- Email: [email protected]
Measurement of tidal and residual currents in the Strait of Hormuz
Estuarine, Coastal and Shelf Science 178 (2016) 101e109
Contents lists available at ScienceDirect
Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss
Measurement of tidal and residual currents in the Strait of Hormuz Jafar Azizpour a, *, Seyed Mostafa Siadatmousavi b, Vahid Chegini a a b
Iranian National Institute for Oceanography and Atmospheric Science, No. 3, Etemadzadeh St., West Fatemi Ave., 1411813389, Tehran, Iran Iran University of Science and Technology, Narmak, 1684613114, Tehran, Iran
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 December 2015 Received in revised form 24 May 2016 Accepted 5 June 2016 Available online 7 June 2016
Quantifying the current in the Strait of Hormuz (SH) is vital for understanding the circulation in the Persian Gulf. To measure the current in the strait, four subsurface moorings were deployed at four different stations close to SH from early November 2012 to the end of January 2013. Tidal current were dominant in the SH. The tides in the SH were complex partially standing waves and the dominant pattern varied from being primarily semi-diurnal to diurnal. The phase difference between tidal constituents of current and sea level elevation time series was used as an index to show the partially progressive wave pattern inside the study area. At mooring positions 3 and 4, located to the left of SH, the phase differences were close to 160 and 100 , respectively. It indicates partially progressive waves in opposite direction at these stations. K1 and M2 were the two main constituents at all stations inside the study area. At surface, the magnitude of semi-major axis of ellipses for M2 constituent was larger than corresponding value for K1 whereas at the bottom layer, the opposite pattern was observed. The M2 rotary coefficients at mooring 1 illustrated that current vector at the bottom layer rotated in opposite direction compared to current vectors at the middle and surface layers. The rotation was counterclockwise in the bottom layer, while it was clockwise in the surface and middle layers. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Strait of Hormuz Tidal currents Residual currents Field measurements
1. Introduction The Persian Gulf (PG) is a shallow, semi-enclosed basin with a mean depth of approximately 40 m. The PG is connected to the deep Gulf of Oman through the narrow Strait of Hormuz (SH). It consists of a shallow and wide shelf along its southern part (depths of less than 50 m), and a deep northern part with a maximum depth of 100 m separated from the coast by a narrow shelf (Pous et al., 2015). Circulation in the PG is composed of two spatial scales: basin scale and meso-scale (Thoppil and Hogan, 2010). The circulation in the PG is primarily driven by the Shamal wind and heat fluxes, while thermohaline forcing, and tides exert secondary impacts on that. Although tide is important for moving and vertical mixing of water on a horizontal scale of order 10 km, it is not important in the large-scale residual circulation of the PG and SH. In terms of temporal scales, tide is important on periods of less than 24 h (Reynolds, 1993). Influence of tide in basin scale circulation of the PG is insignificant, except in the vicinity of SH and along its
* Corresponding author. E-mail addresses: [email protected] (J. Azizpour), [email protected] (S.M. Siadatmousavi), [email protected] (V. Chegini). http://dx.doi.org/10.1016/j.ecss.2016.06.004 0272-7714/© 2016 Elsevier Ltd. All rights reserved.
northern coast close to Iran (Blain, 1998; Pous et al., 2015). Over the short time scales, the strait circulation is dominated by tides (Reynolds, 1993). In the past few decades, some numerical studies have been carried out on currents and circulation in the PG and SH using finite €mpf and difference and finite element methods (Blain, 1998; Ka Sadrinasab, 2006; Lardner et al., 1982, 1993; Lardner and Das, 1991; Pous et al., 2012; Pous et al., 2015; Yao and Johns, 2010). In addition, several short period measurements have been conducted exclusively for the SH (Johns et al., 2003; Matsuyama et al., 1998; Pous et al., 2004) and some limited measurements exist in which SH was part of the basin-wide survey (Brewer and Dyrssen, 1985; Emery, 1956; Reynolds, 1993). This study presents the results of the first long time field measurements of currents in the Iranian part of the SH. The objective of this study is to present a more realistic view of the hydrodynamics close to SH using the measured currents at its northern part. Characteristics of tidal currents will be presented in terms of the tidal ellipse parameters, and the spectral analysis will be performed on the current time series to obtain the energy contents at tidal frequencies.
102
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
2. Measurements and methods
Table 1 Position and deployment period of moorings used in this study.
2.1. Field of study The SH is a relatively narrow waterway located at the mouth of the PG, with an average width of almost 56 km and the maximum depth of approximately 120 m in its southern part. At its narrowest section, the strait width is less than 39 km. It extends between 25.5 N and 27.3 N and between 55.5 E and 57 E (Fig. 1), and connects PG to the Gulf of Oman and Indian Ocean. It is one of the most important waterways in the world due to petroleum export of oil-rich countries around the PG. Iran borders the northern flank of SH, while on the south, the United Arab Emirates and Musandam, an exclave of Oman, are located. Almost 20% of the world’s petroleum, and approximately 35% of the petroleum traded by sea, passes through the SH. Close to the SH, there is a boundary between the extra-tropical weather system from the northwest, and the tropical weather system of the Arabian Sea and Indian Ocean from southeast, which are affected mainly by monsoon circulation. The southerly wind dominates in summer and north-west wind occurs frequently in winter (Reynolds, 1993).
Latitude Mooring Mooring Mooring Mooring
1 2 3 4
25 26 26 25
Longitude 0
53.807 N 0 48.595 N 0 26.394 N 0 56.391 N
57 56 55 55
0
6.652 E 0 37.812 E 0 48.307 E 0 2.390 E
Period 3.Nov.2012e5.Feb.2013 2.Nov.2012e29.Jan.2013 4.Nov.2012e30.Jan.2013 4.Nov.2012e31.Jan.2013
meters was close to the bed (~8 m above bottom), the next one was close to the mid depth (50 m from surface) and the last one was near the surface (25 m from surface). The 3 other moorings (2, 3 and 4) hosted two current meters, one close to the surface and another one close to the bed (50, 65 and 75 m of water depth, respectively). The surface current meters were located in depth between 25 and 32 m to avoid any interference with passing ships. The moorings locations were selected based on ship traffic zone, and also political and scientific considerations. All RCM9 were equipped with temperature, conductivity and pressure sensors. The subsurface moorings also contained temperature/depth sensors, which provided vertical profiles with a 10 m resolution. Sampling intervals for all instruments were set to 20 min. The moorings were deployed early in November 2012 and continued sampling for approximately 3 months (Table 1).
2.2. Observation 2.3. Data processing The in situ measurements of current and sea level fluctuations over the SH are rare and fragmentary. This study reports a longterm monitoring of currents (more than one month) which was conducted in the east, west and middle part of the SH in four subsurface mooring lines (Fig. 1). Mooring 1 included three RCM9 (AANDERAA) and was deployed at 110 m depth. One of the current
For all RCM9 data, quality control was carried out by removing spikes and bad data using phase-space method (Goring and Nikora, 2002). The missing data within the gaps lasting less than 3 h were linearly interpolated. Harmonic analysis was utilized to determine the tidal
Fig. 1. Study area, the moorings locations and the isobaths (GEBCO_08, 2010). The inset shows the location of SH with respect to the Persian Gulf and the Gulf of Oman.
