Temporal and spatial variations in three-dimensional seismic oceanography

Seismic oceanography is a new cross-discipline between geophysics and oceanography that uses seismic reflection data to image and study the oceanic water column. Previous work on seismic oceanography was largely limited to two-dimensional (2D) seismic data and methods. Here we explore and quantify temporal and spatial variations in three-dimensional (3D) seismic oceanography to address whether 3D seismic imaging is meaningful in all directions and how one can take advantage of the variations. From a 3D multichannel seismic survey acquired for oil and gas exploration in the Gulf of Mexico over a 6-month period, a 3D oceanic seismic volume was derived. The 3D seismic images exhibit both temporal and spatial variations of the ocean, and theoretical and data analyses were used to quantify their contribution. Our results suggest that temporal variation is more prominent in the crossline direction than in the inline direction, causing discontinuities in crossline images. However, a series of 3D inline images can be seen as snapshots of the water column at different times, capturing temporal variation of thermohaline structures induced by ocean dynamics. Our findings suggest the potential uses of marine 3D seismic data in studying time-evolving mesoscale ocean dynamics.

Geostatistical Seismic Inversion for Temperature and Salinity in the Madeira Abyssal Plain

A two-dimensional multichannel seismic reflection profile acquired in the Madeira Abyssal Plain during June 2016 was used in a modeling workflow comprising seismic oceanography processing, geostatistical inversion and Bayesian classification to predict the probability of occurrence of distinct water masses. The seismic section was processed to image in detail the fine scale structure of the water column using seismic oceanography. The processing sequence was developed to preserve, as much as possible, the relative seismic amplitudes of the data and enhance the shallow structure of the water column by effectively suppressing the direct arrival. The migrated seismic oceanography section shows an eddy at the expected Mediterranean Outflow Water depths, steeply dipping reflectors, which indicate the possible presence of frontal activity or secondary dipping eddy structures, and strong horizontal reflections between intermediate water masses suggestive of double diffuse processes. We then developed and applied an iterative geostatistical seismic oceanography inversion methodology to predict the spatial distribution of temperature and salinity. Due to the lack of contemporaneous direct measurements of temperature and salinity we used a global ocean model as spatial constraint during the inversion and nearby contemporaneous ARGO data to infer the expected statistical properties of both model parameters. After the inversion, Bayesian classification was applied to all temperature and salinity models inverted during the last iteration to predict the spatial distribution of three distinct water masses. A preliminary interpretation of these probabilistic models agrees with the expected ocean dynamics of the region.

Observations of fine structure changes in shoaling internal solitary waves based on seismic oceanography method

<p>In the study of shoaling internal solitary waves, the observation and research on the internal fine structure and the effect of the topography are still insufficient. We try to make up for such insufficiency by seismic oceanography method. A first-mode depression internal solitary wave was observed propagating on the continental slope in the northeast South China Sea near Dongsha Atoll. We used common offset gathers (COGs) to obtain a series of images of this internal solitary wave that evolved over time, and studied the changes in internal fine structure by analyzing the seismic events in COG migrated sections. We found that the seismic events were broken during the shoaling, which was caused by the instability induced by internal solitary wave. We picked six events which represent six waveform and analyzed their evolution. It was found that the change in shape of waveform at different depths is different. The waveform in deep water deforms before that in shallow water, and the waveform in shallow water deforms to a greater degree. In addition, we also counted four parameters of phase velocity, amplitude, wavelength, and slopes of front and rear during the shoaling. The results show that the phase velocity and amplitude of waveform in shallow water increases, the wavelength decreases, and the slope of rear gradually becomes larger than that of the front. We have compared the observed changes with previous study made by numerical simulation.</p>

Seismic oceanography study of a mode-2 internal solitary wave in the northeast South China Sea

