Juan Perdomo

Combining rock physics analysis, full waveform prestack inversion and high-resolution seismic interpretation to map lithology units in deep water: A Gulf of Mexico case study

A successful seismic-based reservoir properties estimation effort has three steps: accurate seismic inversion in 3D to obtain relevant reservoir parameters, rock physics transformation to relate reservoir parameters to the seismic parameters, and mapping these parameters in 3D. This problem is nonunique and thus any available information—specifically geologic interpretation—should be used to improve our ability to infer the reservoir properties of interest with confidence. Moreover, uncertainty associated with the different predicted values (i.e., confidence interval and estimate of misclassification probability) must be provided as well, so that proper decisions can be made. Thus, it is evident that this involves interdisciplinary effort that includes rock physics, geologic interpretation, and seismic inversion technology. However, for quantitative description of reservoir properties, one must derive a way to quantify the errors and uncertainties associated with the process.

Angle gathers for RTM using extended imaging conditions

We can use extended imaging conditions in general, and the time lag imaging condition in particular, to compute angle gathers for reverse-time migration (RTM). The transformation of time lags to angles was proposed previously by other authors; however, the proposed workflow requires a dip field and is increasingly inaccurate in areas with conflicting dips and for dips greater than 60◦. In this paper, we propose methods to overcome the dip limitation and to do the angle transformation without explicitly using the dip field. We discuss the angular resolution present on these gathers and allude to the effects of acquisition geometry. We use a synthetic 2D model to compare results obtained using our proposed methods and the conventional workflow. Finally, we show RTM angle gathers for the SEAM model using our methods.

Imaging with complex decomposition: Numerical applications to seismic processing in difficult areas

Measured seismic wavefields have complexity that cannot be fully modeled by seismic processing, such as wave equation migration. Traditionally, a separation of data into signal and noise components is attempted, but it is difficult to isolate the noise component that cannot be modeled to be discarded. In this paper we use an alternative approach and assume that most energy in a measured wavefield is signal, and then we use variational methods to separate this signal into a simple part, which can be modeled by conventional seismic processing, and a complex part that is not amenable to investigation by these methods. In this approach, the complex part does have interpretation value and should not be discarded as noise. The simple part may be input to numerical processing such as depth migration, residual static after depth migration, and multiple attenuation, among others.

Resolution analysis for targeted illumination using two‐way wave‐equations

Recent advances in imaging prestack seismic data, in particular, more accurate migration algorithms such as reverse-time migration (RTM), have made higher-quality images possible. However, tools to understand the factors affecting the quality of the results such as illumination are not readily available. Typically, illumination and resolution studies involve ray-based methods, and although they capture the effects of lateral variations in velocities, intensity and directional information of wavefield propagation, and irregular acquisition geometries, they suffer from severe limitations in accuracy in complex regions where the high-frequency asymptotic approximation of rays might fail. Wave-equation methods generally do not generate the directional information of the wave propagation, which prevents us from using a waveequation method directly for illumination studies.

Propagating seismic data quality into rock physics analysis and reservoir property estimation: Case study of lithology prediction using full waveform inversion in clastic basins

Rock physics analysis can provide the relation between the parameters (or seismic attributes) that govern seismic wave propagation (e.g., Vp, Vs, and density in isotropic media) and the reservoir property of interest such as lithology, porosity, and saturation. In this process, we need to account for the quality of the seismic data and derive the appropriate uncertainties associated with the seismic data, such as noise, resolution, and inversion artifacts into the reservoir property estimation. In this case study, we show how to quantitatively propagate seismic data quality issues such as resolution, noise, and inversion accuracy into the lithology estimation in a clastic basin. The use of full waveform inversion and Bayesian classification techniques provides a mathematical framework that enable us to model and directly relate data quality input into the uncertainty associated with lithology prediction.

Imaging Beneath Complex Layering: Processing Seismic Data With Complex Decomposition And Renormalization Group

Seismic Imaging beneath complex layering is challenging and problematic. Examples of complex layering are complex shale, carbonate, basalt and salt. Both the quality of the image and wavelet fidelity of the reservoir beneath the complex layering are usually not satisfactory below complex layering, largely because of the associated multiples and elastic waves. These effects interfere with the primary reflection. The problem is more acute for seismic reservoir analysis where agreement between observed logs data and seismic is important. Apache Corporation and WesternGeco undertook a Strategic Technology Alliance Project of a modeling study to investigate old and new seismic processing methodology. The purpose is to do an in-depth analysis of assumptions and approximations in the conventional seismic processes. The study identifies which of these programs are ineffective in the presence of interference due to strong interaction in the overburden. Then we can replace these programs with the new processing programs like complex decomposition and renormalization group.