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
constituents (Foreman, 1978). Using the MATLAB toolbox U_tide (Codiga, 2011) both horizontal components of current vector were employed for harmonic analysis simultaneously. The M2 partial tide was selected among the results to show the phase lag between water level and current time series. The M2 constituent was the most important tidal frequency over the entire study area (see Table 2). Along-shore and cross-shore components of water velocities were also computed. To carry out the projection, the principal component analysis (Preisendorfer, 1988) was implemented, and results are presented in Table 3. The phases of constituents of the tidal currents and the rotary coefficient were calculated based on Gonella (1972) outlined method to investigate the tidal current rotation along the water column. 3. Results and discussion 3.1. Observations According to Reynolds (1993), the three main components of flow through SH have been recognized including: (1) a tidal current, which is mainly barotropic (Matsuyama et al., 1998), (2) a barotropic sub inertial (low-frequency) component, which is driven by the atmospheric pressure fluctuations over the PG and Indian ocean Monsoon, and (3) a baroclinic/barotropic long-term component. The tidal current is highly variable over a few hour time scales while its vertical variations are negligible close to the strait. It suggests that the tidal current is predominately barotropic (Matsuyama et al., 1998). As mentioned in section 2.3, the U_tide toolbox was used to determine the contribution of tide in total flow regime in the study area (Codiga, 2011). Table 3 exhibits contributions of along-shore tidal current at each measured site. Contributions of tidal current were considerable at all stations and the maximum value was 88.91 percent, which took place at the uppermost current meter of mooring 4. Relatively lower values at mooring 2 compared to other moorings were resulted from the fact that the along-shore and cross-shore components of tidal current were of the same order at mooring 2 due to the geometry of the SH. Figs. 2e6 show along-shore and cross-shore components of current vector at mooring stations. Each panel consists of original current (raw data), tidal current, and residual current. The residual current or non-tidal current is defined as the difference between the time series of original current and tidal current (Boon, 2013). The residual current was much smaller than tidal current in the time series shown in Figs. 3e6. During spring tide, the current speed was higher in positive direction than in the negative one at mooring 1. In other words, the flood velocity was larger than ebb one during the spring tide. At neap tide, the ebb velocity was larger than flood velocity (see alongshore components of Figs. 2 and 3).
103
Table 3 Contributions of along-shore tidal currents in percent and the angle of axes rotation in degree for mooring sites. Mooring 1 Surface layer Middle layer Bottom layer Surface layer Middle layer Bottom layer
73.53 78.10 82.93 60.89 61.55 59.19
Mooring 2 69.78 e 75.16 27.02 e 39.45
Mooring 3
Mooring 4
86.79 e 85.66 30.82 e 30.29
88.91 e 83.25 9.28 e 7.60
The maximum quantities of along-shore current velocities for surface, middle and bottom layer at mooring 1 were 64.35, 74.59 and 60.67 cm/s, respectively. For cross-shore current, the maximum speeds were 28.74, 23.83 and 21.74 cm/s, respectively. The amplitudes of M2 and K1 constituents were dominant at all layers of mooring 1; the magnitude of the semi-major axis of the tidal ellipses for M2 (K1) constituent in the surface, middle and bottom layers were 17.5 (18.5), 17.6 (19.4), and 17.5 (10.4) cm/s, respectively (see Fig. 7). Tidal and residual currents of three layer of mooring 1 are shown in Fig. 3. First three upper panels (A, B, and C) and lower panels (D, E, and F) show along-shore and cross-shore components, respectively. The relative importance of residual currents in the water column increased from summer to winter, especially for along-shore currents. The residual current of the cross-shore current during neap tide was considerable. Fig. 4 illustrates along-shore and cross-shore currents at the mooring 2 position. Fig. 4A, and B shows raw data at surface and bottom layers, respectively. The intensity of along-shore and crossshore currents were rather the same. Incoming (out coming) water at this station bifurcated such that most of water entered (leaved) the PG from the southern part of the Qeshm Island (see Fig 1.), which is related to the along-shore components of the current vector. Part of water enters (leaves) the PG through the Khuran Strait on the north of the Qeshm Island that is related to the cross-shore components of the current vector. The 10 days low-pass filtered data confirm this pattern (not shown here). Contribution of cross-shore components of current vectors was substantial, reaching up to 33 and 52% of the magnitude of the corresponding along-shore components at surface and bottom current meters, respectively. Deviation of the main axis at surface and bottom were 27.0 and 39.5 , respectively. It is an indication of significant contributions of cross-shore bottom layer current in comparison with the surface layer. The maximum magnitudes of along-shore current velocities for the surface and bottom layers were 46.56, and 45.31 cm/s, respectively while the values for cross-shore currents were 29.79 and 34.0 cm/s, appropriately. Similar to mooring 1, the M2 and K1 constituents were dominant
Table 2 Amplitude and phase of main tidal constituents, and portion of constituent in tide for all moorings. Constituents Mooring 1
Mooring 2
Mooring 3
Mooring 4
Amplitude (m) Phase (deg.) Contribution (%) Amplitude (m) Phase (deg.) Contribution (%) Amplitude (m) Phase (deg.) Contribution (%) Amplitude (m) Phase (deg.) Contribution (%)
M2
K1
S2
O1
MK3
MSF
F
0.719 266.0 59.31 0.813 286.0 66.16 0.711 309.0 65.78 0.372 320.0 49.37
0.447 36.6 22.92 0.403 56.8 16.30 0.367 84.4 17.54 0.160 131.0 9.17
0.263 307.0 7.96 0.284 322.0 8.09 0.196 359.0 5.03 0.041 14.5 0.60
0.143 40.8 2.35 0.191 50.3 3.66 0.063 83.8 0.51 0.045 261.0 0.73
0.058 286.0 0.39 0.009 38.3 0.01 0.134 53.0 2.35 0.150 115.0 8.02
0.074 319.0 0.64 0.037 302.0 0.14 0.069 306.0 0.61 0.192 5.80 13.19
0.60
0.54
0.47
0.49
104
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
Fig. 2. Raw data time series of along-shore and cross-shore current at surface (A), middle (B) and bottom (C) layers of mooring 1.
Fig. 3. Tidal (black line) and residual (red line) time series of along-shore (A, B, and C) and cross-shore (D, E, and F) current at surface, middle and bottom layers of mooring 1. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
frequencies of the current. Along-shore and cross-shore tidal and residual currents are illustrated in Fig. 4CeF. Due to the geometry and relatively lower depth of postion of mooring 2, the magnitudes of tidal and residual currents were rather the same, particularly during neap tide. The magnitudes of the semi-major axis of the tidal ellipses for these constituents at the surface (bottom) layer were 9.46 (13.1), and 11.8 (12.0) cm/s, consequently (Fig. 7). Fig. 5 shows the current components at the surface and bottom layers of the mooring 3. Along-shore and cross-shore original data are shown in Fig. 5A, and B. Contributions of cross-shore current were
insignificant in this station and do not exceed 5% of the magnitude of along-shore components in entire water column. Magnitudes of ebb velocities (positive) were high in contrast to flood velocities during spring tide and vice versa. Asymmetries of original and tidal currents were clear for along-shore currents in the water column (see Fig. 5AeD). In the period of observations, 10 days low-passed filtered data (not shown here) showed density-driven inflow from the Indian Ocean Surface Water (IOSW). Only on 12 and 15 January 2013, current direction reversed noticeably, associated with the dense Persian Gulf Water (PGW). The maximum density
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
105
Fig. 4. Time series of along-shore (black line) and cross-shore (red line) current at surface and bottom of mooring 2; A and B are raw data; C and D are tidal current; E and F are residual current at the surface and bottom layers, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Time series of along-shore (black line) and cross-shore (red line) current at surface and bottom of mooring 3; A and B are raw data; C and D are tidal current; E and F are residual current at the surface and bottom layers, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
