Indian Journal of Geo-Marine Sciences Vol. 43(1), January 2014, pp. 17-21 Acoustic propagation affected by environmental parameters in coastal waters Sanjana M C, G Latha, A Thirunavukkarasu & G Raguraman National Institute of Ocean Technology, Velachery-Tambaram Road, Pallikaranai, Chennai-600 100, India. [E-mail: sanjana@niot.res.in] Received 13 July 2011; revised 26 March 2013 In coastal regions various factors affect the propagation of acoustic signals such as wind, tidal effects, off shore currents and even river outflows. Shallow water waveguide is also characterised by site-specific source nature, bathymetry, sediment properties and sound speed profile. Monterey-Miami Parabolic Equation (MMPE) model has been used to study the propagation in very shallow waters of Arabian Sea with respect to environmental parameters. Sound speed Profiles (SSP) and bottom sediment samples measured at the site characterize the water column and the ocean bottom respectively. Bottom is clay (soft bottom) which can lead to absorption of acoustic intensity into the sediment leading to a decrease in reflected acoustic rays. Under these conditions propagation to large distances will be associated with large losses in acoustic energy. Critical angle of propagation determined theoretically is found to be ±11.42 o with respect to horizontal representing low order trapped modes. SSP is downward refracting enhancing propagation along the bottom. In this paper Transmission Loss (TL) computation have been carried out for different frequencies with respect to range and depth. Signal loss and arrival structure and the travel time have been computed for different ranges from 0 to 6 km. Model results have been evaluated by comparing with field measurements and are found to be suitable for modelling extremely shallow water environments. [Keywords: Acoustic propagation, Sound speed profile, Transmission loss, Shallow water] Introduction The complex environment in coastal waters is manipulated by temporal and spatial fluctuations which can alter the existing sound speed profile which decides the propagation of sound. Propagation in shallow water is also influenced by the sediment bottom properties since multiple reflections takes place owing to the wave guide nature of the environment. River influx is also an entity which is variable across the year and can influence sites at very shallow waters. Modified Parabolic Equation (PE) models have been used successfully in such complex coastal environments to study propagation effects 1-3. Monterey-Miami Parabolic Equation Model (MMPE) have been used to relate environmental parameters to Transmission Loss (TL) in two extremely shallow water environments 4. This work examines the propagation for a 32 m water column with a downward refracting sound velocity profile and a clay sea bed with sandy sub bottom using Monterey-Miami Parabolic Equation model(mmpe) in the frequency band up to 5 khz. PE models are best suited for range dependent environments and also for broad frequency band as is being considered here. Critical angle of seabed has been determined from the measured bottom and water column properties and inputs from Hamilton model. Characteristics of near surface source propagation in terms of travel time and arrival structure and Transmission loss in db with respect to range and depth for different frequencies is investigated for a downward refracting environment. Materials and Methods Study site The environment considered is a sloping bottom shallow water wave guide with depth ranging from 16 m to 32 m and a range extent of 6 km, with clayey bottom (Fig.1a). Two major rivers Muvattupuzha and Periyar empty into the site which leads to variation in the distribution of water properties. Near to the shore the sound velocity is less and increases as we go offshore. Sound velocity profile is typically
18 INDIAN J MAR SCI VOL. 43(1), JANUARY 2014 speed profiles have been considered starting from the initial point and every 2 km apart in range. We are considering 1 m layer of clay and below that sandy layer. Sound speed in the sediment is 1596 m/s with a density of 1.26 g/cc with a bottom attenuation of 0.1 db/km/hz. A sub bottom layer with sound speed of 1705 m/s, density of 1.5 g/cc and attenuation of 0.1 db/km/hz is also considered. Source depth is considered at just below the surface and the frequency band considered is up to 5 khz. Wide angle source (approximates to a point source) is considered initially. To perform broadband analysis centre frequencies of 2520 Hz was used with a bandwidth of 5000 Hz. Results & Discussion The critical angle at this site determined from water column and bottom properties is èc=arccos(1538/1569)=11.42 o. Propagation is characterized by normal modes corresponding to waves striking the bottom at grazing angles lower than the critical angle <11.42 o. So waves striking the bottom at angles less than 11.42p will be totally internally reflected whereas rays with grazing angles greater than 11.42p are heavily attenuated on encountering the bottom. However in shallow water most of the energy that propagates is along the horizontal (Fig. 2). Fig. 1(a) Study site and (b) Sound speed profiles at different depths. downward refracting enhancing propagation along the bottom (Fig.1b). Sediment samples have been collected at the site and sieve analysis has been carried out. Bottom is mainly clay with admixture of sand leading to absorption of sound into the sediment. Shallow Water propagation modelling The second release of the Monterey-Miami Parabolic Equation (MMPE) propagation Model has been used to understand the shallow water propagation at this site. This model was developed by 5 based on Split Step Fourier (SSF) technique and since then many improvements have been made on this. It is a full wave underwater acoustic propagation model that utilizes the split step Fourier marching algorithm. Parabolic equation is popular for solving range dependent propagation problems. Total 4 sound Fig. 2 Reflection coefficient for the site. Understanding how sound is propagating at a site is important in assessing the ambient noise field and the contribution due to different sources. For a broad band point source at 0.1 m and 3 m depth from the water surface, at a water column depth of ~16 m, the output starting field data is given in Fig. 3. As the frequency increases the noise is restricted more and more towards the surface and maximum noise seen at the source depth for all frequencies. For frequencies above 2 khz, sound waves are mostly restricted to the upper 4 m and 6 m of the water column at the start of the simulation for source depths of 0.1 and 3 m respectively.
