What is and is not retained of the rock record in seismic reflection data? Experimental strata formed in a new laboratory basin offer a novel opportunity for exploring this question to a level of detail not possible in the field. The basin, developed at St. Anthony Falls Laboratory of the Unviersity of Minnesota, is used to simulate strata formation under controlled rates of base level change, sediment supply, and subsidence, which can be varied spatially and temporally across the basin floor. The resulting deposits approach the appearance and complexity of natural strata found in sedimentary bodies ranging in scale from bedforms to continental margins. Furthermore, the experimental strata are completely dissected and digitally imaged in three dimensions.
We are using the digital images to generate synthetic seismic data via a number of techniques. These range from a simple convolutional model, which produces the equivalent of ideal post-stack, time-migrated seismic data, to numerically solving the two-way wave equation, which we use to produce a synthetic version of pre-stack, unmigrated seismic data. These synthetic data are being compared to the digital images of the experimental strata to assess the amount of stratigraphic information preserved in the modeled seismic data.
Pratson, L.F. and Wences G., in press, Seismic simulations of experimental strata, submitted to the American Association of Petroleum Geologists Bulletin for expanded publication Feb 2002.
The XES Basin prototype is 1.6 m long by 1 m wide, with up to 0.8 m of accommodation space for deposition (A). Spatial and temporal variations in accommodation space are created by the basin's subsiding floor. Underlying the floor are 10 hexagonal-shaped subsidence cells arranged in a honeycomb pattern (B). At the beginning of an experiment, these cells are buried beneath a layer of dry, well-sorted commercial gravel (A). The top of the gravel is then covered with a thin rubber membrane, which forms the basin floor. The membrane subsides by withdrawing a small volume of gravel from the bottoms of the underlying subsidence cells. The subsidence is smooth and continuous in time and space, and can be varied between adjacent cells to produce slopes in the membrane of up to 60 degrees.
In the experiment, water mixed with a 50%:50% blend (by volume) of fine (120 mm) quartz and coal sand was fed from a single source point into one side of the basin while base level was adjusted from the other side. The more buoyant coal sands (specific gravity of 1.3 kg*m^-2*s^-2) were a visibly distinct proxy for sediments that were finer-grained and slower-settling than the quartz sands (specific gravity 2.65 kg*m^-2*s^-2).
Subsidence rates were varied across the basin so as to produce a simple bowl-shaped geometry centered over the middle of the basin. The rates of subsidence and of water and sediment discharge were held constant throughout the experiment. The only factor that was varied was base level. This history included a slow fall and rise of base level followed hours later by a rapid fall and rise.
After the experiment was completed, the basin strata were sliced lengthwise at intervals of 2.5 cm. Each cross-sectional slice was digitally photographed, allowing the strata to be digitally reassembled and viewed in three dimensions (e.g. below left). The photographic panel shown below to the right is an example of such a cross section. This particular panel is located just off the center dipline through the basin. Labels on the photograph identify a range of deposits and structures in the strata and indicate the relative timing of their formation with respect to the changes in base level that occurred during the experiment (graph below panel).
The experimental strata clearly capture in miniature the sequence stratagraphic architecture of basin strata, i.e., the structures, stratal patterns and facies distributions. Therefore, because of the natural, scale-independent appearance of the experimental strata, we simply assume that its architecture can be resized to basin proportions and used to generate synthetic seismic data at exploration frequencies.
Click on the images below to see enlargements.
In the digital photos of the experimental strata, the lightest areas in the deposit are quartz sand, and the darkest areas are coal, which, as mentioned above, is used as a proxy for mud. The gray scales in the photos are converted into acoustic models, which in turn are then used to simulate seismic data. One of the simplest procedures for doing this is as follows:
--Assign
dimensions to each pixel in the digital photo (in the case shown, dx = 1m,
dy = 5m).
--Convert the gray scale value of each pixel (0 - 225) to a value of bulk
density (A)
--Compute porosities from the bulk densities.
--Compute P-wave velocities (B) from the porosities using an empirical relation
(e.g. Hamilton, 1980).
--Compute models of acoustic impedance and reflectivity (C) from the densities
and velocities.
--Convolve the reflectivity model with a seismic source wavelet (D) to produce
normal-incidence seismic data (F).
Convolution
Model
Simple convolutional modeling of seismic data simulates
ideally processed seismic data. It does not simulate seismic wave propagation,
nor its effects on the recorded seismograms. Therefore, the convolutional
simulation is theoretically the "best" data that can be obtained
with a given seismic source.
One-Way
Wave Equation
More realistic simulations of seismic data are obtained
by numerically solving the wave equation. The so-called "exploding reflector"
model, which solves the wave equation for upward propagating waves, accounts
for diffractions and lateral wave propagation.
Two-Way
Wave Equation
Solution of the two-way wave equation simulates the
full propagation of seismic waves, from their source, through multi-layered
strata, to their arrival at surface receivers. This more realistic model of
wave propagation in two dimensions accounts for such additional effects as
geometrical spreading, multiples, refractions, and reflections at interfaces.
The model can be used to simulate multichannel seismic data. An example of
multichannel shot gathers is shown to the right.
The following simulations illustrate the relationship between the experimental strata and the synthetic seismic data produced by the convolution model, the one-way wave equation model, and the two-way wave equation model. For these comparisons, the one-way and two-way model data were migrated. But before the two-way data were migrated, they underwent additional processing including:
--Deconvolution.
--Muting of first arrivals.
--Velocity analysis.
--Normal moveout correction and stacking.
--Dip moveout correction.
--Inverse normal moveout.
--Velocity analysis.
--Final stacking.
Click on the images or names below to see the simulations.
St. Anthony Falls Laboratory of the Unviersity of Minnesota
Paul Heller, University of Wyoming (click on "Paul's research and teaching Web page")
Heller, P.L., C. Paola, I-G Hwang, B. John, and R. Steel, 2001, Geomorphology and sequence stratigraphy due to slow and rapid base-level changes in an experimental subsiding basin: American Association of Petroleum Geologists Bulletin, v.
Paola, C., 2000, Quantitative models of sedimentary basin filling: Sedimentology, v. 47, p. 121-178.
Back to Lincoln
Pratson's home page
Back to the Duke University Solid
Earth Home Page
Back to the Duke University Surface
Processes Home Page
Return
to the EOS home page
All images and movies are copyrighted.
Last updated November 9, 2001