differences from the surface to the bed, reached 2 kg/m3 (Azizpour et al., 2014), which indicated enough density anomaly to drive currents. The maximum magnitudes of along-shore density-driven current velocities for the surface and bottom layers were 10.63 and 6.84 cm/s, correspondingly. The maximum quantities for alongshore (cross-shore) surface and bottom layers were 91.69 (81.44), and 39.59 (28.90) cm/s, respectively. Similar to moorings 1 and 2, contribution of tidal current was significant. Similar to the results
shown by Johns et al. (2003), the magnitudes of residual currents increased for along-shore currents from summer to winter. Variations of residual currents in the cross-shore currents were irregular but generally, from summer to winter, these variations increased. The magnitudes of the semi-major axis of the tidal ellipses for M2 and K1 constituents at the surface (bottom) layer were 26.0 (36.4) and 25.2 (26.2) cm/s, as a consequence (Fig. 7). The current components for mooring 4 are depicted in Fig. 6. Along-shore and cross-
106
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
Fig. 6. Time series of along-shore (black line) and cross-shore (red line) current at surface and bottom of mooring 4; A and B are raw data; C and D are tidal current; E and F are residual current at the surface and bottom layers, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
shore raw data are shown in two upper panels (Fig. 6A, and B). The residual current was significant in the cross-shore components (Fig. 6E, and 6F), especially at surface layer. In contrast, the tidal current components have the main contributions in the alongshore currents (Fig. 6C, and 6D). Ebb (to SH) current velocities magnitudes were much more in contrast flooding (to PG) velocities. The difference between the maximum ebb and flood velocities in the along-shore currents for surface and bottom layers were 40.50 and 14.89 cm/s, respectively. The maximum values of along-shore (cross-shore) current component for the surface and near bottom layers were 101.45 (86.81) and 80.89 62.84 cm/s, consequently. The magnitude of the semi-major axis of the tide ellipses for the M2 and K1 constituents at the surface (bottom) layer were 26.8 (25.8) and 28.4 (32.0) cm/s, appropriately (Fig. 7). Harmonic analysis of the current time series reveals several important constituents. Current ellipses for four main constituents (two semidiurnal, M2 and S2, and two diurnal, K1 and O1) are plotted for surface and bottom layers of mooring sites (Fig. 7). Altogether, these constituents represented more than 92% of tidal currents. Vertical structure of the currents indicates that the tidal currents were mainly barotropic. Barotropic properties of tidal currents are obvious from the time series of currents. Although ellipses shrinked toward the bottom (except mooring 2), magnitudes of semi-minor axes increased. This result shows that the relative importance of cross-shore currents increased toward the bed. Employing the amplitude ratio (Defant, 1961), ðAmp:O1 þAmp:K 1 Þ F ¼ ðAmp:M , it revealed that the tide in this area was a 2 þAmp:S 2 Þ mixed tide with semidiurnal tide dominated over the diurnal ones (see Table 2). The tide has a strong fortnightly (two-week) cycle of spring and neap tides, especially in the position of mooring 4. The dominant tidal constituent in most stations was the M2 tide. Table 2 shows the amplitude and phase of main constituents for all moorings. In general, the tidal amplitudes decreased uniformly from mooring 1 to 4. Tidal ranges for moorings 1 to 4 were 3.23, 3.48, 2.83 and 1.82 m, respectively. Tidal ranges of ebb tide at the moorings 1 and 2 were greater than flooding tide. In contrast, mean
tidal ranges of flooding tide at the moorings 3 and 4 were larger than ebb tide. It means that the magnitude of peak ebb exceeds that of the peak flood at mooring 1 and 2 locations; whereas at mooring 3 and 4, in which the magnitude of flood peak exceeds that of the ebb peak (asymmetry of tide). It should be noted that the tidal constituent MSF, with frequency of 0.0028 cph, is classified as long period tidal waves. Its amplitude was highly variable as moving toward semi-diurnal amphidromic point. It could be related to inability of harmonic analysis to distinguish between the astronomical tidal forcing and the non-tidal low-frequency meteorological signal (Ursella et al., 2014) or owing to the interaction between the low-frequency tidal and meteorological signals of the same order.
3.2. Current phases Tides in the PG enter from the Arabian Sea through the SH and propagate as Kelvin waves (Lehr, 1984). Tides move forward along the Iranian coast in the northern part and along Arabian countries coast in the southern part in a counterclockwise pattern. According to Defant (1961), the PG tidal waves have oscillation periods between 21.6 and 27 h. At the moorings locations, the phase difference between the M2 constituent of water level and vertically averaged tidal currents were calculated. The difference for moorings 1 to 4 were 60 , 45 , 160 and 100 , accordingly. The phase differences between water level and the currents were representative of partially progressive wave pattern (Walters et al., 1985) in moorings 1, 2 and 4. The difference of 160 deg, at the mooring 3 might be occurred due to topographic slope and density stratification. Ursella et al. (2014) also reported the phase difference between water level and currents around 270 in the Strait of Otranto. They argued that opposite energy flux direction confirmed this difference. The difference of 100 at mooring 4, is probably resulted from the PG dense water (stratification) and Abu Musa Island nonlinear effects and geometry (See Fig 1. For location of Abu Musa Island). Similar effects of stratification on tidal current were reported before by Howarth (1998) at Oyster Ground, North Sea.
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
107
Fig. 7. Current ellipses for the four major tidal constituents (M2, S2, K1, and O1) obtained by harmonic analysis of mooring sites measurements. Clockwise (counterclockwise) rotations are given by dashed (solid) lines.
108
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
Fig. 8. Rotary spectra for the moorings: A. mooring 1 at surface, middle and bottom layers, B. mooring 2 at surface, bottom layers, C. mooring 3 at surface, bottom layers, D. mooring 4 at surface, bottom layers.
Stratification and physical properties of water in the SH and Iranian part of the PG investigated by Azizpour et al. (2014) in more details. 3.3. Tidal analysis In this section, a rotary coefficient is used to describe the rotational direction and eccentricity of tidal ellipses. The rotary coefficient, as defined by Gonella (1972), can be expressed in terms of 1 A2 tidal ellipse parameters as CR ¼ 2A , where A1 and A2 are the A21 þA22 semi-major and semi-minor ellipse axes, respectively. Conventionally, the semi-minor axis gives a negative value for clockwise rotation of the tidal current vectors. Thus, the sign of CR is opposite to that of the semi-minor axis (by this convention). It is positive for clockwise rotation of the tidal vector, negative for counterclockwise rotation of the tidal vector, and zero for reversing tides. By defining the tidal ellipse parameters, CR was calculated for M2 constituent of measured currents at all mooring positions. Values for mooring 1 at surface, middle and bottom layers were 0.030, 0.042 and 0.097, correspondingly. These values showed that the rotations of water column as a resultant of M2 constituent at the bottom layer was counterclockwise, while at surface and middle layers were clockwise. At moorings 2 and 3 positions, entire water column rotation was counterclockwise (0.954, 0.968 at bottom and surface layers for mooring 2, and -0.142 and 0.091 for mooring 3, consequently). For mooring 4, the direction of rotation was different and water column rotation due to M2 was clockwise; i.e. CR values were 0.069 and 0.086 at bottom and surface layers, respectively. It is worthy to mention that the rotation of Cartesian coordinate to have alongshore and onshore axis did not change the tidal ellipse parameters sign. 3.4. Spectral analysis Fig. 8 presents the rotary spectra at different layers of the water column at different moorings locations. Using harmonic analysis (U_tide package), the tidal signal was decomposed into 35 constituents. Fig. 8 also confirms that the diurnal component K1 and semi-diurnal component M2 were the most energetic tidal constituents in the study area. The diurnal tidal band was more energetic than the semi-diurnal one in the bottom layer of all moorings. Moreover, at the position of mooring 1 the diurnal one (K1) was the main peak in the spectrum, where surface and mid layers show
