WO2001001350A1

WO2001001350A1 - Multimedia techniques for multidimensional data interpretation

Info

Publication number: WO2001001350A1
Application number: PCT/US2000/017382
Authority: WO
Inventors: David J. Monk; David Ridyard; H. Roice Nelson
Original assignee: Continuum Resources, Corp.
Priority date: 1999-06-25
Filing date: 2000-06-23
Publication date: 2001-01-04
Also published as: AU5887300A

Abstract

A method and system for investigating seismic data sets using the attributes presented in combinations of visual, audible, and haptic forms, using a 3D visual display (604) and the data explorer (602). The explorer processes the seismic data to generate output values, which are conveyed to the user as corresponding values of an audible, visible, or haptic signal attribute (604). A sound generated having a frequency, amplitude, beat frequency, and sounding pattern (606), is adjusted to correspond to the explorer output. The user control is altered in proportion to the explorer output. The color, transparency and texture of the explorer screen (604) similary can be modified to reflect the seismic output values of the explorer. The explorer output parameters are mapped to provide simultaneous access to a large number of parameters.

Description

Multimedia Techniques For Multidimensional Data Interpretation RELATED APPLICATIONS

The present application claims priority to U.S. Provisional Patent Application 60/141,177 filed June 25, 1999. This application is hereby incorporated herein by reference. BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

The present invention generally relates to the field of multidimensional data analysis and interpretation, and more specifically relates to the use of multimedia techniques for presenting data. Description of the Related Art

Science has progressed to a point which generally allows investigators to gather an enormous amount of data relating to any given problem. Often, however, so much data is gathered that it becomes difficult to manage. Scientists have even coined a phrase for the problem, saying that "the solution is hidden in the data". Seismology is only one example of one area where this problem is acute. While the ensuing discussion centers around seismology, the disclosed techniques are expected to be applicable to the fields of medicine, financial modeling, and scientific modeling, among others. In other words, although this document describes the invention in geoscience terms, the invention has broad application across all sciences.

The field of seismology focuses on the use of seismic waves to locate mineral deposits such as hydrocarbons, ores, water, and geothermal reservoirs. Seismology is also used for archaeological purposes and to obtain geological information for engineering. Exploration seismology provides data that, when used in conjunction with other available geophysical, borehole, and geological data, can provide information about the structure and distribution of rock types and their contents. Oil companies use seismology to select sites for drilling wells. While it is difficult to directly identify petroleum reservoirs using surface methods, seismic data is used to map geological structures which may provide some indirect indication of the production potential. Experience has shown that the use of seismic data greatly improves the likelihood of a successful venture. Seismic data (land, marine or transition zone) reveals information about the geology of layers below the surface. These data are acquired using a seismic source (generally an array of high pressure air guns for marine data, and dynamite or vibratory seismic sources for land data), which generates a source signature near the surface. The sound travels downwards, and is reflected from interfaces between different geologic layers below the surface. The reflected energy is recorded using pressure or velocity sensitive detectors, at or below the surface. Subsequent processing of the recorded data can yield an image of the geologic subsurface.

A variety of methods for acquiring and processing seismic survey data are well known. The final result of such methods is typically either a set of two-dimensional (2D) "slices" through the earth, or a three-dimensional (3D), seismic volume representation of an earth volume. Ideally, the seismic data is a presented in dimensions "x", "y" (surface position) and "t" time or "z" (depth) below the surface, with the amplitude at each defined point being representative of the acoustic impedance at that point in the earth. Potentially, multiple 3D seismic surveys conducted over a period of time (4D seismic) can reveal changes in the subsurface which may be the result of producing hydrocarbons from a reservoir. Figure 1 shows a typical seismic representation of a 2D slice of the earth. The dark areas have high acoustic reflectivity values indicating large changes in the rock properties at those points in the earth. After acquisition and processing of seismic data, the information is "interpreted". As a rule, interpretation has historically been performed visually.

Interpretation of seismic data involves identification of the major boundaries between the rock types, and using well log or other information to gain an understanding of the rock properties which may lead to the acoustic changes that are seen in the seismic data. Typically the inteφreter needs to examine and integrate many different sources of information to perform an inteφretation.

The seismic sections show "reflectivity", "acoustic impedance" or implied velocity, none of which is of particular interest to the geologist and petroleum engineer. Similarly, they are not interested in the gamma ray absoφtion, or other well log information. Rather, they are interested in the rock properties which lead to the results seen in a seismic section or well log. The inteφreter uses these data to infer information about structure, stratigraphy, and rock properties such as "permeability", "porosity", fluid pressure, and the presence or absence of commercially viable hydrocarbon reserves.

