Wang, Jingjing

An integrated project can take many forms depending on available data. As simple as a horizontally isotropic model with estimated hydraulic fracture geometries used for simple approximations, to a large scale seismic to simulation workflow. Presented is a large-scale workflow designed to take into consideration a vast source of data.

In this study, the team investigates a development area in the Eagle Ford rich in data acquisition. We develop a robust workflow, taking into account field data acquisition (seismic, 4D seismic and chemical tracers), laboratory (geomechanical, geochemistry and PVT) measurements and correlations, petrophysical measurements (characterization, facies, electrical borehole image), real time field surveillance (microseismic, MTI, fracture hit prevention and mitigation program through pressure monitoring) and finally integrating all the components of a complex large scale project into a common simulation platform (seismic, geomodelling, hydraulic fracturing and reservoir simulation) which is used to run sensitivities.

The workflow developed and applied for this project can be scaled for projects of any size depending on the data available. After integrating data from various disciplines, the following primary drivers and reservoir understanding can be concluded. At a given oil price, optimum well spacing for a given completion strategy can be developed to maximize rate of return of the project. Many operators function in isolated teams with a genuine effort for collaboration, however genuine effort is not enough for a successful integrated modelling project, a dedicated multidisciplinary team is required.

We present what is to our knowledge, one of the most complete data sets used for an integrated modelling project to be presented to the public. The specific lessons from the project are applied to future Eagle Ford projects, while the overall workflow developed can be tailored and applied to any future field developments.

Summary

We make comparison analysis of skewness and kurtosis criteria for constant-phase wavelet estimation, which plays a significant part in high-resolution seismic data processing and interpretation. Synthetic examples illustrate that skewness criterion has higher dynamic range and better stability than kurtosis criterion. As a high-order statistics, skewness criterion breaks the limitation that the bandwidth of input signal must be greater than center frequency and reduces the influence of noise on wavelet phase estimation. Like kurtosis criterion, the estimated phase corresponding to maximum skewness is also affected by the distribution of reflectivity.

Introduction

Seismic wavelet estimation plays an important role in data processing and interpretation. In particular, the phase of wavelet is paramount because it directly has an impact on the result of forward modeling, deconvolution and inversion (e.g. Lu, 2005; Yuan et al, 2011; Yu et al, 2011; Yuan and Wang, 2013). Deterministic approach for dealing with the wavelet estimation utilizes the well logs. However, well logs are not always accessible. Different logging information sometimes predicts various results of wavelet estimation. Therefore, it is demanded for statistical method to estimate the wavelet phase from the seismic record itself.

Under the assumption that wavelet phase does not depend on the changing frequency, Levy and Oldenburg (1987), Longbottom et al. (1988), and Van der Baan (2008) proposed the method of wavelet phase estimation based on kurtosis, whose maximum indicates the best phase of wavelet. As the fourth-order statistics of data, kurtosis is applied to measuring the deviation of a signal from Gaussianity. Many researchers improved it and made great achievements. White (1988) used this criterion for wavelet phase estimation and proposed the maximum kurtosis phase correction. Considering that kurtosis is sensitive to large reflection coefficient, Lu and Liu (2007) introduced an improved kurtosis criterion. Also other criteria were studied to indicate the phase of wavelet. Fomel and Baan (2014) estimated the local phase of seismic data by the skewness and defined a new seismic attribute through expanding this criterion to the space domain. As a new criterion function, skewness is used to process nonstationary data and achieves a good result. Compared with kurtosis, skewness has a higher dynamical range. But the deep distinction of kurtosis and skewness is not explicitly elaborated in previous study.