Brian J. Zook

Distributed computation of electromagnetic scattering problems using finite-difference time-domain decompositions

Finite-Difference Time-Domain (FDTD) is an important numerical method for solving electromagnetic scattering problems. Solutions for these problems are computationally intensive, In addition, large problems require large amounts of memory, making distributed memory desirable. This paper describes distributed FDTD methods and their implementations on a distributed workstation network using PVM. We examine the parallelism of the methods and evaluate computing performance by studying load balancing, communication pattern variations, and scalability of the programs on the distributed network systems. This work should be relevant to anyone solving partial differential equations using finite-difference techniques.<<ETX>>

Interwell seismic logging for formation continuity at the Gypsy test site, Oklahoma

The continuity logging (or in-seam) method has been used extensively in the coal industry to determine continuity. Continuity logging is an alternative approach to multiple-offset crosswell seismic measurements using guided waves propagating parallel to the layering; in particular, to determine the continuity of sand and shale stratigraphy in oil and gas fields. The present work is a study of rock physical properties and seismic data for reservoir continuity using interwell logging techniques. Processing methods are used to delineate the reservoir boundaries and to relate seismic waves with rock physical parameters. The specific application has been to identify and analyze the propagation characteristics of guided waves in the Gypsy-fluvial sandstone formation at the Gypsy test site in Oklahoma. The integration of well logs and interwell seismic data, including trapped seismic events, suggests the presence of a continuous low-velocity clean sand in the Gypsy sandstone interval. The correlation of sand and shale stratigraphy with well logs and guided waves is consistent with available pressure pulse test data, indicating that the sands are in pressure communication between the wells and that an impermeable barrier exists between the top and bottom of the sand units in the Gypsy sandstone reservoir.

A field test of an electrodeless arc discharge, borehole seismic source

Waveforms generated by an impulsive, 1.2 kJ, seven‐conductor wireline electrodeless arc discharge borehole seismic source or sparker at Texaco’s Humble, TX field test site were recorded by three borehole sensor arrays: two free‐hanging hydrophone streamers in in‐line boreholes at 82 m and 170 m from the source well and a grouted, three‐component geophone string in a borehole 110 m from the source well. A repeatability test of the source, consisting of single firings of the source at a rate of 1 firing per 5 s, showed very clean, very strong, Ricker‐like wavelets. Despite a high‐degree of attenuation (exact value of Q is not known), the useful frequency passband of the wavelets was from 200 Hz to 1200 Hz for the data recorded by the 82-m offset hydrophones and 200 Hz to 500 Hz for the 170-m hydrophones. Using 62 single‐firing wavelets recorded in the 82-m offset well gave mean and median crosscorrelations greater than 0.96 with standard deviations less than 0.02. A stack test, consisting of 1, 2, 4, 8, 16,...

Transmission and detection of guided seismic waves in attenuating media

We can use guided seismic waves to map properties of reservoirs between wells, with the low-velocity layers acting as waveguides. When guided waves are detected, they are an indication of the continuity of the bed examined. Guided waveforms are characterized by time-frequency representations to study important physical properties of the beds acting as waveguides. We used full waveform seismic modeling in viscoelastic media to examine the required velocity contrasts and distances over which guided-wave signals can be used. In one set of models, sandstones are the central waveguide lithology; in another set, shales. We applied these models, referred to here collectively as shaly sandstone waveguides, to a range of geological circumstances where either the sands or the shales represent the low-velocity layers within a reservoir. To study the distances over which guided waves can be detected, we compared the amplitudes of the signals computed for the models, using a realistic source strength, to the signal levels determined from published borehole noise studies. In shaly sandstone waveguides, we find it is feasible to use particle velocity measurements to record guided waves above seismic noise levels in the frequency range of 60 to 800 Hz at well separations exceeding a distance of 800 m. However, pressure detectors such as hydrophones may only be useful up to distances of 400 m between wells. In addition to the issues of shaly sandstone waveguides and practical distances between wells, we present an application of guided waves using crosswell seismic data from the Gypsy test site in Oklahoma (a site originally established by British Petroleum). In this field example within a sandstone reservoir, we demonstrate the sensitivity of leaky mode amplitudes to source-receiver location. Another telltale characteristic of continuity in the type of reservoir studied at the Gypsy test site, where there is a low shear velocity contrast between the host medium and the waveguide, is the head wave followed by the leaky mode.

Interwell seismic imaging at the Savannah River Site, South Carolina

Crosswell and continuity logging seismic measurements were made beneath a large tank (27 m diameter) used for processing radioactive waste at the Department of Energy (DOE) Savannah River Site in the Atlantic Coastal Plain of South Carolina. We used the data to delineate a low‐velocity zone (soft materials) and image the connectivity of a clay unit between wells. The low‐velocity zone depicted on the crosswell seismic tomogram integrated with data from cores and well logs revealed soft materials in the region between 150 and 180 ft (46–55 m). The bottom boundary of this low‐velocity zone correlates with a reflection observed in the crosswell seismic image at a depth of 180 ft (55 m). This reflection corresponds to the impedance contrast between the soft materials and the more rigid Tinker Formation. The low‐velocity zone of soft materials indicates a dissolution margin of a carbonate unit (which is part of the Utley limestone) and the presence of loose sands of the Griffins Landing Member. Ray tracing and...

Reverse VSP and crosswell seismic imaging at the Savannah River site

Analysis of crosswell and three-component seismic data integrated with well logs have produced information on the distribution of subsurface heterogeneities below the In-Tank Precipitation facility at the Savannah River Site (SRS). The travel time P-wave tomogram and reflection imaging delineate lateral and vertical structural details of the formations. In particular, the high-resolution P-wave tomogram captures a low-velocity zone within the carbonates. This zone is surrounded by reflection events between depths of 150 and 200 ft. in the reflection imaging. The reflections are caused by the acoustic impedance contrast between the low velocity zone of `soupy` sand mixtures of unconsolidated materials and the more rigid and dense competent surrounded medium. The time-frequency analysis of full waveforms particle velocity identifies guided waves in form of leaky and normal modes at the depths of about 138 to 150 ft. This resulting change in lithology associated with the presence of guided waves is consistent with a velocity low observed in the vertical velocity profile determined from the inversion of three-component seismic data. This low-velocity zone intercepted by the wells H-BOR-34 and H-BOR-50 correlates with the conductive Griffins Landing Member, which is located above the carbonates. The result of the experiments demonstrate that the present high-resolution crosswell seismic measurement technique (using frequencies up to 1500 Hz) meets the resolution requirements to map geological and geotechnical targets in the vicinity of the In-Tank Precipitation facility at the Savannah River Site

Detectability of guided seismic waves in the presence of attenuation: Model study

Continuity logging measurements conducted between wells at large separations require understanding the propagation characteristics of normal and leaky modes in lossy waveguides. In this paper we determine the feasibility of detecting guided seismic waves between wells by taking into account the seismic source signal strength and receiver noise levels for a buried point source mechanism. Several computer models were constructed by varying the impedance contrast, bed thickness, Q{sub p}, Q{sub s}, and the source peak frequency. These models were used to produce waveguide responses at various well separations using a multimode analytical solution to evaluate if the trapped energy in the waveguide will reach the desired distance when the layers are continuous. The preliminary results suggested that it is feasible to detect guided waves above electronic and seismic noise levels in shaly sandstone formations at a well separation of 800 m.