Inteφretation of seismic data, at its most basic level, involves establishing and identifying a series of "surfaces" which are the interfaces between rock layers. For example, Figure 2 shows some typical 2D seismic data, and Figure 3 shows how that data may be inteφreted with some initial surface information. Figure 4 shows how the use of color may assist in the inteφretation of the data. To inteφret seismic data, the inteφreter is often required to examine many "attributes" of the data. A list of exemplary seismic attributes is provided in Tables 1-3. Note that there are over

Table 1 Summary of Instantaneous attributes

Name Description Applicability

Instantaneous real The time-domam vibration amplitude of traces Traditionally widely used in structural and stratigraphic interpretations amplitude f(t) at the selected sample It is the default Often used as the base with other amplitude attributes to isolate high expression of most seismic trace data or low amplitude areas such as bright spots and dim spots

Instantaneous quadrature Stemmed from the complex seismic trace The phase-delay feature is useful in quality control of the vertical amplitude q(t) analysis, being the time-domain vibration variation of instantaneous phase, also useful in identifying some AVO amplitude whose phase delay is 90^' from the anomalies from thin-layer reservoirs since they may only be instantaneous real amplitude observable at specific phases

Instantaneous phase γ(i) Defined as the angle whose tangent is Tends to enhance weak mtra-reservoir events but also enhance noise (q(t)/f(t)) the phase (modulation) of traces at Color scale on final map should take into account circular nature of results the selected sample, in degrees or radians (ι e γ_leo =v,_κ) Since hydrocarbons often cause local phasing this attribute is often used with others as one of the hydrocarbon indicators

Cosine of instantaneous An attribute derived from instantaneous Since its fixed bounds (-1 to +1 ) are easier to understand, it is often phase cos (γ(t» phase used with instantaneous phase for better displaying its variation

Product of mst real This composite attribute is designed to This attribute enhances peak and trough amplitudes and turns all amplitude and cosine of mst enhance peak trough amplitudes, especially trough amplitudes as pseudo peak amplitudes for better structural phase f(t)cos (γ(t)) for zero-phase seismic data interpretation

Instantaneous frequency Defined as the derivative of instantaneous Often used to estimate seismic attenuation Oil and gas reservoirs usually ψ(α) phase with time, dγ (t)/dt, in degrees/ms or cause drop-off of high frequency components It helps to measure cyclic- radians/ms ity of geological intervals Tends to be unstable in the presence of noise

Amplitude weighted Instantaneous frequency weighted by Provides a more robust/smoothed estimate of instantaneous frequency instantaneous frequency instantaneous amplitude formulated as and is less prone to noise ∑(f(t)²dγ(t)/dt) ∑(f(t)²)

Energy weighted Instantaneous frequency weighted by Provides the most robust estimate of instantaneous frequency But instantaneous frequency theinslantaneous energy Aft) as such a smoothing can also suppress anomaly information in the trace Σ(A(t)=dγ(t)/dt)/Σ(A( ³)

Slope of instantaneous Defined as the rate of mst frequency change Often used to indicate rate of attenuation and absorption Since gas, frequency d(d-y(t)/dt)/dt oil, and waterbπne saturation cause different attenuations, this attribute for high-resolution data can indicate those fluid boundaries, useful in time lapse 3-D

Reflection strength A(t) Defined as (f(t)^! + q(t)²)'² Alternative names Useful in identifying bπght/dim flat spots Often used to determine lateral are "Instantaneous amplitude, "Amplitude fluid, lithologic and stratigraphic variations in reservoirs As the absolute envelope" value of the complex trace magnitude, it loses some vertical resolution dB-based reflection strength Decibel scale of reflection strength, or 20 The decibel scale is often used in frequency domain to display the power times the common logarithm of reflection spectrum Here this attπbute is used to examine the variation or anomaly strength of reflection strength in decibel scale Fractal analysis of this attπbute that is a model-based descnption of seismic waveforms can yield some fractal mdex to stratigraphic sequence and possible hydrocarbon anomaly ed-liltered energy of Defined as the time-domain median-filtered This median-filtered energy attπbute enhances the peak anomalies of reflection strength energy of reflection strength reflection strength, similar to the role played by another attπbute peπgram, a filtered reflection strength dB-based energy of Decibel scale of time-domain energy of Often used with dB-based reflection strength to examine the time reflection strength reflection strength domain energy anomaly in decibel scale

Slope of reflection strength The rate of reflection strength change over Very useful in characterizing vertical stratigraphic sequences and time vertical variation in reservoir fluid content such as in time lapse 3-D

Product of filtered reflection Filtered reflection strength or peπgram is Useful in analysis of amplitude anomalies by mapping enhanced high- strength & cosine of mst obtained by subtracting the dc/average amplitude and continuous events phase component from the reflection strength