SANJANA et al.: ACOUSTIC PROPAGATION IN COASTAL WATERS... 19 Fig. 3 Output starting field data for a source beneath the surface at 3m and 0.1 m, in ~ 16 m water column. Transmission loss arrival structure: At 16 m water depth, for a near surface source, the arrival is crisp with no signal distortion with a sound speed fluctuation of only 1 m/s (Fig. 4). The transmission of the high frequency signal produced more multipath arrival structure and a weaker head wave due to more attenuation in the floor sediment. At 25 m depth, the sound speed gradient is 4.5 m/s and we can observe slight distortion in the reflected rays. In order to test the validity of the model, the 2.5 khz source responses Fig. 4 Transmission loss arrival structure for the source signal at 16 m, 25 m & different ranges.
20 INDIAN J MAR SCI VOL. 43(1), JANUARY 2014 were recomputed out to a range of 6 km. It is clear that the perturbation does impact the propagation at larger ranges. Travel time for the signal at different ranges for centre frequency of 2.5 khz is given in Fig. 4. At 0.5 km range ( 20 m depth) sound is seen to decay fast due to intense surface and bottom reflections compared to 6 km range (32 m depth). Since the waves are striking at much greater angles than the critical angle, energy is lost fast into the bottom. Transmission loss: The term transmission loss describes the drop in sound energy level as it propagates from one point in the ocean to another. It can be considered as the signal which is lost due to the sum of geometric spreading, refraction, interference and the loss due to attenuation in the ocean. Since the environment is shallow water, cylindrical spreading is assumed to dominate the propagation. In cylindrical spreading the intensity decreases linearly with increasing range. Transmission loss outputs from MMPE are in three different forms, TL at a single frequency versus range and depth, TL at a single range versus frequency and depth and TL at a single depth versus frequency and range. Transmission loss for different frequencies with respect to tange and depth are given in Fig.5. For 0.1 & 0.25 khz, there is more penetration into the sediment both surface and subsurface layers whereas for higher frequencies the penetration is only into the surface clay layer. Fig. 5 Transmission loss for different frequencies (a) with range at 15 m depth (b) with depth at 250 m range (c) with depth at 6 km range. Transmission loss for different ranges in the frequency band up to 5 khz is given in Fig. 6. At 0.1 km range the sound speed profile is almost is o-velocity Fig. 6 Transmission loss for 0.1 km range. with a gradient of 1 m/s between surface and bottom and the corresponding Fig. shows noise spread out along the depth with penetration into the bottom especially at low frequencies. Model Evaluation Difference techniques measure the distance between the model prediction and a standard (which can comprise field measurements or outputs from other models) in terms of db differences at a given range, or over a set range interval. These techniques are best suited to comparative model evaluations conducted in research environments 2. Model accuracy was evaluated by examining field measurements with respect to model output as a function of range and frequency (Fig.7). Conclusions The sound propagation cone for the shallow water channel is defined in terms of critical angle. The critical angle of propagation at the site is estimated
SANJANA et al.: ACOUSTIC PROPAGATION IN COASTAL WATERS... 21 Fig. 7 Comparison of model with field results. to be 11.42 o based on measurements and the source in such a waveguide propagates at an angle confined to a cone of 2*è c =22.8 o. The sound speed profile is downward refracting enhancing propagation along the bottom with a clay sediment layer which leads to loss in acoustic energy. Transmission loss analysis was done to illustrate the effects of shallow water variability on sound propagation. The signal arrival structure has been determined and signal distortion in terms of sound speed fluctuation examined. Signal distortion is observed for sound speed gradients at long ranges. On examining the TL for different frequencies over range and depth it shows that the loss with respect to range is more for low frequencies and decreases with increase in frequency. Transmission loss at short ranges clearly shows the influence of lossy bottom on acoustic propagation for low frequencies. From the different model runs, it is found that the model is able to incorporate water column variability, sediment properties, broad band sources etc and is suitable for modelling extremely shallow water environments. Model results have been compared with field measurements at the site and are found to be satisfactory. Acknowledgements Authors thank Director, NIOT for his support in carrying out this work. Prof. Gopu R Potty, University of Rhode Island, USA is gratefully acknowledged for his valuable guidance in running the model. References 1. Jensen F B Numerical models in Underwater acoustics, in Hybrid Formulation of Wave Propagation and scattering (Martinus Nijhoff, Dordrecht), (1984), pp. 295-335. 2. Etter P C, Underwater Acoustic modeling (E & FN SPON. London), (1996). 3. Smith K B, Convergence, stability and variability of shallow water acoustic predictions using a split-step Fourier Parabolic Equation model, J. of Comput. Acoust., 9 (2001), 243-285. 4. Miksis-Olds J L, & Miller J H. Transmission loss in Manatee habitats, J. Acoust. Soc. Am., 120(4) (2006), 2320-2327. 5. Smith, K B and Tappert, F D, UMPE: The University of Miami Parabolic Equation Model, Version 1.3, Marine Physical Laboratory, SIO Technical Memorandum No. 432, (1993).