Acoustic wave equation based seismic modeling using numerical methods requires effective absorbing boundary conditions to limit the domain size, maintain stability, and contain the computational cost. Numerically the boundary sides and bottom of model regions must be carefully managed to reduce the unwanted internally reflected energy. both analytic and numerical methods have been developed for use in computer codes. Using a measure of reflected wave energy it is possible to find good sponge zone weights for various numbers of grid points. Further, using this metric of reflected wave energy it is possible to construct a contour map for a range of sponge weights and numbers of grid points. This contour map has a global minimum and for the best (nearly) optimal weight and length a remarkably good seismic modeling result is obtained.
Some of the principal targets for newly available AGG surveys are kimberlite pipes. These bodies may be small in extent and have anomaly amplitudes close to current detection limits. In this paper estimates of source depth and horizontal location of vertical pipes have been recovered from synthetic AGG data with errors mostly less than 80 metres and 40 metres respectively (+/−25% and +/− 13% of source depth), at line spacings of up to 200 metres. With the same quality data at line spacings of 400 metres recovery of source parameters is much less reliable.
For crosswell seismic waveform inverse problems, anelastic parameters for heterogeneous media need to be taken into account. The Fre´chet derivatives allow solving inverse problem with a high number of parameters; however, for the anelastic case, the Fre´chet derivatives do not have simple expressions. This is due to the time convolution in the general linear constitutive relation. The viscoelastic model, based on the superposition of relaxation Zener mechanisms, allows turning the time convolution relation into a set of first order partial differential equations which can be numerically handled. Equations are rewritten to model an approximately constant quality factor for frequency bandwidth. Thus, by using this simplified constitutive relation and the associated Fre´chet derivatives, a new formulation for the viscoelastic anisotropic seismic inverse problem is derived. This approach reduces the computer time and memory requirements to the point where computations are practical. We illustrate the feasibility of the method by a full-wave inversion of a synthetic crosswell synthetic data set for a 2D viscoelastic isotropic medium.
A seismic field test is described where comparison shot records were recorded using both a conventional seismic explosive and dBX*. Comparison of test records on an individual shot basis finds the dBX product more energetic. Though desirable, this is only one aspect to be considered when selecting seismic explosive source parameters. A seismic survey is necessarily made up of a multiplicity of source points. Analysis of the consistency of the seismic data, in both signal and noise content, generated across multiple shot points can provide insight as to how the field data will interact with later data processing work.
All Nuclear Magnetic Resonance (NMR) applications require a static magnetic field to polarise the nuclei under investigation. For Surface NMR (SNMR) this field is the Earth's geomagnetic field and the nuclei of interest are the hydrogen protons associated with water (and to a much lesser extent, hydrocarbons). Using synthetic data, the effects of moderate horizontal gradients in the Earth's geomagnetic field on groundwater parameters derived from SNMR measurements are investigated. It is shown that although gradients have a minimal effect on the theoretical amplitude and phase of the initial SNMR response, the subsequent time domain signal is distorted relative to the non-gradient case. This distortion leads to significant errors in the estimation of observed groundwater parameters when processing SNMR time series data.
In this paper we present a new practical and efficient method to calculate approximate seismic images without a priory knowledge of the subsurface model or velocity analysis. By analogy of the use of Feynman path integrals in waveform modeling, we develop an algorithm for model-independent imaging by summation of elementary signals over a sample of all possible paths between source and receiver. The constructive and destructive interference of these signals produces and approximate image of the subsurface.
The effects of mis-positioning in time-lapse seismics are studied, focusing on irregular sampling due to receiver mis-positioning and absence of data due to surface obstacles. Wavefield reconstruction is used to regularize data and to reconstruct data where data-gaps are present. The method is tested on time-lapse 2D synthetic and Troll streamer data. Quality control is performed using conventional di_erence plots, NRMS percentage and the 4D attributes re ection coefficient ratio and two-way travel-time shift.
Chalks and marl reservoirs may maintain high porosity at more than two kilometers of burial depths due to overpressure and the presence of hydrocarbons. As a result, formation stability is compromised considerably and wells are is danger of being lost. Analyses of the variations in velocity, attenuation, and static compression modulus in carbonates allow us to better interpret seismic measurements in terms of subsurface petrophysical parameters and to understand the failure and damage potential. We present a study of variations in dynamic moduli in various carbonate reservoirs. Our study includes log and laboratory data from velocity, porosity, permeability, and attenuation measurements.
Coupling of detectors on the seafloor is well understood and has been characterized by damped oscillatory systems that vary in amplitude and phase as a function of frequency. Furthermore, this response is very different than the response on land due to inertial effects of water. When seafloor motion is out of phase with water motion, larger resonances and phase distortions can result. However, this phenomenon should be alleviated for permanently installed detectors that are buried below the seafloor and provide near-perfect coupling for time-lapse observations. At Teal South in the Gulf of Mexico, buried and non-buried cables were deployed to evaluate differences in coupling. In this study an orientation and scale factor analysis using the direct arrival suggests that there is no benefit to burying detectors for the part of the bandwidth associated with these high frequency P-waves. A full-bandwidth cross-equalization analysis indicates around a 6 dB smaller amount of resonance at lower frequencies for S-waves on the crossline component, and reduced torsional modes on the vertical component. However, these benefits are small for the buried detectors and could be compromised by poor response of gimbaled detectors or uncompacted sediments in the trench where detectors are buried.
Rodríguez, Omar Delgado (Instituto Mexicano del Petréleo) | Shevnin, Vladimir (Instituto Mexicano del Petréleo) | Valdés, Jesús Ochoa (Instituto Mexicano del Petréleo) | Ryjov, Albert (Moscow State Geological Prospecting Academy)
Resistivity method is used extensively in environmental impact studies. In this work, the results of the geoelectrical characterization of a hydrocarbons contaminated site are presented. Although the contamination grade of the study area is low, were mapped two contaminated zones into sandy aquifer. In addition, petrophysical parameters were estimated by recalculate of ground and water resistivity values in clay content, porosity and CEC values. Anomalous values of clay content, porosity and CEC indicate the presence of hydrocarbon contaminants. The correlation between geoelectrical results, petrophysical parameters and hydrocarbons contamination was verified in laboratory by electrical measurements made in pure and contaminated sand samples.