First Break - Volume 40, Issue 9, 2022

Towed-streamer data are often contaminated by low-frequency, high-amplitude noise caused by the streamer motion, cable depth controllers, or sea surface waves. So-called swell-noise usually affects a number of neighbouring traces and appears on the shot record as vertical striping. Attenuating this strong noise over a weak reflection signal can be a significant challenge. In this work, we describe a deep learning approach for estimating and subtracting such noise from the recorded data. Inspired by the ideas of the residual network and the generative adversarial network, we have developed a conditional generative adversarial network to estimate swell noise which could also be applicable to other types of noise found in seismic data. We demonstrate the effectiveness of the proposed network by estimating a high-quality swell noise model on a field data example.

Borehole and surface joint seismic exploration is a 3D seismic exploration method formed by the combination of surface seismic and VSP survey simultaneously. Using the 3D DAS-VSP data, accurate time-depth relationship, formation velocity, deconvolution operator, spherical diffusion compensation factor, absorption attenuation factor, anisotropy parameters and high-resolution structural imaging around the wellbore can be obtained. These parameters can be used to significantly enhance the surface 3D seismic data processing.

This paper describes the 3D DAS-VSP data acquired by a downhole armoured optical cable simultaneously with an OBN data acquisition project in East China Sea, and the results of 3D DAS-VSP data imaging processing. The usual 3D VSP data imaging processing steps include: observation system definition, preprocessing, first arrival picking, static correction, amplitude compensation, deconvolution, wavefield separation, velocity analysis and imaging. Based on the characteristics of offshore 3D VSP downgoing multiples, the offshore 3D VSP downgoing multiple imaging technique has been innovatively developed. This greatly expands the 3D DAS-VSP imaging range and improves the overall 3D DAS-VSP imaging quality. The 3D DAS-VSP downgoing multiple reflection wave imaging shows significant imaging quality improvement in comparison with the vintage 3D OBC data imaging. The detailed structure, tracing of the reservoir and formation tops becomes easier and less ambiguous using both new OBN data imaging and 3D DAS-VSP data imaging when compared with the vintage 3D OBC data imaging.

3D seismic MegaMerges provide valuable information over large areas in mature petroleum provinces. While these surveys in combination with borehole data are useful for regional geological and geophysical analyses, they also offer an enormous opportunity for big-data analysis leading to new understandings. In this project, we combine seismic 3D-merges and well data in a geological context. The target is to predict petrophysical properties such as P-wave velocity (VP), Bulk Density (RHOB), Porosity (∅), and Clay Volume (Vclay) from well and seismic data using machine learning (ML) techniques.

Our workflow comprises five steps. In the first step, seismic 3D surveys are conditioned, reprocessed, and finally merged. In the second step, a low-frequency Acoustic Impedance (ZP) model is generated using the well-log data over the whole MegaMerge before carrying out a model-based inversion to obtain a ZP cube. In the third step, Multi-linear and different neural network algorithms are used for the ML training and validation analysis. In step 4, the well-log trained ML model is applied to predict VP and RHOB from seismic. And in the final step 5, petrophysical relations and well data are used to predict Vclay and ∅ from the ML-derived properties.

In the current example of the Northern North Sea (NNS) MegaMerge, this new approach shows excellent results, providing new insights into the basin geology, petroleum system, reservoir architecture and reservoir quality..

The Permian Basin contains significant oil and gas-bearing shale deposits. Extending over 55 counties in West Texas and southeast New Mexico, it covers a vast region. It is the largest contributor to the oil shale boom in the US and in February 2022 accounted for more than 40% of US oil, Figure 1 (Federal Reserve Bank of Dallas 2022). Figure 2 displays the increase in produced oil from this basin from 2018 to 2022, Chevron News Room 2022.

During the first years of production from these tight shales, seismic data weren’t considered to be essential for locating and drilling high-producing wells. However, the value of seismic data are now recognized and extensively used to design and execute lateral drilling programmes into the most prolific shale units. The shale deposits are widespread but highly heterogeneous in composition. Specifically notable is the lateral variability of Total Organic Content (TOC) and rock properties, such as porosity within the formations. In addition to varying shale properties, carbonate units were deposited simultaneously and are difficult to distinguish when only using reflectivity. Traditionally, the sonic and density values were inputted into supervised and non-supervised neural network applications to estimate rock properties. However, it is difficult to calculate density from conventional seismic data, and the results are typically questionable and have a high uncertainty element. To calculate density from seismic data, we need long offsets that are usable; most of the time, these data are not available (Aki, K., 1980). This study showcases a unique approach and its application using the LambdaRho-MuRho cross plot to estimate rock properties. The LambdaRho and MuRho values are derived from pre-stack inversions and are correlated to existing well data over the zone of interest. The results indicate that this is a more elegant and valid methodology, especially since it seems to produce a better, more accurate distinction between shale and carbonate domains.