Airgun Source Research Articles

The Eastern North American Margin Community Seismic Experiment: An Amphibious Active‐ and Passive‐Source Dataset

Abstract The eastern North American margin community seismic experiment (ENAM‐CSE) was conceived to target the ENAM Geodynamic Processes at Rifting and Subducting Margins (GeoPRISMS) primary site with a suite of both active‐ and passive‐source seismic data that would shed light on the processes associated with rift initiation and evolution. To fully understand the ENAM, it was necessary to acquire a seismic dataset that was both amphibious, spanning the passive margin from the continental interior onto the oceanic portion of the North American plate, and multiresolution, enabling imaging of the sediments, crust, and mantle lithosphere. The ENAM‐CSE datasets were collected on‐ and offshore of North Carolina and Virginia over a series of cruises and land‐based deployments between April 2014 and June 2015. The passive‐source component of the ENAM‐CSE included 30 broadband ocean‐bottom seismometers (OBSs) and 3 onshore broadband instruments. The broadband stations were deployed contemporaneously with those of the easternmost EarthScope Transportable Array creating a trans‐margin amphibious seismic dataset. The active‐source portion of the ENAM‐CSE included several components: (1) two onshore wide‐angle seismic profiles where explosive shots were recorded on closely spaced geophones; (2) four major offshore wide‐angle seismic profiles acquired with an airgun source and short‐period OBSs (SPOBSs), two of which were extended onland by deployments of short‐period seismometers; (3) marine multichannel seismic (MCS) data acquired along the four lines of SPOBSs and a series of other profiles along and across the margin. During the cruises, magnetic, gravity, and bathymetric data were also collected along all MCS profiles. All of the ENAM‐CSE products were made publicly available shortly after acquisition, ensuring unfettered community access to this unique dataset.

Assessing Short‐Term Clock Errors and Drifts of Temporary Seismic Networks Using the Active Airgun Source in Binchuan, Yunnan

ABSTRACT We conducted a short‐term airgun experiment at the Binchuan Fixed Airgun Signal Transmission Station from 14 to 20 February 2017, and two different types of seismometers (Güralp CMG‐40T and QS05A) recorded 62 airgun shots triggered under the same conditions. However, we observed significant clock errors and drifts in seismic data recorded by four QS05A seismometers. To assess the short‐term clock errors and drifts for seismometers, we propose a new method that measures the P‐wave arrival‐time differences between airgun signals recorded at a station pair, using the matched filter method. We find ∼1.0 s absolute clock errors for two Güralp CMG‐40T stations (CKT2 and 53261) and one QS05A station (STA05), as well as ∼0.5 s timing leaps for four QS05A stations (STA19, STA21, STA31, and STA33) during the experiment. Furthermore, all the QS05A seismometers exhibit clock drifts with similar linear trends. Additionally, we use teleseismic waveforms to verify the absolute clock errors for stations CKT2, 53261, and STA05. After double‐checking several possible factors, we determine the hardware failure or malfunctioning that may cause clock errors for the two types of seismometers.

Effects of the Rough Sea Surface on the Signature of a Single Air Gun

The calculation of air gun source signatures gives insight into applications such as air gun array design, deghosting and the impact of sound on marine life. Single air gun source signatures were calculated from the numerical solution of a set of differential equations based on different branches of physics. Some characteristic parameters of air guns were obtained from the Svein Vaage broadband air gun data set (SVBAD) measurements to calibrate the model. The comparison between measured and modeled air gun signals helps to investigate the accuracy of air gun source models. The modeled air gun signatures compared well with measurements from the SVBAD for the case of a calm sea surface of sea state 2 or less. The source ghost signal modeled for a rough sea surface showed amplitude and phase changes, affecting the ghost notches, which may explain discrepancies between the SVBAD measurements and modeled air gun signals at frequencies above 150 Hz.

Comparison of Propagation Models for the Characterization of Sound Pressure Fields

This paper aimed at comparing different propagation models for the calculation of sound pressure fields in the ocean as well as sound exposure levels. The chosen models are freely available and commonly used by the general underwater acoustics community when sound propagation losses need to be predicted in an ocean environment. The models used were Ocean Acoustics and Seismic Exploration Synthesis (OASES), RAM, Bellhop, and Kraken, which are all based on different approximations of the wave equation. The comparison exercise was carried out by calculating propagation loss, sound pressure waveforms, and sound exposure levels produced by a modeled air gun source. Calculations were performed for a simple range-independent problem of a Pekeris waveguide and produced insights into applicability of each model with regards to near- and far-fields, as well as low and high frequencies for example. Two of the models, RAM and Bellhop, which are suitable to range-dependent problems, were also used to solve a slightly complex case having a sloping bottom. It is hoped that the results presented here will be particularly useful to those who need to use propagation models for marine environmental and bioacoustics studies.

