Elsevier

Applied Ocean Research

Volume 102, September 2020, 102303
Applied Ocean Research

Revisiting a coastal acoustic tomography experiment in Hiroshima Bay: Temporal variations in path-averaged currents and its relation to wind

https://doi.org/10.1016/j.apor.2020.102303Get rights and content

Abstract

Remote sensing is a desirable method for measuring ocean currents on various spatial and temporal scales. Coastal acoustic tomography (CAT) can measure current velocity in coastal and shallow seas, but the application of this method in stratified shallow estuaries has not been sufficiently tested. This study revisits the acoustic transmission data obtained from the 2013 Hiroshima Bay CAT experiment and investigates what information could be obtained from acoustic data when a relatively strong southward wind blew in the bay. We determined path-averaged currents of the first arrival rays and compared the temporal variations in this path-averaged current with those in the wind, water temperature, and detided sea level. When the southward wind was intensifying, the sea level at the northern coast of the bay decreased and then recovered when the southward wind subsided. These sea level variations must be accompanied by net volume transport variations. When sea level was recovering, the path-averaged current exhibited a positive bias, reflecting an increase of volume transport toward the inner part of the bay. In contrast, the path-averaged current did not show a noticeable variation when the sea level was decreasing although the surface layer current was expected to be intensified southward. This is because the first arrival ray did not propagate through the surface layer during the period, and therefore, the path-averaged current from the first arrival rays did not contain surface current velocity information. CAT is promising method for measuring variations in current velocity fields associated with winds; however, in the application to stratified and shallow estuaries such as Hiroshima Bay, it is crucial to consider acoustic ray-path variations when analyzing and interpreting travel time data time series.

Introduction

Coastal and shallow sea circulation consists of multiple processes such as tidal currents, wind-driven currents, and density-driven currents. Remote sensing methods that can observe variations due to those processes on different spatial and temporal scales are desirable. One such method is measuring the travel times of sound pulses that are reciprocally transmitted and received between horizontally separated transceiver systems. A sound pulse traveling along an acoustic ray path in water travels faster in the current velocity direction as well as in warmer and more saline water. The path-averaged current velocity and sound speed can be measured, respectively, using the differential travel times (the difference of reciprocally determined travel times) and the summed travel times (e.g. [6], [10], [19]). Deploying multiple transceivers enables estimation of the spatial distributions of sound speed and current by solving an inverse problem or using data assimilation schemes. The method, known as ocean acoustic tomography (OAT), was proposed by Munk and Wunsch [11] to measure and understand the behaviors of mesoscale (and larger) features associated with general ocean circulation [10]. OAT has been applied to shallow waters, with some applications focusing on the horizontal variation of tidal currents. This application is referred to as coastal acoustic tomography (CAT; [9], [14], [20], [22], [25], [27]). Although those studies have focused mainly on the depth-averaged tidal current, recent studies have also attempted to reconstruct three-dimensional current structures using data assimilation schemes [2], [29]. Additionally, CAT has recently been used to investigate relatively small magnitude currents such as overtidal and residual currents [17], [28].

In 2013, a CAT experiment was conducted in the northern portion of Hiroshima Bay, Japan, to investigate the upwelling caused by winds and sea level anomalies due to internal surges [2], [23], [24]. Hiroshima Bay is a semi-enclosed bay in the western Seto Inland Sea, Japan (Fig. 1). In the northern Hiroshima Bay, density stratification exists because of low salinity water associated with Ota River inflow on the surface as well as high temperature during the summer. Water circulation has been studied in Hiroshima Bay (e.g. [[5], [18]]). Aside from tidal variations, estuarine circulation induced by the Ota river inflow may be the primary circulation process in Hiroshima Bay [21]. Winds also play an essential role as a forcing driver [12], particularly for upwelling, which lifts sub-surface water with high dissolved nutrient content to the surface. In Hiroshima Bay, winds may also cause upwelling of oxygen-deficient water and spread toxic plankton cysts from bottom sediments [16]. Because Hiroshima Bay is closely related to human activities such as fisheries, aquaculture, marine leisure, transportation, and reclamation, it is important to monitor the bay’s water environment. CAT is a promising tool that acquires temporal and horizontal variations of current fields in shallow and semi-enclosed bays such as Hiroshima Bay. One of CAT’s advantages is that reciprocal acoustic transmissions do not interrupt fishing and shipping activities as the acoustic transceivers are deployed from quays or near shore. Additionally, if the power supplies are supported by solar cells [1], acoustic transmissions can be performed over a long period with sufficiently short intervals.