particularly high energy at negative frequencies (indicating a clockwise rotation of the tidal vector). At the mooring 2 position, semi-diurnal component (M2) was more energetic in the positive frequencies at whole water column and has a primary peak in the spectrum, which corresponds to the counterclockwise rotation. Inertial oscillations, occurred at period of ~27 h, and shown as a negative peak frequency. At all the measurements period, the energy of inertial oscillations was less than corresponding values for the main tidal constituents. In addition, rotary spectra revealed other energetic semi-diurnal (S2 and N2) and diurnal (O1 and Q1) constituents. Some other mixed and non-linear constituents (with frequency of 0.122e0.242 cph), exist as secondary peaks in the spectra (e.g. M4, M6 and MK3). Comparing to diurnal and semidiurnal constituents energies, the secondary peaks were especially observed at moorings 2 and 4. Energy of fortnightly constituent (MFS) were considerable and this constituent revealed a peak in all spectra (Fig. 8). 4. Conclusion Results of current observation in the Strait of Hormuz revealed that tidal oscillation have the main contributions in the measured current throughout the water column. The time scale of the tidal current is short (less than 24 h). Partially progressive standing waves propagated in the area and the phase difference between currents and sea level fluctuations shows a lag in the sea level fluctuations. From mooring 1 to 4, the sea level fluctuations decreased gradually, while water current velocities (especially along-shore components) increased i.e. the opposite pattern in tidal fluctuation and tidal current were observed. The strongest currents were measured at mooring 4, which is close the location of an amphidromic point for semidiurnal harmonic constituents. Amphidromic points in the PG reported by several studies, for €mpf and Sadrinasab (2006) and Reynolds (1993). Acinstance Ka cording to Dronkers (1986), the flood asymmetry (long slow ebb, and short fast flood) was dominated at locations of mooring 1 and 2 while ebb asymmetry (long slow flood, short fast ebb) was dominated at mooring locations 3 and 4. At the moorings sites, the magnitude of semi-major axis of ellipses for M2 constituent at surface layer was larger than K1. In deeper layers, the magnitudes of semi-major axis of ellipses for diurnal constituent (K1) was larger than semi-diurnal ones. Fortnightly cycle has a secondary peak at
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
mooring positions together with mixed tidal constituents. Energy of main diurnal and semi-diurnal constituents were high in contrast other constituents. On the other hand, inertial oscillation energies in mooring’s stations were considerable. The spectral characteristics of the moorings show that the variations of currents were distributed mainly among the tidal oscillations, inertial oscillations and also long-period motion. Long period motion was generated by IOSW inflow from surface on the northern flank and PG dense water on the southern flank (mooring 4). Contributions of energy for diurnal and semi-diurnal tidal band at the surface and bottom layers were rather same. With decreasing depth, energy of non-linear tidal constituents (M4 and M6) was increased, especially in the bottom layer (Fig. 8 panel B and D). These are resultant of Islands and geometry effects located near the mooring lines. Due to the generally strong meteorologically induced sub-tidal flow, energy of surface layer was considerable in contrast to bottom layer for all mooring stations. Overall, smallest amount of energy were recognizable at mooring 2 station due to bifurcation of the flow. References Azizpour, J., Chegini, V., Khosravi, M., Einali, A., 2014. Study of the Physical Oceanographic Properties of the Persian Gulf, Strait of Hormuz and Gulf of Oman Based on PG-GOOS CTD Measurements. J. Persian Gulf Mar. Sci. 5, 37e48. Blain, C.A., 1998. Barotropic tidal and residual circulation in the Arabian Gulf. Estuar. Coast. Model. 166e180 (1997). ASCE. Boon, J.D., 2013. Secrets of the Tide: Tide and Tidal Current Analysis and Predictions, Storm Surges and Sea Level Trends. Elsevier. Brewer, P.G., Dyrssen, D., 1985. Chemical oceanography of the Persian Gulf. Essays on oceanography: a tribute to John Swallow. Prog. Oceanogr. 41e 55, 1e4. Codiga, D.L., 2011. Unified Tidal Analysis and Prediction Using the UTide Matlab Functions. Graduate School of Oceanography, University of Rhode Island Narragansett, RI, p. 60. Defant, A., 1961. Physical Oceanography, vol. 2. Dronkers, J.J., 1986. Tidal asymmetry and estuarine morphology. J. Sea Res. 20 (2/3), 117e131. Emery, K.O., 1956. Sediments and water of the Persian Gulf. Bull. Amer. Assoc. Petrol. Geol. 40, 2354e2383. Foreman, M.G.G., 1978. Manual for Tidal Currents Analysis and Prediction. Institute of Ocean Sciences, Patricia Bay, Sidney, p. 57. GEBCO_08, G, 2010. Version 20100927. British Oceanographic Data Centre (BODC). Accessed 30/01/2010, Available online. http://www.gebco.net.
109
Gonella, J., 1972. A rotary-component method for analysing meteorological and oceanographic vector time series. Deep Sea Res. Oceanogr. Abstr. 833e846. Elsevier. Goring, D.G., Nikora, V.I., 2002. Despiking acoustic Doppler velocimeter data. J. Hydraul. Eng. 128, 117e126. Howarth, M., 1998. The effect of stratification on tidal current profiles. Cont. Shelf Res. 18, 1235e1254. Johns, W.E., Yao, F., Olson, D.B., Josey, S.A., Grist, J.P., Smeed, D.A., 2003. Observations of seasonal exchange through the Straits of Hormuz and the inferred heat and freshwater budgets of the Persian Gulf. J. Geophys. Res. 108 (C12), 3391. €mpf, J., Sadrinasab, M., 2006. The circulation of the Persian Gulf: a numerical Ka study. Ocean Sci. 2, 27e41. Lardner, R., Das, S., 1991. On the computation of flows driven by density gradient: residual currents in the Arabian Gulf. Appl. Math. Model. 15, 282e294. Lardner, R., Belen, M., Cekirge, H., 1982. Finite difference model for tidal flows in the Arabian Gulf. Comput. Math. Appl. 8, 425e444. Lardner, R.W., Al-Rabeh, A.H., Gunay, N., Hossain, M., Reynolds, R.M., Lehr, W.J., 1993. Computation of the residual flow in the Gulf using the Mt Mitchell data and the KFUPM/RI hydrodynamical models. Mar. Pollut. Bull. 27, 61e70. Lehr, W.J., 1984. A brief survey of oceanographic modelling and oil spill studies in the KAP region. UNESCO report in marine sciences 28. In: El-Sabh, M.I. (Ed.), Oceanographic Modelling of the Kuwait Action Plan Region, pp. 4e11. Matsuyama, M., Kitade, Y., Senjyu, T., Koike, Y., Ishimaru, T., 1998. Vertical structure of a current and density front in the Strait of Hormuz. Offshore Environ. ROPME after War Relat. Oil-Spill 23e34. Pous, S., Carton, X., Lazure, P., 2004. Hydrology and circulation in the Strait of Hormuz and the Gulf of OmandResults from the GOGP99 Experiment: 1. Strait of Hormuz. J. Geophys. Res. Oceans 1978e2012, 109. Pous, S., Carton, X., Lazure, P., 2012. A process study of the tidal circulation in the Persian Gulf. Open J. Mar. Sci. 2, 131. Pous, S., Lazure, P., Carton, X., 2015. A model of the general circulation in the Persian Gulf and in the Strait of Hormuz: Intraseasonal to interannual variability. Cont. Shelf Res. 94, 55e70. Preisendorfer, R.W., 1988. Principal Component Analysis in Meteorology and Oceanography. Elsevier, Amsterdam. Reynolds, R.M., 1993. Physical oceanography of the Gulf, Strait of Hormuz, and the Gulf of OmandResults from the Mt Mitchell expedition. Mar. Pollut. Bull. 27, 35e59. Thoppil, P.G., Hogan, P.J., 2010. A modeling study of circulation and eddies in the Persian Gulf. J. Phys. Oceanogr. 40, 2122e2134. Ursella, L., Kova cevi c, V., Ga ci c, M., 2014. Tidal variability of the motion in the Strait of Otranto. Ocean Sci. 10, 49e67. Walters, R., Cheng, R., Conomos, T., 1985. Time scales of circulation and mixing processes of San Francisco Bay waters. Tempor. Dyn. estuary San Franc. Bay 13e36. Springer. Yao, F., Johns, W.E., 2010. A HYCOM modeling study of the Persian Gulf: 1. Model configurations and surface circulation. J. Geophys. Res. 115, C11017.