Guest Editorial Special Issue on Verification and Validation of Air Gun Source Signature and Sound Propagation Models

The papers in this special issue focuses on the verification and validation of models for predicting the sound field radiated by air guns and air gun arrays. This issue aims to address the demand for increasingly precise assessments of environmental impacts, requiring accurate characterization of the air gun array sound field, including the interface between source signature models and sound propagation models. To address this need, a modeling workshop was held in July 2016. Participants in the 2016 workshop were asked to predict, for specified scenarios, the time series of sound pressure and sound particle acceleration at different positions relative to the source, as well as the sourcewaveform. Air gun modeling experts worldwide, including the 2016 workshop participants, were invited to contribute to this special issue.

Parabolic Equation Modeling of a Seismic Airgun Array

The numerical modeling of seismic airgun sources, with a specific emphasis on the impact of sound on the marine ecosystem, is a challenging problem. In this paper, the simplified range-independent isovelocity problem, set up as part of the International Airgun Modelling Workshop, is solved using the parabolic equation (PE). The acoustic pressure and particle acceleration are computed including the arrival time series and source energy level in decidecade bands for ranges spanning from 30 m to 30 km. The particle acceleration is computed by taking the spatial first-order finite difference of the pressure on the PE computational grid (~1/16th wavelength). The two significant challenges this problem poses to the PE solution are the very large bandwidth (5-4500 Hz) and the computational accuracy required for time-series and particle acceleration computations. Each of the metrics outlined in the statement of the problem is computed and presented. In addition to computing the required fields, the 17-gun volumetric array is placed in a 3-D ocean environment in deep water in the Gulf of Mexico, with peak sound pressure computed out to 30 km and the band integrated sound exposure level (L <sub xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">E,p</sub> ) computed to 400 km.

Crustal-scale depth imaging via joint full-waveform inversion of ocean-bottom seismometer data and pre-stack depth migration of multichannel seismic data: a case study from the eastern Nankai Trough

Abstract. Imaging via pre-stack depth migration (PSDM) of reflection towed-streamer multichannel seismic (MCS) data at the scale of the whole crust is inherently difficult. This is because the depth penetration of the seismic wavefield is controlled, firstly, by the acquisition design, such as streamer length and air-gun source configuration, and secondly by the complexity of the crustal structure. Indeed, the limited length of the streamer makes the estimation of velocities from deep targets challenging due to the velocity–depth ambiguity. This problem is even more pronounced when processing 2-D seismic data due to the lack of multi-azimuthal coverage. Therefore, in order to broaden our knowledge about the deep crust using seismic methods, we present the development of specific imaging workflows that integrate different seismic data. Here we propose the combination of velocity model building using (i) first-arrival tomography (FAT) and full-waveform inversion (FWI) of wide-angle, long-offset data collected by stationary ocean-bottom seismometers (OBSs) and (ii) PSDM of short-spread towed-streamer MCS data for reflectivity imaging, with the former velocity model as a background model. We present an application of such a workflow to seismic data collected by the Japan Agency for Marine-Earth Science and Technology (JAMSTEC) and the Institut Français de Recherche pour l'Exploitation de la Mer (IFREMER) in the eastern Nankai Trough (Tokai area) during the 2000–2001 Seize France Japan (SFJ) experiment. We show that the FWI model, although derived from OBS data, provides an acceptable background velocity field for the PSDM of the MCS data. From the initial PSDM, we refine the FWI background velocity model by minimizing the residual move-outs (RMOs) picked in the pre-stack-migrated volume through slope tomography (ST), from which we generate a better-focused migrated image. Such integration of different seismic datasets and leading-edge imaging techniques led to greatly improved imaging at different scales. That is, large to intermediate crustal units identified in the high-resolution FWI velocity model extensively complement the short-wavelength reflectivity inferred from the MCS data to better constrain the structural factors controlling the geodynamics of the Nankai Trough.