During the 2013 Hiroshima Bay CAT experiment, a relatively strong northerly wind (southward wind) blew over a 48 h period along with a typhoon passage. By applying an inversion technique to the travel times, Zhang et al. [24] reconstructed the horizontal water temperature distribution and examined upwelling in northern Hiroshima Bay caused by this cross-shore wind. The mean water depth of northern Hiroshima Bay is approximately 18 m, and in such shallow waters, upwelling is not only explained by classical Ekman layer theory but also by cross-shore winds (e.g. [8]). Chen et al. [2] assimilated travel time data into a numerical ocean model to further explore wind-induced upwelling and mixing. In northern Hiroshima Bay, southward winds cause a reverse flow, i.e., a southward flow in the upper layer and a northward flow in the lower layer, which results in upwelling at the northern coast [2], [16]. The observation results of Takasugi et al. [16] showed that the time variation of surface currents synchronized with that of the wind. By contrast, Chen et al. [2] found a time lag of approximately 24 h between the maximum southward wind and associated surface transport. However, they did not discuss this time lag in detail.

This study revisits data obtained by the 2013 CAT experiment in Hiroshima Bay and explores acoustic arrival data characteristics. Understanding acoustic arrival data characteristics is essential to appropriate application of inversion techniques or data assimilation schemes to travel time data. In particular, we investigate what information was detected by acoustic travel times during the southward wind event. For this purpose, we compare the path-averaged current with other available data such as wind, sea level variation, and water temperature. Although Zhang et al. [24] and Chen et al. [2] focused mainly on horizontal variation, we keep our focus on the path-averaged value as it can detect small variations due to external forces. For example, in the Seto Inland Sea, differential travel times can detect variations in the mean (time-averaged) current caused by the sea level difference between the west and east entrances (the Bungo and Kii channels) of the Seto Inland Sea [17]. Because a relatively strong wind blew, the path-averaged current would capture some signals related to the wind. By explaining the path-averaged current’s time variation, we show that the reciprocal acoustic transmissions (and CAT) have potential to measure variations in current velocity fields associated with winds. The remainder of this paper is organized as follows. The experiment and data to be investigated are described in Section 2; Results and discussion are presented in Sections 3 and 4, respectively; and a summary of this study is provided in Section 5.

Section snippets

Wind, water temperature, and sea level

This study discusses the time variations of the path-averaged current and sound speed (see Section 2.2) together with those of wind, water temperature, and sea level. The wind data used in this study are hourly wind speed and direction measured at the Hiroshima Local Meteorological Observatory (square symbol in Fig. 1b). The wind direction is expressed in 16 directions: north, north northeast, northeast, and so on.

During the CAT experiment, the vertical water temperature profile was measured

Wind, water temperature, and sea level

Fig. 2 shows temporal variations of wind, water temperature, sea level, and detided sea level. The wind shows periodic sea/land breezes (i.e., wind blowing toward land during the day and offshore at night), except during a southward wind event on September 15–17. The southward wind reached a maximum speed of approximately 14 m/s around 2:00 on September 16. Water temperatures showed diurnal variations at the surface before the southward wind event. Surface temperature rapidly decreased as the

Discussion

Our primary purpose in this study is to investigate what information could be extracted from path-averaged current data from the 2013 Hiroshima Bay CAT experiment, particularly during a strong southward wind event. The estimated path-averaged current showed a positive bias (intensification toward the H2 station) when the southward wind subsided and the depressed sea level on the northern coast of Hiroshima Bay recovered. The recovery of sea level must be accompanied by an increase in net volume

Summary and conclusion

This study revisited the acoustic transmission data obtained by the 2013 Hiroshima Bay CAT experiment [2], [24]. We reconsidered what information could be obtained from differential travel times (i.e., path-averaged current) during the southward wind event by comparing the time variations of the path-averaged current with those of the wind, water temperature, and detided sea level. When the southward wind was subsiding and the sea level on the northern coast of Hiroshima Bay was recovering, the

CRediT authorship contribution statement

Naokazu Taniguchi: Conceptualization, Formal analysis, Writing - original draft. Yuji Sakuno: Funding acquisition, Visualization, Writing - review & editing. Hidemi Mutsuda: Funding acquisition, Conceptualization, Writing - review & editing. Masazumi Arai: Conceptualization, Formal analysis, Writing - review & editing.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgments

A part of the 2013 Hiroshima Bay CAT experiment data was downloaded from the institutional repository of Hiroshima University (http://ir.lib.hiroshima-u.ac.jp/en/00043744) made by Chen et al. [2], and the full acoustic transmission data were provided by Arata Kaneko of Hiroshima University. Wind and atmospheric pressure data were downloaded from the website of the Japan Meteorological Agency (http://www.data.jma.go.jp/gmd/risk/obsdl/), and sea level data were downloaded from the Japan

References (29)

  • C.-F. Huang et al.

    Mapping of ocean currents in shallow water using moving ship acoustic tomography

    J. Acoust. Soc. Am.

    (2019)
  • S.J. Lentz et al.

    The Wind- and Wave-Driven Inner-Shelf Circulation

    Annu. Rev. Mar. Sci.

    (2012)
  • J. Lin et al.

    Accurate imaging and prediction of Kanmon Strait tidal current structures by the coastal acoustic tomography data

    Geophys. Res. Lett.

    (2005)
  • W. Munk et al.

    Ocean acoustic tomography

    (1995)
  • Cited by (0)

    View full text