Contents lists available at ScienceDirect
Estuarine, Coastal and Shelf Science journal homepage: www.elsevier.com/locate/ecss
Measurement of tidal and residual currents in the Strait of Hormuz Jafar Azizpour a, *, Seyed Mostafa Siadatmousavi b, Vahid Chegini a a b
Iranian National Institute for Oceanography and Atmospheric Science, No. 3, Etemadzadeh St., West Fatemi Ave., 1411813389, Tehran, Iran Iran University of Science and Technology, Narmak, 1684613114, Tehran, Iran
a r t i c l e i n f o
a b s t r a c t
Article history: Received 3 December 2015 Received in revised form 24 May 2016 Accepted 5 June 2016 Available online 7 June 2016
Quantifying the current in the Strait of Hormuz (SH) is vital for understanding the circulation in the Persian Gulf. To measure the current in the strait, four subsurface moorings were deployed at four different stations close to SH from early November 2012 to the end of January 2013. Tidal current were dominant in the SH. The tides in the SH were complex partially standing waves and the dominant pattern varied from being primarily semi-diurnal to diurnal. The phase difference between tidal constituents of current and sea level elevation time series was used as an index to show the partially progressive wave pattern inside the study area. At mooring positions 3 and 4, located to the left of SH, the phase differences were close to 160 and 100 , respectively. It indicates partially progressive waves in opposite direction at these stations. K1 and M2 were the two main constituents at all stations inside the study area. At surface, the magnitude of semi-major axis of ellipses for M2 constituent was larger than corresponding value for K1 whereas at the bottom layer, the opposite pattern was observed. The M2 rotary coefficients at mooring 1 illustrated that current vector at the bottom layer rotated in opposite direction compared to current vectors at the middle and surface layers. The rotation was counterclockwise in the bottom layer, while it was clockwise in the surface and middle layers. © 2016 Elsevier Ltd. All rights reserved.
Keywords: Strait of Hormuz Tidal currents Residual currents Field measurements
1. Introduction The Persian Gulf (PG) is a shallow, semi-enclosed basin with a mean depth of approximately 40 m. The PG is connected to the deep Gulf of Oman through the narrow Strait of Hormuz (SH). It consists of a shallow and wide shelf along its southern part (depths of less than 50 m), and a deep northern part with a maximum depth of 100 m separated from the coast by a narrow shelf (Pous et al., 2015). Circulation in the PG is composed of two spatial scales: basin scale and meso-scale (Thoppil and Hogan, 2010). The circulation in the PG is primarily driven by the Shamal wind and heat fluxes, while thermohaline forcing, and tides exert secondary impacts on that. Although tide is important for moving and vertical mixing of water on a horizontal scale of order 10 km, it is not important in the large-scale residual circulation of the PG and SH. In terms of temporal scales, tide is important on periods of less than 24 h (Reynolds, 1993). Influence of tide in basin scale circulation of the PG is insignificant, except in the vicinity of SH and along its
* Corresponding author. E-mail addresses: [email protected] (J. Azizpour), [email protected] (S.M. Siadatmousavi), [email protected] (V. Chegini). http://dx.doi.org/10.1016/j.ecss.2016.06.004 0272-7714/© 2016 Elsevier Ltd. All rights reserved.
northern coast close to Iran (Blain, 1998; Pous et al., 2015). Over the short time scales, the strait circulation is dominated by tides (Reynolds, 1993). In the past few decades, some numerical studies have been carried out on currents and circulation in the PG and SH using finite €mpf and difference and finite element methods (Blain, 1998; Ka Sadrinasab, 2006; Lardner et al., 1982, 1993; Lardner and Das, 1991; Pous et al., 2012; Pous et al., 2015; Yao and Johns, 2010). In addition, several short period measurements have been conducted exclusively for the SH (Johns et al., 2003; Matsuyama et al., 1998; Pous et al., 2004) and some limited measurements exist in which SH was part of the basin-wide survey (Brewer and Dyrssen, 1985; Emery, 1956; Reynolds, 1993). This study presents the results of the first long time field measurements of currents in the Iranian part of the SH. The objective of this study is to present a more realistic view of the hydrodynamics close to SH using the measured currents at its northern part. Characteristics of tidal currents will be presented in terms of the tidal ellipse parameters, and the spectral analysis will be performed on the current time series to obtain the energy contents at tidal frequencies.
102
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
2. Measurements and methods
Table 1 Position and deployment period of moorings used in this study.
2.1. Field of study The SH is a relatively narrow waterway located at the mouth of the PG, with an average width of almost 56 km and the maximum depth of approximately 120 m in its southern part. At its narrowest section, the strait width is less than 39 km. It extends between 25.5 N and 27.3 N and between 55.5 E and 57 E (Fig. 1), and connects PG to the Gulf of Oman and Indian Ocean. It is one of the most important waterways in the world due to petroleum export of oil-rich countries around the PG. Iran borders the northern flank of SH, while on the south, the United Arab Emirates and Musandam, an exclave of Oman, are located. Almost 20% of the world’s petroleum, and approximately 35% of the petroleum traded by sea, passes through the SH. Close to the SH, there is a boundary between the extra-tropical weather system from the northwest, and the tropical weather system of the Arabian Sea and Indian Ocean from southeast, which are affected mainly by monsoon circulation. The southerly wind dominates in summer and north-west wind occurs frequently in winter (Reynolds, 1993).
Latitude Mooring Mooring Mooring Mooring
1 2 3 4
25 26 26 25
Longitude 0
53.807 N 0 48.595 N 0 26.394 N 0 56.391 N
57 56 55 55
0
6.652 E 0 37.812 E 0 48.307 E 0 2.390 E
Period 3.Nov.2012e5.Feb.2013 2.Nov.2012e29.Jan.2013 4.Nov.2012e30.Jan.2013 4.Nov.2012e31.Jan.2013
meters was close to the bed (~8 m above bottom), the next one was close to the mid depth (50 m from surface) and the last one was near the surface (25 m from surface). The 3 other moorings (2, 3 and 4) hosted two current meters, one close to the surface and another one close to the bed (50, 65 and 75 m of water depth, respectively). The surface current meters were located in depth between 25 and 32 m to avoid any interference with passing ships. The moorings locations were selected based on ship traffic zone, and also political and scientific considerations. All RCM9 were equipped with temperature, conductivity and pressure sensors. The subsurface moorings also contained temperature/depth sensors, which provided vertical profiles with a 10 m resolution. Sampling intervals for all instruments were set to 20 min. The moorings were deployed early in November 2012 and continued sampling for approximately 3 months (Table 1).
2.2. Observation 2.3. Data processing The in situ measurements of current and sea level fluctuations over the SH are rare and fragmentary. This study reports a longterm monitoring of currents (more than one month) which was conducted in the east, west and middle part of the SH in four subsurface mooring lines (Fig. 1). Mooring 1 included three RCM9 (AANDERAA) and was deployed at 110 m depth. One of the current
For all RCM9 data, quality control was carried out by removing spikes and bad data using phase-space method (Goring and Nikora, 2002). The missing data within the gaps lasting less than 3 h were linearly interpolated. Harmonic analysis was utilized to determine the tidal
Fig. 1. Study area, the moorings locations and the isobaths (GEBCO_08, 2010). The inset shows the location of SH with respect to the Persian Gulf and the Gulf of Oman.
