ALBERT

Notes: The increase in the number of geophone groups in production records during recent years and the requirement for accurate basic static corrections for high resolution records have made it necessary to develop sufficiently accurate automatic techniques for the determination of static corrections.A fully automatic method is presented which makes use of the delay-time method in order to compute static corrections at each shot position. Delay times, weathering and subweathering velocities are determined from automatic picks of the first arrivals on common-offset trace collections.It is assumed that the weathering is a single layer and that the dip of the subweathering layer under the geophone groups is small.The picking routine is fully automatic and successful in most cases, provided the signal-to-noise ratio is sufficiently high.The subsequent filtering of erroneous values for picked times is performed by means of statistical techniques, using curves of picked times on common-offset trace collections. If the distance between receivers and shot-points on the profile is sufficiently short, one can expect only little change in the picked times of two contiguous traces.The method is well adapted to end-on spreads with a great number of traces, where distances between geophone groups are short.Examples are presented showing the possibilities of the method for the determination of long wavelength as well as short wavelength components of static corrections.

Notes: A new deterministic technique for estimating a wavelet suggested by Loewenthal and Jakubowicz requires measurement of both pressure and vertical particle velocity. Through construction of the impedance function a deterministic estimation of the wavelet and the reflectivity can be obtained.This idea is tested for a one-dimensional model. The test is carried out by forming a synthetic seismogram of both pressure and particle velocity and checking the formulas for obtaining the estimated wavelet under noisy conditions.

Notes: Four methods for the determination of Q in marine sediments are compared: two traditional methods, i.e. the risetime and the spectral ratio method, and two newly established ones, the spectrum modeling and the wavelet modeling method. In the latter one Q and the reflection time T are determined simultaneously, which gives a much better accuracy for T than reading it from the seismogram. The risetime and the spectral ratio methods are used for obtaining Q directly from the data. The principle of the modeling methods is to calculate the effect of absorption and dispersion on a reference wavelet or its spectrum for various values of Q, and the best fit between the observed and the calculated data leads to the optimum result. Numerical tests on synthetic data show that a precision of more than 25% for data containing noise or superposed arrivals can hardly be achieved; in any case, wavelet modeling is the superior method. Application to data from a vertical reflection profile in the Baltic Sea yields Q in the range of 15–100 for different layers, which is to be expected in the sedimentary environment of this area.The computations were performed in the Computer Center of Kiel University. The authors thank R. Meissner for his comments on the manuscript.

Notes: Small offsets in hard coal seams can be detected with the aid of seam (channel) waves. Transmission and reflection of seam waves depend, among other parameters, upon the symmetry properties of the sequence rock/coal/rock. Two typical unsymmetrical sequences are found in European coal deposits: (i) coal seams with roof and floor of differing acoustic impedance and (ii) coal seams interlayered with rock and soil.Two-dimensional analog models with appropriate impedance contrasts are used to study the effect of the unsymmetrical layers upon the propagation of Rayleigh seam waves. Data analysis is based upon amplitude measurements both parallel and perpendicular to the layers and dispersion curves.The effect of unsymmetrical roof (rock 1) and floor (rock 2) was studied with models containing homogeneous coal seams. Leaky mode wave groups with phase velocities (cR) in the range between the SV-wave velocities (βr1, 2) of the two rock materials, i.e. βr1≥cR 〉 βr2, form a characteristic part of the Rayleigh seam wave signal. Using Knott's energy coefficient calculations it is shown that in that range energy leakage into the surrounding rock by refracted SV-waves is restricted to only one of the two interfaces, namely coal/rock 2. At the other interface, coal/rock 1, all waves are totally reflected. Thus, the high amplitudes of these leaky mode wave groups are explained by “quasi-normal mode” features.The influence of a dirt bed on wave propagation was studied in models where the roof and the floor have the same elastic properties. The maximum thickness of the dirt bed did not exceed 20% of the total seam thickness. The effect of the bed's location within the seam was also investigated.For all recorded normal-mode wave groups either the total seam or the coal layers could be regarded as wave guides. This was shown by the fact that the phases could be associated with the phase velocity dispersion curves calculated for the symmetrical sequence rock/coal/rock. These curves are relevant under the condition that the thickness of the coal layer assumed under the calculation coincides with the thickness of the effective wave guide of the respective wave groups.Wave groups guided in the total seam are not influenced by either the thickness or the position of the dirt bed. On the other hand, for wave groups guided in the coal layers, the quotient of signal amplitudes in the coal layers is influenced by the position of the dirt bed.

Notes: Shotpoint gathers from conventional reflection seismic surveys contain both reflected and refracted waves. In this study shot records were processed and analyzed, and the data were modeled with reflected, refracted, and reflected-refracted waves to fit the recorded data. The result is a detailed velocity model. The inverse problem for refracted waves was solved by using the Wiechert-Herglotz inversion.A 500-km-long 26-fold reflection seismic line from the Barents Sea, north of Norway, has been investigated. The data show high velocities, multiple reflections, and various types of noise. To test the method a total of 34 shot gathers were analyzed along this line. The aim of the interpretation was to determine the velocity in the seafloor and the near-surface sediments. It is possible to map the vertical as well as the lateral velocity distribution in detail. Depending on the length of the streamer and the velocity gradient in the sediments, the calculated depth varies between 300 and 500 m below the seafloor. These velocities were also compared to the stacking velocities obtained from the reflection seismic data to see how the velocities determined by different methods were related.The velocity distribution in the sediments is one of the key factors in seismic interpretation. The technique discussed in this paper can contribute to velocity information both in the processing and interpretation of seismic data.

Notes: In the course of investigations concerning disposal of radioactive waste, DC-geoelectric soundings were performed in the salt mine Asse in 1982. The survey resulted in the determination of resistivities of various salt formations. A comparison of various resistivities obtained in salt formations in the mine shows that it is possible to estimate the content of free water in salt using Archie's equation.The significant result of this survey is that salt formations in the mine can be divided into two categories according to their resistivities:1. Salt with resistivity between 107 and 108Ωm (particularly Na2Sp and Na3β). The particularly pure halitic formations are to be found within this resistivity range.2. Salt with resistivity between 103 and 105Ωm (Na2Tl, K2, transition salt). The salt formations with thin layers of anhydrite and a high content of hygroscopic salts are classified in the paper.

Notes: Various exact methods of inverting the complete waveform of vertical seismic reflection data to produce acoustic impedance profiles have been suggested. These inverse methods generally remain valid for nonvertical, plane-wave data, provided total reflection does not occur. Thus, in principle, the “seismogram” at each ray parameter in a slant stack can be interpreted separately.Rather than invert each plane-wave seismogram separately, they can all be interpreted simultaneously and an “average” model thus obtained. Inversion for both the velocity and the density also becomes possible when two or more plane-wave seismograms are simultaneously inverted. The theory for a noniterative inversion method, based on the time-domain Riccati equation, is discussed. Numerical examples of inversions using this technique on synthetic data demonstrate its numerical stability and the advantage of simultaneous inversion of several seismograms to reduce the effect of noise in the data and increase the stability of the inversion process.