic reciprocal transmission experiment was carried out on June 27, 2017, in ChaoJing Bay nearby Keelung City, where the water depth varies from 20 m to 65 m. A total of three tomographic sensors were deployed and they were installed on an AUV, a fishing boat, and a bottom-moored buoy. Reciprocal acoustic transmissions between the mobile platforms were used to estimate ocean currents, which required accounting for the Doppler effects on the acoustic arrival patterns and the resulting differential travel times. The estimated areal currents show consistency with the ADCP current measurement from the boat when the covered area was near the deeper water. With the increased availability of highly maneuverable unmanned surface and underwater vehicles, abundant ocean environmental data can now be collected. Most environmental surveys by unmanned vehicles conduct point measurements of the ocean properties along survey lines. This study uses tomographic techniques to extend the survey area covered by the autonomous vehicles and obtain a synoptic ocean current distribution. An acoustic reciprocal transmission experiment was carried out on June 27, 2017, in ChaoJing Bay nearby Keelung City, where the water depth varies from 20 m to 65 m. A total of three tomographic sensors were deployed and they were installed on an AUV, a fishing boat, and a bottom-moored buoy. Reciprocal acoustic transmissions between the mobile platforms were used to estimate ocean currents, which required accounting for the Doppler effects on the acoustic arrival patterns and the resulting differential travel times. The estimated areal currents show consistency with the ADCP current measure...

Variation of Residual Current in the Seto Inland Sea Driven by Sea Level Difference Between the Bungo and Kii Channels

Understanding residual (i.e., total minus tidal) currents in coastal seas is important because the residual currents affect long-term material transports. In the Seto Inland Sea, Japan (SIS), a Bungo Channel to Kii Channel sea level difference causes a horizontal pressure gradient in the SIS and thus affects the residual current in the SIS. This study applies a linear regression method to examine how the residual current responds to the Bungo-Kii sea level difference. The residual current is obtained using the reciprocal acoustic transmission data collected in the eastern portion of the Aki-Nada sea area in 2012. The residual currents are estimated in the following three periods: from 12 April to 9 June, from 15 June to 21 July, and from 20 September to 27 October. In the regression analysis, an additional term is included to account for the fortnightly variation of the tide-induced residual current. More than 75% of the observed residual currents can be explained by the sea level difference and the fortnightly variation. For the three periods, the variations of the residual current along the acoustic transmission line are, respectively, 0.20 6 0.01, 0.22 6 0.01, and 0.30 6 0.01 cm s per 1 cm of the Bungo-Kii sea level difference. The corresponding variations in the volume transport are 920 6 149, 1; 0406170, and 1; 3906223 m s per 1 cm sea level difference. Comparing with the wind-induced volume transport, we find that the sea level difference can cause a comparable volume transport variation.

Autonomous underwater vehicle navigation using a single tomographic node

This study investigates AUV navigation using the one-way travel-time from an ocean tomographic sensor node. For the study of ocean acoustic tomography, several tomographic nodes are moored in the study area and broadcast m-sequence signals periodically. Each moored node consists of an acoustic transceiver, a GPS module, and a processing unit. The pulse-per-second signal from the GPS module is used to ensure the time synchronization of all the moored nodes. A tomographic sensor is installed on an AUV, serving as a moving node. To obtain highly accurate one-way travel-time between moored nodes and AUV, a chip scale atomic clock is operated on the AUV. Based on dead-reckoning and travel-time measurement, Extended Kalman Filter (EKF) is employed to improve the AUV localization. An experiment was conducted using an AUV and one moored node deployed in a shallow water environment southwest of Taiwan. The measurements from a Doppler velocity logger and a compass are used to calculate the current position of vehic...

Moving ship tomography experiments in a coastal sea

This study considers measuring differential travel times (DTT) between moored and ship-towed transceivers in shallow waters as a feasibility study to extend the concept of the moving ship tomography (Cornuelle et al., 1989) to current field reconstructions. The delay-Doppler ambiguity function (AF) method is used to estimate and compensate the waveform distortions induced by the relative motion between transceivers. The Doppler shift (the relative speed between transceivers) is estimated by the peak of the AF, and the delay time series associated with the peak is the Doppler compensated arrival pattern. The DTTs are determined using the compensated arrival patterns in reciprocal directions. For the current estimation the estimated relative speed is used for removing the effect of the transceiver motion from the DTTs. A reciprocal sound transmission experiment was conducted using a ship-towed transceiver and two moored transceivers in the Sizihwan coastal near Kaohsiung, Taiwan. Due to the complicated arri...