Apparent polarity Defined as the polarity of reflection strength Useful for checking the lateral variation of polarity along a reflection layer Often used in conjunction with reflection strength

Response phase Derived from the instantaneous phase around An alternative way to track the time and spatial-vaπant phase change the lobes of reflection strength of the seismic wavelet

Response frequency Derived from the instantaneous frequency An alternative way to track the time and spatial variant dominant around the lobes of reflection strength frequency change of the seismic wavelet

THE LEADina tocε MAY 1997

Table 3 Summary of multitrace-windowed attributes

Correlation KLPC1 The first pnnapal component of multrtracβ and Used as a measure of linear coherence of events in the multitrace zero-lag cross correlation matrix KLPC reprewindow The standard value is 1, and lower values indicate degrees of sents the rjnncipalanarysis method or K-L trans- discontinuity or incoherence such as geological events with steeper fcrm which stemmed from Karhunen and Loev dips or more random noise Useful in detecting seismic discontinuity There are eight options of trace patterns avail like faults and unconformities able for all these mult-trace attribute extraction.

Correlation KLPC2 The second pπnαpal component of multi-trace While the seismic data are pπmanfy characterized by correlabon KLPCt and zero-lag cross correlation matrix correlation KLPC2 gives a secondary indication of residual features in the data, usually the displayed feature is similar to KLPC1 feature but the value range is different. Table 3. Summary of multitrace-windowed attributes

Name Description Applicability

Correlation KLPC3 The third principal component of ulti trace It gives a third indication of residual features in the data. and zero-lag cross correlation matrix

Correlation KLPC ratio The ratio of differences in those principal A comprehension of the correlation KLPC1, KLPC2, and KLPC3, components, (PC1-PC3)/(PC1-PC2) usually resulting in the feature characteπzation close to that of correlation KLPC1

Correlation length Average distance (in traces) for cross correlaAn indicator of lateral continuity Useful in determining the persistence tion of all traces to drop to a value of 05 If of faαes (especially shales) within the window interval. correlation is still above 05 at the edge of the trace window, the distance is given as half the ' interval length.

Average correlation Average of the cross correlations within the | Uselul in detecting seismic discontinuities, different trace patterns may trace window (excluding the autocorrelation of indicate anisotropy of the discontinuities central trace)

Concentrative correlation j More weight is given to traces adjacent to the Useful in detecting seismic discontinuities. This allows traces closer to I central trace m summing the cross correlations I the central trace to influence the result, thus improving the image where more than three traces are used.

Correlation kurtosis Fourth power of average correlation in the Useful in enhancing seismic discontinuities, often for enhancing low trace. correlation spots or lineaments.

Minimum correlation The minimum of the cross correlation within Useful in detecting seismic discontinuities and indicates the dominant the trace window seismic discontinuity around each trace

Maximum correlation The maximum of the cross correlation within Often used with minimum correlation, and their difference helps with the trace window seismic discontinuity inteφretation

Semblance coefficient The energy of the stack normalized by the The deviation from the unity value indicates the degree of difference mean energy of the components of the stack between traces and can indicate geologic and or stratigraphic The standard value is unity discontinuities

50 single trace attributes, and another dozen multi-trace attributes. Each of these attributes is normally expressed visually as needed for inspection and interpretation. Perhaps the most common attributes are: amplitude, frequency, coherence, and the like; and the Complex attributes as described by Tanner (1979), which include instantaneous frequency, instantaneous phase, and absolute amplitude. Such attributes are not easily examined more than one or two at a time.

In many cases there is additional information available beyond the seismic survey data, for example when there are existing wells on a potential reservoir. This may occur, for example, when seismic exploration is used to further define an existing field. Wells drilled for hydrocarbon exploration and exploitation are "logged" by means of a 'suite' or 'ensemble' of logging tools. These tools measure electrical, physical and radioactive properties of rocks, and the measurements are recorded at the surface. A wealth of information is typically available. Unfortunately little of the information relates directly and unambiguously to the properties which are needed. The log data must be inteφreted to obtain estimates of: amount of hydrocarbon bearing sand, porosity, fraction of water in pore space, permeable intervals, hydrocarbon type, etc.. Frequently, individual layer characteristics will allow some formation layers to be identified in many wells so that such layers can be mapped in much the same way as seismic data is mapped. Of course well information is generally very sparse (in a spatial sense) compared to seismic data sampling.

Examples of log types are: spontaneous potential, gamma ray, electrical resistivity, sonic, formation density, and neutron. Spontaneous potential (S.P.) logs measure potentials generated in bore holes by the flow of electrically conductive mud filtrate into permeable formations. Shale/sand boundaries have electrochemical effects that also produce potentials measurable by an S.P. log. The S.P. log is therefore an indicator of the presence of permeable formations, although the actual permeability value is generally not discernable from the S.P. log. Formation water resistivity can be calculated from good quality S.P. logs and this is valuable in resistivity log inteφretation when formation water samples are not available.