Seismic on floating ice: data acquisition versus flexural wave noise

ABSTRACTGeophysical surveying of the Arctic will become increasingly important in future prospecting and monitoring of the terrestrial and adjacent areas in this hemisphere. Seismic data acquired on floating ice are hampered with extensive noise due to ice vibrations related to highly dispersive ice flexural waves generated by the seismic source. Several experiments have been conducted on floating ice in van Mijenfjorden in Svalbard in the Norwegian Arctic to specifically analyse the extent of flexural waves recorded with various seismic receivers and sources deployed both on top of ice and in the water below. The data show that flexural waves are severely damped at 5 m or deeper below the ice and hydrophone data suffer less from these vibrations compared with data recorded on the ice. Aliasing of single receiver hydrophone data can to some extent be suppressed using an in‐line line source of detonating cord. Experiments on ice on shallow water show prominent guided wave modes often referred to as Scholte waves propagating along the seabed. In this case, both flexural and Scholte waves interfere and make a complicated pattern of coherent noise. On shallow water, the positioning and type of the seismic source must be evaluated with respect to the coherent noise generated by these waves. Geophone strings of 25 m effectively suppress both flexural and Scholte waves due to their relative short wavelengths. An airgun generates relative more low‐frequency energy than a surface source of detonating cord. Accordingly, seismic mapping of deep seismic horizons seem to be best achieved using geophone strings of such length and an airgun source. For shallow targets, the use of hydrophones in combination with detonating cord is an appropriate solution. Seismic surveying in the Arctic always have to follow environmental restrictions of not disturbing or harming wildlife and not causing permanent footprints into the vulnerable tundra, which implies that the choice of seismic acquisition strategy might occur as a trade‐off between optimum data quality and environmental constraints.

What controls the initial peak of an air-gun source signature?

Seismic air guns are broadband sources that generate acoustic waves at many frequencies. The low-frequency waves can be used for imaging, whereas the high-frequency waves are attenuated and/or scattered before they can reflect from targets of interest in the subsurface. It is desirable to reduce the amplitude of the high-frequency acoustic waves because they are thought to be disruptive, and potentially damaging, to marine life and are not useful for geophysical purposes. The high-frequency acoustic waves are primarily associated with the initial expansion of the air-gun bubble and associated peak in the acoustic pressure time series, which is commonly referred to as the source signature of the air gun. We have developed a quasi-1D model of a seismic air gun coupled to a spherical bubble that accounts for gas dynamics and spatially variable depressurization inside the firing chamber to investigate controls on the initial peak of the source signature. The model is validated against data collected during field tests in Lake Seneca, New York. Simulations and field data show that the initial peak is primarily dependent on the operating pressure. A lower gun pressure results in a smaller peak amplitude and a slower rise time. The slope, the amplitude of the initial peak divided by the rise time, is used as a proxy for environmental impact and can decrease by as much as 50% when the air-gun pressure is reduced from 2000 to 1000 psi. The low frequencies are controlled by the total discharged mass, which is dependent upon the gun volume and pressure. Decreasing the operating pressure while simultaneously increasing the gun volume will reduce the high frequencies while maintaining the desirable low-frequency signals.

Characterisation of focused gas hydrate accumulations from the Pegasus Basin, New Zealand, using high-resolution and conventional seismic data

Gas hydrates are reported widely in seismic data from New Zealand’s Hikurangi Margin (east coast of the North Island). Over the last decade, conventional petroleum exploration interests in this region have led to the collection of several regional seismic datasets. These data have greatly improved our understanding of hydrate accumulations in the area; however, the resolution of industry multichannel seismic surveys is limited by the bandwidth of the airgun sources used. We present preliminary results from an academic high-resolution generator-injector (GI) airgun seismic survey, undertaken in mid 2015, that targeted focused gas hydrate accumulations lying within thrusted accretionary units in the Pegasus Basin, at the south end of the Hikurangi Margin. Each feature was surveyed with 5 to 10 closely spaced seismic lines that provide an opportunity to examine the three-dimensional structure and stratigraphy with better resolution than the original lines. Two main processes of natural gas migration and resulting accumulation as hydrate are examined more fully: (1) vertical transport into the shallow seafloor driven by overpressure and (2) inclined transport upward along dipping permeable beds. In both of these cases, our data show significant three-dimensional variability is needed to focus fluid migration from below into hydrate trapping configurations within the hydrate stability field nearer to the seafloor. Due to the absence of well data in this basin, the high-resolution seismic data also help to constrain interpretations of basin stratigraphy which plays a significant role in hosting and trapping hydrate accumulations.