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
constituents (Foreman, 1978). Using the MATLAB toolbox U_tide (Codiga, 2011) both horizontal components of current vector were employed for harmonic analysis simultaneously. The M2 partial tide was selected among the results to show the phase lag between water level and current time series. The M2 constituent was the most important tidal frequency over the entire study area (see Table 2). Along-shore and cross-shore components of water velocities were also computed. To carry out the projection, the principal component analysis (Preisendorfer, 1988) was implemented, and results are presented in Table 3. The phases of constituents of the tidal currents and the rotary coefficient were calculated based on Gonella (1972) outlined method to investigate the tidal current rotation along the water column. 3. Results and discussion 3.1. Observations According to Reynolds (1993), the three main components of flow through SH have been recognized including: (1) a tidal current, which is mainly barotropic (Matsuyama et al., 1998), (2) a barotropic sub inertial (low-frequency) component, which is driven by the atmospheric pressure fluctuations over the PG and Indian ocean Monsoon, and (3) a baroclinic/barotropic long-term component. The tidal current is highly variable over a few hour time scales while its vertical variations are negligible close to the strait. It suggests that the tidal current is predominately barotropic (Matsuyama et al., 1998). As mentioned in section 2.3, the U_tide toolbox was used to determine the contribution of tide in total flow regime in the study area (Codiga, 2011). Table 3 exhibits contributions of along-shore tidal current at each measured site. Contributions of tidal current were considerable at all stations and the maximum value was 88.91 percent, which took place at the uppermost current meter of mooring 4. Relatively lower values at mooring 2 compared to other moorings were resulted from the fact that the along-shore and cross-shore components of tidal current were of the same order at mooring 2 due to the geometry of the SH. Figs. 2e6 show along-shore and cross-shore components of current vector at mooring stations. Each panel consists of original current (raw data), tidal current, and residual current. The residual current or non-tidal current is defined as the difference between the time series of original current and tidal current (Boon, 2013). The residual current was much smaller than tidal current in the time series shown in Figs. 3e6. During spring tide, the current speed was higher in positive direction than in the negative one at mooring 1. In other words, the flood velocity was larger than ebb one during the spring tide. At neap tide, the ebb velocity was larger than flood velocity (see alongshore components of Figs. 2 and 3).
103
Table 3 Contributions of along-shore tidal currents in percent and the angle of axes rotation in degree for mooring sites. Mooring 1 Surface layer Middle layer Bottom layer Surface layer Middle layer Bottom layer
73.53 78.10 82.93 60.89 61.55 59.19
Mooring 2 69.78 e 75.16 27.02 e 39.45
Mooring 3
Mooring 4
86.79 e 85.66 30.82 e 30.29
88.91 e 83.25 9.28 e 7.60
The maximum quantities of along-shore current velocities for surface, middle and bottom layer at mooring 1 were 64.35, 74.59 and 60.67 cm/s, respectively. For cross-shore current, the maximum speeds were 28.74, 23.83 and 21.74 cm/s, respectively. The amplitudes of M2 and K1 constituents were dominant at all layers of mooring 1; the magnitude of the semi-major axis of the tidal ellipses for M2 (K1) constituent in the surface, middle and bottom layers were 17.5 (18.5), 17.6 (19.4), and 17.5 (10.4) cm/s, respectively (see Fig. 7). Tidal and residual currents of three layer of mooring 1 are shown in Fig. 3. First three upper panels (A, B, and C) and lower panels (D, E, and F) show along-shore and cross-shore components, respectively. The relative importance of residual currents in the water column increased from summer to winter, especially for along-shore currents. The residual current of the cross-shore current during neap tide was considerable. Fig. 4 illustrates along-shore and cross-shore currents at the mooring 2 position. Fig. 4A, and B shows raw data at surface and bottom layers, respectively. The intensity of along-shore and crossshore currents were rather the same. Incoming (out coming) water at this station bifurcated such that most of water entered (leaved) the PG from the southern part of the Qeshm Island (see Fig 1.), which is related to the along-shore components of the current vector. Part of water enters (leaves) the PG through the Khuran Strait on the north of the Qeshm Island that is related to the cross-shore components of the current vector. The 10 days low-pass filtered data confirm this pattern (not shown here). Contribution of cross-shore components of current vectors was substantial, reaching up to 33 and 52% of the magnitude of the corresponding along-shore components at surface and bottom current meters, respectively. Deviation of the main axis at surface and bottom were 27.0 and 39.5 , respectively. It is an indication of significant contributions of cross-shore bottom layer current in comparison with the surface layer. The maximum magnitudes of along-shore current velocities for the surface and bottom layers were 46.56, and 45.31 cm/s, respectively while the values for cross-shore currents were 29.79 and 34.0 cm/s, appropriately. Similar to mooring 1, the M2 and K1 constituents were dominant
Table 2 Amplitude and phase of main tidal constituents, and portion of constituent in tide for all moorings. Constituents Mooring 1
Mooring 2
Mooring 3
Mooring 4
Amplitude (m) Phase (deg.) Contribution (%) Amplitude (m) Phase (deg.) Contribution (%) Amplitude (m) Phase (deg.) Contribution (%) Amplitude (m) Phase (deg.) Contribution (%)
M2
K1
S2
O1
MK3
MSF
F
0.719 266.0 59.31 0.813 286.0 66.16 0.711 309.0 65.78 0.372 320.0 49.37
0.447 36.6 22.92 0.403 56.8 16.30 0.367 84.4 17.54 0.160 131.0 9.17
0.263 307.0 7.96 0.284 322.0 8.09 0.196 359.0 5.03 0.041 14.5 0.60
0.143 40.8 2.35 0.191 50.3 3.66 0.063 83.8 0.51 0.045 261.0 0.73
0.058 286.0 0.39 0.009 38.3 0.01 0.134 53.0 2.35 0.150 115.0 8.02
0.074 319.0 0.64 0.037 302.0 0.14 0.069 306.0 0.61 0.192 5.80 13.19
0.60
0.54
0.47
0.49
104
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
Fig. 2. Raw data time series of along-shore and cross-shore current at surface (A), middle (B) and bottom (C) layers of mooring 1.
Fig. 3. Tidal (black line) and residual (red line) time series of along-shore (A, B, and C) and cross-shore (D, E, and F) current at surface, middle and bottom layers of mooring 1. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
frequencies of the current. Along-shore and cross-shore tidal and residual currents are illustrated in Fig. 4CeF. Due to the geometry and relatively lower depth of postion of mooring 2, the magnitudes of tidal and residual currents were rather the same, particularly during neap tide. The magnitudes of the semi-major axis of the tidal ellipses for these constituents at the surface (bottom) layer were 9.46 (13.1), and 11.8 (12.0) cm/s, consequently (Fig. 7). Fig. 5 shows the current components at the surface and bottom layers of the mooring 3. Along-shore and cross-shore original data are shown in Fig. 5A, and B. Contributions of cross-shore current were
insignificant in this station and do not exceed 5% of the magnitude of along-shore components in entire water column. Magnitudes of ebb velocities (positive) were high in contrast to flood velocities during spring tide and vice versa. Asymmetries of original and tidal currents were clear for along-shore currents in the water column (see Fig. 5AeD). In the period of observations, 10 days low-passed filtered data (not shown here) showed density-driven inflow from the Indian Ocean Surface Water (IOSW). Only on 12 and 15 January 2013, current direction reversed noticeably, associated with the dense Persian Gulf Water (PGW). The maximum density
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
105
Fig. 4. Time series of along-shore (black line) and cross-shore (red line) current at surface and bottom of mooring 2; A and B are raw data; C and D are tidal current; E and F are residual current at the surface and bottom layers, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Fig. 5. Time series of along-shore (black line) and cross-shore (red line) current at surface and bottom of mooring 3; A and B are raw data; C and D are tidal current; E and F are residual current at the surface and bottom layers, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