<p>The nonlinear internal solitary waves (ISWs) are ubiquitous and recently many mode-1 ISWs have been reported to be detected in the northeast South China Sea by using the seismic oceanography method. However, few mode-2 ISWs are discovered in seismic data in the South China Sea. Thus, waveform characteristics and kinematics parameters of the mode-2 ISWs in this region need further study.</p><p>In this paper, one convex mode-2 ISW is presented near Dongsha Plateau on September 20th, 2009, and is analyzed by the combination of reprocessed seismic section and reanalysis hydrographic data. The seismic events of the multi-channel seismic section are extracted to obtain the vertical amplitude distribution and water depth of the mode-2 ISW. The seismic events can be used to analyze the structural characteristics in a snapshot, while different pre-stack common-offset gathers (COGs) can observe the seismic fine structures of the mode-2 ISW in chronological order. Furthermore, we use COGs method to calculate the apparent phase velocities of the peak and trough part of the mode-2 ISW on the seismic section and then correct the phase velocities according to the seismic measurement direction and ISWs propagation direction derived from satellite data. Theoretically, the reanalysis hydrographic data can be used to calculate the vertical structure and propagation speed of ISW based on the KdV model, and the theoretical results can be compared with those from seismic observations.</p><p>In total, 10 seismic events are extracted to obtain wave amplitudes and corresponding water depth distribution. Among the seismic events, only 2 events are elevation wave types and the rest 8 events are depression wave types. The maximum amplitude is about 25.5m of a depression wave event at 200m water depth. The dimensionless amplitude is 2.56, this number shows that the mode-2 ISW is of large amplitude. Moreover, the pycnocline is displaced over 20% from the mid-depth of the total seawater depth, illustrating the mode-2 ISW is of asymmetry. The fine structures of the mode-2 ISW observed on COGs also show the asymmetric and complex wave disturbance in different acquisition times. The apparent phase velocity of the crest is 1.59m/s, while the apparent phase velocity of the trough (the maximum amplitude) is 0.8065, the results indicate that the elevation waves of the mode-2 ISW may move faster than the underlying depression waves. Finally, the corrected phase speed of the mode-2 ISW is consistent with the propagation speed calculated by the KdV equation. More pieces of evidence are needed to explain the generation and to predict further evolution of the asymmetric mode-2 ISW, and seismic oceanography may be one of the key techniques to answer these questions.</p>

Temporal and spatial variations in three-dimensional seismic oceanography

Seismic oceanography is a new cross-discipline between geophysics and oceanography that uses seismic reflection data to image and study the ocean water column. Previous works on seismic oceanography were largely limited to two-dimensional seismic data and methods. We present a complete three-dimensional (3D) oceanic seismic study and explore its imaging capability in seismic oceanography. From a 3D multichannel seismic survey acquired for oil and gas exploration in the Gulf of Mexico over six months period, a 3D water-column seismic volume was derived. The 3D seismic volume exhibits both temporal and spatial variations of the ocean, and theoretical and empirical analyses were performed to discriminate their contribution. Our analyses shows that temporal variation is largely embedded along crossline direction, hence deteriorating the quality of seismic images; however, inline sections, which can be seen as snapshot representation of the water column, allow to capture not only thermohaline structure but also the temporal evolution of the ocean dynamics. Our finding highlights that appropriately processed and analyzed 3D seismic data not only provide superior images of the ocean structure but also can be useful for investigation of temporal evolution of mesoscale ocean dynamics.

Random noise attenuation of sparker seismic oceanography data with machine learning

Seismic oceanography (SO) acquires water column reflections using controlled source seismology and provides high lateral resolution that enables the tracking of the thermohaline structure of the oceans. Most SO studies obtain data using air guns, which can produce acoustic energy below 100 Hz bandwidth, with vertical resolution of approximately 10 m or more. For higher-frequency bands, with vertical resolution ranging from several centimeters to several meters, a smaller, low-cost seismic exploration system may be used, such as a sparker source with central frequencies of 250 Hz or higher. However, the sparker source has a relatively low energy compared to air guns and consequently produces data with a lower signal-to-noise (S∕N) ratio. To attenuate the random noise and extract reliable signal from the low S∕N ratio of sparker SO data without distorting the true shape and amplitude of water column reflections, we applied machine learning. Specifically, we used a denoising convolutional neural network (DnCNN) that efficiently suppresses random noise in a natural image. One of the most important factors of machine learning is the generation of an appropriate training dataset. We generated two different training datasets using synthetic and field data. Models trained with the different training datasets were applied to the test data, and the denoised results were quantitatively compared. To demonstrate the technique, the trained models were applied to an SO sparker seismic dataset acquired in the Ulleung Basin, East Sea (Sea of Japan), and the denoised seismic sections were evaluated. The results show that machine learning can successfully attenuate the random noise in sparker water column seismic reflection data.