Marine vibrators: the new phase of seismic exploration

ABSTRACTMarine seismic vibrators are generally considered to be less intrusive than airguns from an environmental perspective. This is because they emit their energy spread out in time, rather than in a single, high‐intensity pulse. There are also significant geophysical benefits associated with marine vibrators, and they stem from the ability to specify in detail the output acoustic waveform. The phase can be specified independently at each frequency. Such detailed control cannot be achieved with conventional airgun sources, where the phase can only be modified using simple overall time delays. The vibrator phase can be employed in several different ways: it can be applied to the overall source phase in a sequence so that it varies from one source point to the next; it can be applied to the individual vibrators within the source array so the source directivity is changed; it can be applied to the overall source phase of each source in a simultaneous source acquisition. Carefully designed phase sequences can attenuate the residual source noise, and this in turn allows extra source points to be interleaved between the conventional ones. For these extra source points, the relative phase of the vibrators within the array can be chosen to create a transverse gradient source, which illuminates the earth predominantly in directions out of the plane of the sail line without left/right ambiguity. If seismic vibrator data are acquired using interleaved conventional and transverse gradient sweeps, more information is collected per kilometre of vessel travel than is the case in conventional acquisition. This richer data acquisition leads to the possibility of acquiring all the necessary seismic data in a shorter time. Three‐dimensional reconstruction techniques are used to recover the same image quality that would have been obtained using the conventional, more time‐consuming acquisition. For a marine vibrator to be suitable for these techniques it must, in general terms, have ‘high fidelity’. The precise device specifications are defined through realistic end‐to‐end simulations of the physical systems and the processing. The specifications are somewhat more onerous than for a conventional vibrator, but they are achievable. A prototype vibrator that satisfies these requirements has been built. In a simulated case study of a three‐dimensional deep‐water ocean bottom node survey, the seismic data could have been acquired using marine vibrators in one third of the time that it would have taken using airguns.

Estimation of subbottom geoacoustic properties from offshore airgun surveys in Atlantic Canada

This paper estimates seabed geoacoustic profiles at 14 sites offshore of the eastern Canadian coast using sound levels from a single airgun. A 210 in3 airgun was operated along two track lines at each site and a calibrated seabed-mounted Autonomous Multichannel Acoustic Recorder equipped with a single omnidirectional hydrophone recorded the airgun sounds. High-resolution bathymetry was acquired along the tracks using a ship-mounted multibeam system and conductivity-temperature-depth profiles were measured at each site. Seabed geoacoustic profiles were estimated in a non-linear trans-dimensional Bayesian inversion using energy spectral density levels at six frequencies between 10 and 320 Hz as a function of range. The inversion estimated the unknown number of subbottom layers, geoacoustic properties (including compressional and shear wave speeds and attenuations, density, and layer thicknesses), airgun source levels, a range-correction factor, and error statistics. Data were fit with error standard deviations of 0.6–5.6 dB. Convergence of the posterior probability density function was not reached within the time limits of the study so parameter uncertainty was estimated using a linearized formulation of the inverse problem centered around the model that minimized the Bayesian Information Criterion.This paper estimates seabed geoacoustic profiles at 14 sites offshore of the eastern Canadian coast using sound levels from a single airgun. A 210 in3 airgun was operated along two track lines at each site and a calibrated seabed-mounted Autonomous Multichannel Acoustic Recorder equipped with a single omnidirectional hydrophone recorded the airgun sounds. High-resolution bathymetry was acquired along the tracks using a ship-mounted multibeam system and conductivity-temperature-depth profiles were measured at each site. Seabed geoacoustic profiles were estimated in a non-linear trans-dimensional Bayesian inversion using energy spectral density levels at six frequencies between 10 and 320 Hz as a function of range. The inversion estimated the unknown number of subbottom layers, geoacoustic properties (including compressional and shear wave speeds and attenuations, density, and layer thicknesses), airgun source levels, a range-correction factor, and error statistics. Data were fit with error standard deviati...