differences from the surface to the bed, reached 2 kg/m3 (Azizpour et al., 2014), which indicated enough density anomaly to drive currents. The maximum magnitudes of along-shore density-driven current velocities for the surface and bottom layers were 10.63 and 6.84 cm/s, correspondingly. The maximum quantities for alongshore (cross-shore) surface and bottom layers were 91.69 (81.44), and 39.59 (28.90) cm/s, respectively. Similar to moorings 1 and 2, contribution of tidal current was significant. Similar to the results
shown by Johns et al. (2003), the magnitudes of residual currents increased for along-shore currents from summer to winter. Variations of residual currents in the cross-shore currents were irregular but generally, from summer to winter, these variations increased. The magnitudes of the semi-major axis of the tidal ellipses for M2 and K1 constituents at the surface (bottom) layer were 26.0 (36.4) and 25.2 (26.2) cm/s, as a consequence (Fig. 7). The current components for mooring 4 are depicted in Fig. 6. Along-shore and cross-
106
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
Fig. 6. Time series of along-shore (black line) and cross-shore (red line) current at surface and bottom of mooring 4; A and B are raw data; C and D are tidal current; E and F are residual current at the surface and bottom layers, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
shore raw data are shown in two upper panels (Fig. 6A, and B). The residual current was significant in the cross-shore components (Fig. 6E, and 6F), especially at surface layer. In contrast, the tidal current components have the main contributions in the alongshore currents (Fig. 6C, and 6D). Ebb (to SH) current velocities magnitudes were much more in contrast flooding (to PG) velocities. The difference between the maximum ebb and flood velocities in the along-shore currents for surface and bottom layers were 40.50 and 14.89 cm/s, respectively. The maximum values of along-shore (cross-shore) current component for the surface and near bottom layers were 101.45 (86.81) and 80.89 62.84 cm/s, consequently. The magnitude of the semi-major axis of the tide ellipses for the M2 and K1 constituents at the surface (bottom) layer were 26.8 (25.8) and 28.4 (32.0) cm/s, appropriately (Fig. 7). Harmonic analysis of the current time series reveals several important constituents. Current ellipses for four main constituents (two semidiurnal, M2 and S2, and two diurnal, K1 and O1) are plotted for surface and bottom layers of mooring sites (Fig. 7). Altogether, these constituents represented more than 92% of tidal currents. Vertical structure of the currents indicates that the tidal currents were mainly barotropic. Barotropic properties of tidal currents are obvious from the time series of currents. Although ellipses shrinked toward the bottom (except mooring 2), magnitudes of semi-minor axes increased. This result shows that the relative importance of cross-shore currents increased toward the bed. Employing the amplitude ratio (Defant, 1961), ðAmp:O1 þAmp:K 1 Þ F ¼ ðAmp:M , it revealed that the tide in this area was a 2 þAmp:S 2 Þ mixed tide with semidiurnal tide dominated over the diurnal ones (see Table 2). The tide has a strong fortnightly (two-week) cycle of spring and neap tides, especially in the position of mooring 4. The dominant tidal constituent in most stations was the M2 tide. Table 2 shows the amplitude and phase of main constituents for all moorings. In general, the tidal amplitudes decreased uniformly from mooring 1 to 4. Tidal ranges for moorings 1 to 4 were 3.23, 3.48, 2.83 and 1.82 m, respectively. Tidal ranges of ebb tide at the moorings 1 and 2 were greater than flooding tide. In contrast, mean
tidal ranges of flooding tide at the moorings 3 and 4 were larger than ebb tide. It means that the magnitude of peak ebb exceeds that of the peak flood at mooring 1 and 2 locations; whereas at mooring 3 and 4, in which the magnitude of flood peak exceeds that of the ebb peak (asymmetry of tide). It should be noted that the tidal constituent MSF, with frequency of 0.0028 cph, is classified as long period tidal waves. Its amplitude was highly variable as moving toward semi-diurnal amphidromic point. It could be related to inability of harmonic analysis to distinguish between the astronomical tidal forcing and the non-tidal low-frequency meteorological signal (Ursella et al., 2014) or owing to the interaction between the low-frequency tidal and meteorological signals of the same order.
3.2. Current phases Tides in the PG enter from the Arabian Sea through the SH and propagate as Kelvin waves (Lehr, 1984). Tides move forward along the Iranian coast in the northern part and along Arabian countries coast in the southern part in a counterclockwise pattern. According to Defant (1961), the PG tidal waves have oscillation periods between 21.6 and 27 h. At the moorings locations, the phase difference between the M2 constituent of water level and vertically averaged tidal currents were calculated. The difference for moorings 1 to 4 were 60 , 45 , 160 and 100 , accordingly. The phase differences between water level and the currents were representative of partially progressive wave pattern (Walters et al., 1985) in moorings 1, 2 and 4. The difference of 160 deg, at the mooring 3 might be occurred due to topographic slope and density stratification. Ursella et al. (2014) also reported the phase difference between water level and currents around 270 in the Strait of Otranto. They argued that opposite energy flux direction confirmed this difference. The difference of 100 at mooring 4, is probably resulted from the PG dense water (stratification) and Abu Musa Island nonlinear effects and geometry (See Fig 1. For location of Abu Musa Island). Similar effects of stratification on tidal current were reported before by Howarth (1998) at Oyster Ground, North Sea.
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
107
Fig. 7. Current ellipses for the four major tidal constituents (M2, S2, K1, and O1) obtained by harmonic analysis of mooring sites measurements. Clockwise (counterclockwise) rotations are given by dashed (solid) lines.
108
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
Fig. 8. Rotary spectra for the moorings: A. mooring 1 at surface, middle and bottom layers, B. mooring 2 at surface, bottom layers, C. mooring 3 at surface, bottom layers, D. mooring 4 at surface, bottom layers.
Stratification and physical properties of water in the SH and Iranian part of the PG investigated by Azizpour et al. (2014) in more details. 3.3. Tidal analysis In this section, a rotary coefficient is used to describe the rotational direction and eccentricity of tidal ellipses. The rotary coefficient, as defined by Gonella (1972), can be expressed in terms of 1 A2 tidal ellipse parameters as CR ¼ 2A , where A1 and A2 are the A21 þA22 semi-major and semi-minor ellipse axes, respectively. Conventionally, the semi-minor axis gives a negative value for clockwise rotation of the tidal current vectors. Thus, the sign of CR is opposite to that of the semi-minor axis (by this convention). It is positive for clockwise rotation of the tidal vector, negative for counterclockwise rotation of the tidal vector, and zero for reversing tides. By defining the tidal ellipse parameters, CR was calculated for M2 constituent of measured currents at all mooring positions. Values for mooring 1 at surface, middle and bottom layers were 0.030, 0.042 and 0.097, correspondingly. These values showed that the rotations of water column as a resultant of M2 constituent at the bottom layer was counterclockwise, while at surface and middle layers were clockwise. At moorings 2 and 3 positions, entire water column rotation was counterclockwise (0.954, 0.968 at bottom and surface layers for mooring 2, and -0.142 and 0.091 for mooring 3, consequently). For mooring 4, the direction of rotation was different and water column rotation due to M2 was clockwise; i.e. CR values were 0.069 and 0.086 at bottom and surface layers, respectively. It is worthy to mention that the rotation of Cartesian coordinate to have alongshore and onshore axis did not change the tidal ellipse parameters sign. 3.4. Spectral analysis Fig. 8 presents the rotary spectra at different layers of the water column at different moorings locations. Using harmonic analysis (U_tide package), the tidal signal was decomposed into 35 constituents. Fig. 8 also confirms that the diurnal component K1 and semi-diurnal component M2 were the most energetic tidal constituents in the study area. The diurnal tidal band was more energetic than the semi-diurnal one in the bottom layer of all moorings. Moreover, at the position of mooring 1 the diurnal one (K1) was the main peak in the spectrum, where surface and mid layers show
