Proceedings of ACOUSTICS 2005 9-11 November 2005, Busselton, Western Australia Australian Acoustical Society 467 Simulation of Time Series Received Underwater from Small Explosive Detonations in Shallow Ocean Regions Adrian D. Jones (1) , Alec J. Duncan (2) , Paul A. Clarke (3) and Amos Maggi (2) (1) Program Office, DSTO Edinburgh, SA 5111, Australia (2) Centre for Marine Science & Technology, Curtin University of Technology, Perth WA 6845, Australia (3) Maritime Operations Division, DSTO Edinburgh, SA 5111, Australia ABSTRACT The sound pressure time series received at medium ranges from small underwater explosives, known as "SUS" charges, have been under close study in recent years in relation to the potential impact of the use of such devices on marine fauna, in particular, marine mammals. Past work has centred on investigations of time series measured in shallow oceans in the Australian region. Here, at-sea measured data showed, consistently, received peak levels which were considerably less than published weak shock theory would have suggested. This paper shows the results from the analysis of an extended data set, which includes measurements of SUS signals received along a shallow ocean track in an additional ocean region. Further, this paper shows the results of simulations of the time series received along all these tracks. These simulations of received SUS waveforms, carried out at Curtin University, have been obtained by generating an inverse Fourier transform of the product of the oceanic transfer function and the Fourier transform of an input SUS waveform. The oceanic transfer function has been based on the use of the SCOOTER model at low frequencies and a ray model (BELLHOP gaussian beam ray model) at remaining frequencies. By simulating the received time series in this way, reasons for the discrepancies between measured peak data and expectations based on weak shock theory have been investigated and are presented in this paper. INTRODUCTION Underwater explosive devices have a long history of use as sonar signal sources by defence forces internationally (see, for example, Urick 1983 figure 1.3 and section 4.4). A number of standardised designs were developed for Signal Underwater Sound (SUS) explosive sources, these differing in the weight of the explosive charge and in the depth in the ocean at which detonation was pre-set to occur. The use of SUS as signal sources for anti-submarine warfare (ASW) has, for several decades, been replaced by the use of coherent sources, capable of generating tonal signals of extended duration. However, the SUS charge remains an attractive option as a signal source for characterising the acoustic parameters of a shallow ocean region (eg. transmission loss and reverberation (eg. Hall 1996)). Underwater explosives, such as SUS, are an anthropogenic noise source type which has potential to cause harm or annoyance to marine fauna. As the Australian Defence Force (ADF) wishes to conduct its maritime operations in an environmentally responsible manner, DSTO in conjunction with Curtin University, has been conducting research to understand relevant phenomena and establish essential principles. This paper shows recent results of this work, and provides an update to descriptions published previously by Jones and Clarke (2004, 2005). BACKGROUND TO PRESENT WORK Previous work (Jones and Clarke 2004, 2005) presented measurements of signals received from SUS for two shallow ocean tracks – Track A and Track B. It was shown that the broadband acoustic signal energy received at ranges from several km to about 10 km was in accord with expectations, whereas the broadband peak level received was much less than anticipated for a quiescent ocean of infinite extent. Here, the differences between theory and measurement exceeded 10 dB and approached 20 dB. The reason for this discrepancy was not known with certainty, but it was not considered to be due to a lack of data integrity. It was postulated that the reduced peak level might be due to one or more of the following: lack of coherence in transmission due to medium irregularities; lack of coherence of reflection due to boundary irregularities. A lack of coherence would be associated with time spreading, enhancing the possibility of pulse reduction. To further investigate the relevant phenomena, it was decided to (1) analyse pressure time series data for an additional ocean track, (2) simulate received time series for all measurement scenarios and (3) simulate effects of a small degree of loss of coherence in transmission. MEASURED DATA FOR TRACK C Underwater signals received from SUS charges have been obtained along a number of tracks within continental shelf waters in the Australian region. For each track, in-situ details include water temperature versus depth for at least one point along the track (from which sound speed versus depth has been obtained). All signals selected for analysis were examined for the presence of overload and any exceeding system criteria were rejected (Valentine Flint and Lawrence 1992). In addition, data were selected for study only if the peak level throughout the entire measured waveform was at least 5.5 dB less than the hard clipping limits of the recording system. This latter criterion was selected, as these requirements exceed the maximum possible amount by which the data sampling rate of 20 kHz might cause a peak from the SUS type in question to be underestimated. Here, the peak waveform has been assumed to be in the shape of an instantaneous rise followed by an exponential decay, with a time constant of 0.1 ms, this being applicable to ranges greater than several km. Using this criterion, data reported at 2.2 km on Track A (Jones and Clarke 2004) are now rejected.