Parabolic equation modeling of a seismic airgun array

The numerical modelling of seismic airgun sources, with a specific emphasis on the impact of sound on the marine ecosystem, is a challenging problem. In this paper, the simplified range-independent iso-velocity problem set up as part if the International Airgun Modelling Workshop (IAMW), is solved using the Parabolic Equation (PE). The acoustic pressure and particle acceleration are computed including the arrival time series and source energy level in deci-decade bands for ranges spanning from 30m to 30 km. The particle acceleration is computed by taking the spatial gradient of the pressure on the PE computational grid. The two significant challenges this problem poses to the PE solution are the very large bandwidth (5-4500 Hz) and the computational accuracy required for time-series and particle acceleration computations. Each of the metrics outlined in the statement of the problem are computed and presented. In addition to computing the required fields, the 17-gun volumetric array is placed in a 3D ocean environment in deep water in the Gulf of Mexico, with peak pressure computed out to 30km and the band integrated Signal Energy Level (SEL) computed to 400 km.

Validation of airgun array modelled source signatures

Characterizing the source levels from seismic airgun arrays to increase confidence in model accuracy has been identified as a requirement by the scientific and regulatory community. To comply with regulatory requirements in Australia, a measurement program was conducted to validate the source signature predictions of JASCO’s Airgun Array Source Model. The validation program measured a range of commercial airgun arrays, including a 2380in3 array, and was conducted in 80m of water with array passing directly over the recorder on the seafloor. For the 2380in3 array, the maximum measured PK was 220.62 dBre1μPa. Due to the proximity of the measurements, interesting features such as ghost cavitation and airgun recharge were observed in the data. The measurement study results were used to validate the modelled far-field source levels through a comparison between the measured received sound-levels and predicted received sound-levels at a real receiver point in the far-field of the source. The predictions were made using a wavenumber integral model coupled to the airgun source model. The program measured received sound levels in the endfire, broadside and vertical directions, and as the results showed good agreement with the modelling results, provided a validation of the complete modelled source signatures for the array.

Source signature encoding of a vertical source array to separate the emitted direct wave from its ghost

Marine air-gun sources can be sequence-encoded by firing their individual elements independently over a short period of time. Using near-orthogonal firing sequences, whose crosscorrelation is minimal, as encoding sequences for multiple sets of air-gun sources, enables us to exploit their orthogonality as a separation feature. We find that, by distributing air guns over depths from 5 to 30 m, firing sequences can be designed whose direct, down-going wavefield is close to orthogonal to its source-ghost wavefield. The fundamentally new aspect of this approach is that the source-ghost signal is no longer just a time-delayed, opposite-polarity version of the down-going wavefield, but due to the different air-gun depths results in a different source sequence. This enables the consideration of the ghost wavefield as a separate source. We generate a set of such firing sequences by minimizing the crosscorrelation of these wavefields and optimizing their respective autocorrelations to achieve sharp peaks. The obtained, optimized firing sequences are then used for marine seismic source encoding. By adapting a multifrequency algorithm originally developed for simultaneous source separation, we determine that the ghost-source wavefield can be separated as a separate source from the direct, down-going wavefield.

Seismic imaging in the Krafla high-temperature geothermal field, NE Iceland, using zero- and far-offset vertical seismic profiling (VSP) data

Among geothermal exploration methods, active surface seismic methods have played only a minor role to date. Especially in high-temperature volcanic systems, reflection seismic data often reveal poor delineation of volcanic features, due to the internal heterogeneity of volcanic sequences. To enhance the vertical resolution, one possibility is the application of downhole seismic methods like vertical seismic profiling (VSP). A test experiment was carried out in the Krafla high-temperature geothermal field, NE-Iceland, to assess the ability of VSP to image subsurface structures, such as fractures, zones of high permeability, magmatic bodies, and zones of supercritical fluids and steam. Logging in such hostile environments is technical challenging in many aspects, but mainly due to the high temperature impact on the downhole electronic components of the measuring equipment. This requires a thorough pre-examination and implementation of the measurement, especially to avoid delays and tool failures. This paper presents results of zero- and far-offset VSP data from the K-18 borehole from within the Krafla caldera, which reveal good correlation with the surrounding lithology. The raw three-component seismic data display a good signal-to-noise ratio and dominant signal frequencies between 20 and 40 Hz, down to c. 2200 m depth, for air gun and explosive sources, respectively. A zero-offset source comparison was also conducted to assess the use of different impulsive sources for future VSP surveys in similar settings. By applying a standard VSP processing, we identified stratigraphic boundaries between lavas, hyaloclastites, and intrusions, which are in good agreement with existing well data. For the zero-offset VSP, both P- and S-wave velocity models were calculated and a depth-converted corridor stack was determined. In addition, multicomponent Kirchhoff depth migration and Fresnel volume migration were tested around the borehole. The 3D results are promising, but the specific shape and lateral extent of the reflectors could not be determined due to the restriction to only two sources and the insufficient spatial coverage (aperture). Our study demonstrates that vertical seismic profiles can clearly detect variations in the subsurface volcanic stratigraphy in high-temperature geothermal fields. A more detailed reservoir characterization can be achieved by further data integration, enhanced survey design including more source positions, and improved processing and imaging techniques, such as full-waveform inversion.