particularly high energy at negative frequencies (indicating a clockwise rotation of the tidal vector). At the mooring 2 position, semi-diurnal component (M2) was more energetic in the positive frequencies at whole water column and has a primary peak in the spectrum, which corresponds to the counterclockwise rotation. Inertial oscillations, occurred at period of ~27 h, and shown as a negative peak frequency. At all the measurements period, the energy of inertial oscillations was less than corresponding values for the main tidal constituents. In addition, rotary spectra revealed other energetic semi-diurnal (S2 and N2) and diurnal (O1 and Q1) constituents. Some other mixed and non-linear constituents (with frequency of 0.122e0.242 cph), exist as secondary peaks in the spectra (e.g. M4, M6 and MK3). Comparing to diurnal and semidiurnal constituents energies, the secondary peaks were especially observed at moorings 2 and 4. Energy of fortnightly constituent (MFS) were considerable and this constituent revealed a peak in all spectra (Fig. 8). 4. Conclusion Results of current observation in the Strait of Hormuz revealed that tidal oscillation have the main contributions in the measured current throughout the water column. The time scale of the tidal current is short (less than 24 h). Partially progressive standing waves propagated in the area and the phase difference between currents and sea level fluctuations shows a lag in the sea level fluctuations. From mooring 1 to 4, the sea level fluctuations decreased gradually, while water current velocities (especially along-shore components) increased i.e. the opposite pattern in tidal fluctuation and tidal current were observed. The strongest currents were measured at mooring 4, which is close the location of an amphidromic point for semidiurnal harmonic constituents. Amphidromic points in the PG reported by several studies, for €mpf and Sadrinasab (2006) and Reynolds (1993). Acinstance Ka cording to Dronkers (1986), the flood asymmetry (long slow ebb, and short fast flood) was dominated at locations of mooring 1 and 2 while ebb asymmetry (long slow flood, short fast ebb) was dominated at mooring locations 3 and 4. At the moorings sites, the magnitude of semi-major axis of ellipses for M2 constituent at surface layer was larger than K1. In deeper layers, the magnitudes of semi-major axis of ellipses for diurnal constituent (K1) was larger than semi-diurnal ones. Fortnightly cycle has a secondary peak at
J. Azizpour et al. / Estuarine, Coastal and Shelf Science 178 (2016) 101e109
mooring positions together with mixed tidal constituents. Energy of main diurnal and semi-diurnal constituents were high in contrast other constituents. On the other hand, inertial oscillation energies in mooring’s stations were considerable. The spectral characteristics of the moorings show that the variations of currents were distributed mainly among the tidal oscillations, inertial oscillations and also long-period motion. Long period motion was generated by IOSW inflow from surface on the northern flank and PG dense water on the southern flank (mooring 4). Contributions of energy for diurnal and semi-diurnal tidal band at the surface and bottom layers were rather same. With decreasing depth, energy of non-linear tidal constituents (M4 and M6) was increased, especially in the bottom layer (Fig. 8 panel B and D). These are resultant of Islands and geometry effects located near the mooring lines. Due to the generally strong meteorologically induced sub-tidal flow, energy of surface layer was considerable in contrast to bottom layer for all mooring stations. Overall, smallest amount of energy were recognizable at mooring 2 station due to bifurcation of the flow. References Azizpour, J., Chegini, V., Khosravi, M., Einali, A., 2014. Study of the Physical Oceanographic Properties of the Persian Gulf, Strait of Hormuz and Gulf of Oman Based on PG-GOOS CTD Measurements. J. Persian Gulf Mar. Sci. 5, 37e48. Blain, C.A., 1998. Barotropic tidal and residual circulation in the Arabian Gulf. Estuar. Coast. Model. 166e180 (1997). ASCE. Boon, J.D., 2013. Secrets of the Tide: Tide and Tidal Current Analysis and Predictions, Storm Surges and Sea Level Trends. Elsevier. Brewer, P.G., Dyrssen, D., 1985. Chemical oceanography of the Persian Gulf. Essays on oceanography: a tribute to John Swallow. Prog. Oceanogr. 41e 55, 1e4. Codiga, D.L., 2011. Unified Tidal Analysis and Prediction Using the UTide Matlab Functions. Graduate School of Oceanography, University of Rhode Island Narragansett, RI, p. 60. Defant, A., 1961. Physical Oceanography, vol. 2. Dronkers, J.J., 1986. Tidal asymmetry and estuarine morphology. J. Sea Res. 20 (2/3), 117e131. Emery, K.O., 1956. Sediments and water of the Persian Gulf. Bull. Amer. Assoc. Petrol. Geol. 40, 2354e2383. Foreman, M.G.G., 1978. Manual for Tidal Currents Analysis and Prediction. Institute of Ocean Sciences, Patricia Bay, Sidney, p. 57. GEBCO_08, G, 2010. Version 20100927. British Oceanographic Data Centre (BODC). Accessed 30/01/2010, Available online. http://www.gebco.net.
109
Gonella, J., 1972. A rotary-component method for analysing meteorological and oceanographic vector time series. Deep Sea Res. Oceanogr. Abstr. 833e846. Elsevier. Goring, D.G., Nikora, V.I., 2002. Despiking acoustic Doppler velocimeter data. J. Hydraul. Eng. 128, 117e126. Howarth, M., 1998. The effect of stratification on tidal current profiles. Cont. Shelf Res. 18, 1235e1254. Johns, W.E., Yao, F., Olson, D.B., Josey, S.A., Grist, J.P., Smeed, D.A., 2003. Observations of seasonal exchange through the Straits of Hormuz and the inferred heat and freshwater budgets of the Persian Gulf. J. Geophys. Res. 108 (C12), 3391. €mpf, J., Sadrinasab, M., 2006. The circulation of the Persian Gulf: a numerical Ka study. Ocean Sci. 2, 27e41. Lardner, R., Das, S., 1991. On the computation of flows driven by density gradient: residual currents in the Arabian Gulf. Appl. Math. Model. 15, 282e294. Lardner, R., Belen, M., Cekirge, H., 1982. Finite difference model for tidal flows in the Arabian Gulf. Comput. Math. Appl. 8, 425e444. Lardner, R.W., Al-Rabeh, A.H., Gunay, N., Hossain, M., Reynolds, R.M., Lehr, W.J., 1993. Computation of the residual flow in the Gulf using the Mt Mitchell data and the KFUPM/RI hydrodynamical models. Mar. Pollut. Bull. 27, 61e70. Lehr, W.J., 1984. A brief survey of oceanographic modelling and oil spill studies in the KAP region. UNESCO report in marine sciences 28. In: El-Sabh, M.I. (Ed.), Oceanographic Modelling of the Kuwait Action Plan Region, pp. 4e11. Matsuyama, M., Kitade, Y., Senjyu, T., Koike, Y., Ishimaru, T., 1998. Vertical structure of a current and density front in the Strait of Hormuz. Offshore Environ. ROPME after War Relat. Oil-Spill 23e34. Pous, S., Carton, X., Lazure, P., 2004. Hydrology and circulation in the Strait of Hormuz and the Gulf of OmandResults from the GOGP99 Experiment: 1. Strait of Hormuz. J. Geophys. Res. Oceans 1978e2012, 109. Pous, S., Carton, X., Lazure, P., 2012. A process study of the tidal circulation in the Persian Gulf. Open J. Mar. Sci. 2, 131. Pous, S., Lazure, P., Carton, X., 2015. A model of the general circulation in the Persian Gulf and in the Strait of Hormuz: Intraseasonal to interannual variability. Cont. Shelf Res. 94, 55e70. Preisendorfer, R.W., 1988. Principal Component Analysis in Meteorology and Oceanography. Elsevier, Amsterdam. Reynolds, R.M., 1993. Physical oceanography of the Gulf, Strait of Hormuz, and the Gulf of OmandResults from the Mt Mitchell expedition. Mar. Pollut. Bull. 27, 35e59. Thoppil, P.G., Hogan, P.J., 2010. A modeling study of circulation and eddies in the Persian Gulf. J. Phys. Oceanogr. 40, 2122e2134. Ursella, L., Kova cevi c, V., Ga ci c, M., 2014. Tidal variability of the motion in the Strait of Otranto. Ocean Sci. 10, 49e67. Walters, R., Cheng, R., Conomos, T., 1985. Time scales of circulation and mixing processes of San Francisco Bay waters. Tempor. Dyn. estuary San Franc. Bay 13e36. Springer. Yao, F., Johns, W.E., 2010. A HYCOM modeling study of the Persian Gulf: 1. Model configurations and surface circulation. J. Geophys. Res. 115, C11017.