Special Issue on Verification and Validation of Airgun Source Signature and Sound Propagation Models

Special Issue on Verification and Validation of Airgun Source Signature and Sound Propagation Models

Analysis of sound propagation from the transitional area to deep water

Underwater acoustic propagation over the continental shelf and slope is complicated and is also an important issue. Motivated by a phenomenon in an experiment conducted in the South China Sea indicating that the energy of the received signal around the sound channel axis is significantly stronger than that at shallower depths, we study sound propagation from the transitional area to deep water. Numerical simulations with different source depths are first performed, from which we reach the following conclusions. When the source is placed near the sea surface, acoustic wave will be strongly attenuated by bottom losses in a range-independent deep-water environment, whereas it can propagate to a very long distance because of the continental slope. When the source is mounted on the slope bottom in shallow-water area, acoustic energy will be trapped near the sound channel axis, and it converges more evidently than the case where the source is located near the surface. Then, simulations with different source ranges are performed. By comparing the relative energy level in the vertical direction between the numerical and experimental results, the range of the unknown air-gun source is roughly determined. [Work supported by the National Natural Science Foundation of China under Grant Nos. 11434012 and 41561144006.]

Seismic airgun exploration of continental crust structures

The active sources generate seismic waves transmitting through the deep underground and can be used to image high-resolution subsurface structures. Therefore, an appropriate seismic source is the key factor to active source exploration. In order to study the structure of continental crust and its temporal variations, we selected an artificial seismic source generated from releasing air bubbles in water (airgun source hereinafter) out of a variety of artificial sources like the explosion, electronic sparkers, hammering, eccentric vibration, heavy-duty train vibration, and vibroseis etc. Airgun source is a new type of artificial source that have been proven to be environmentally friendly, safe, and highly efficient. Three Fixed Airgun Signal Transmission Stations (FASTS) have been built a few years ago in Yunnan, Xinjiang, and Gansu provinces in western China and have been continuously running for several years. Seismic waves generated by the airgun sources are highly reproducible and stacking of them can produce good seismograms on stations far away from the source, for instance, seismic waves produced in Xinjiang FASTS can be well recorded nearly 1300 km away after 5000 stacking, covering an area of 6 million km2 and penetrating down to a depth of about 60 km. Establishing about 10 FASTSs in China would enable long-term continuous underground monitoring of all 9.6 million km2 of land area. Treating airgun active sources as lanterns to illuminate subsurface structures, we can achieve the situation with “Beidou surveys from the sky and lantern illuminates underground”.

Determining the behavioural dose-response relationship of marine mammals to air gun noise and source proximity.

The effect of various anthropogenic sources of noise (e.g. sonar, seismic surveys) on the behaviour of marine mammals is sometimes quantified as a dose-response relationship, where the probability of an animal behaviourally 'responding' (e.g. avoiding the source) increases with 'dose' (or received level of noise). To do this, however, requires a definition of a 'significant' response (avoidance), which can be difficult to quantify. There is also the potential that the animal 'avoids' not only the source of noise but also the vessel operating the source, complicating the relationship. The proximity of the source is an important variable to consider in the response, yet difficult to account for given that received level and proximity are highly correlated. This study used the behavioural response of humpback whales to noise from two different air gun arrays (20 and 140 cubic inch air gun array) to determine whether a dose-response relationship existed. To do this, a measure of avoidance of the source was developed, and the magnitude (rather than probability) of this response was tested against dose. The proximity to the source, and the vessel itself, was included within the one-analysis model. Humpback whales were more likely to avoid the air gun arrays (but not the controls) within 3 km of the source at levels over 140 re. 1 µPa2s-1, meaning that both the proximity and the received level were important factors and the relationship between dose (received level) and response is not a simple one.