Gamma ray logs detect the natural radioactivity of rocks. Radioactive elements (mainly potassium) are concentrated in shales and are rare in clean sands. The log is therefore predominantly a qualitative indicator of shale (impermeable) and sand (permeable) intervals. In the North Sea, micaceous sands can give a strong radioactive signal. Electrical resistivity logs measure the electrical resistance of the formation and contents.

All reservoir rocks contain some formation water (a conductive brine) and the electrical resistance depends strongly on the proportions present of nonconductive rock, non-conductive hydrocarbons and conductive brine. Since pore space not occupied by conductive water must be occupied by nonconductive hydrocarbon, resistivity logs give an indirect estimate of hydrocarbon content.

Sonic logs record the time interval necessary for a compressional acoustic wave to travel through one foot of formation. Because of differing velocities through the solid rock matrix and the fluid in the pore space, the 'interval transit time' is related to the proportion of fluid present, and so to the porosity. Formation density logs measure the response of the formation to bombardment by medium energy gamma rays emitted from a radioactive source. Secondary rays scattered by the formation are detected by the tool, the degree of scattering being dependent upon the porosity of the formation.

Neutron logs detect and measure the high energy gamma rays emitted when high energy neutrons are 'captured' by a hydrogen atom. The formation is bombarded with high energy neutrons from a source in the tool. The log response is a hydrogen estimator, and since hydrogen is associated either with water or with hydrocarbons the tool essentially detects porosity. It is not possible to distinguish oil from water in the pore space with this tool, but it may be possible to distinguish gas from oil. Inteφretation of geophysical (seismic or well log) data has historically been by visual inspection of seismic and well data. Traditionally this inspection has been limited to a 2D representation of the information, but recent technology advances have permitted 3D and 4D representation of the data. 4D representations generally show how the reservoir rocks have changed over a period of time; perhaps due to the production of hydrocarbons.

The ability of a human inteφreter to understand the geologic information buried in the huge quantities of data that he is presented with, varies with the "skill" of the inteφreter.

Feagin (1981) showed in a study of 29 inteφreters, who worked with 37 different seismic sections that while there is wide variation in individual ability, the perceptions are heavily influenced by the way in which the information is conveyed to the inteφreter. Feagin limited his study to different visual methods of display.

Additionally in the book "Inteφretation of three dimensional seismic data" by Alistair

Brown, there is discussion about the use of different color schemes being helpful to an inteφreter in understanding the attributes of seismic data. 3D inteφretation of combined (integrated) seismic and well data is possible with current technology, and an example of one such visual display type is shown in Figure 5. Note that current techniques and computer capability now allow the geophysicists to have access to many different data types at the same time. This may become confusing, and often some data is removed. In Figure 7 there is no seismic data displayed, but rather some of the inteφretations which have been made from the data. It is desirable to make some of this additional information still available to the geophysicist as they examine the displayed information. However, showing the original seismic data would obscure the inteφretation which has been made.

The geophysical inteφreter and geologist must gain knowledge from many different information sources. Only a limited number of the data sets can be inteφreted at any one time using by visual inspection of the information. It is desirable to provide a method which assists the inteφreter in understanding different aspects or attributes of the information during a single examination of data. This method may advantageously be beneficial to other fields of endeavor where exploitation of large, interrelated data sets is desired.

The following references provide additional information regarding seismic and well data inteφretation:

Anstey N.A. "Seismic Interpretation: The physical aspects" 1977 published by

International Human Resources Development Coφoration

Brown A.R., "Interpretation of three dimensional seismic data: AAPG Memoir 42" published by the American Association of Petroleum Geologists. Coffeen J.A. "Seismic on Screen : An introduction to interactive interpretatio ^'' 1990 published by PennWell Publishing Company. Doveton J.H. "Geologic log Analysis Using Computer Methods " 1994 published by the American Association of Petroleum Geologists. Feagin F.J. "Seismic data display and reflection perceptibility", Geophysics Vol. 46,

1981, pp. 106-120. Nelson, Jr., H. Roice, New Technologies in Exploration Geophysics, Gulf Publishing

Company, Houston, Texas, 1983, 281 p. Taner, M. T., Koehler, F. and Sheriff, R. E., 1979, Complex seismic trace analysis : Geophysics, 44, no. 06, 1041-1063. (* Errata in GEO-44-11-1896; Discussion in

GEO-45-12-1877-1878; Reply in GEO-45-12-1878-1878) Weimer P and Davis T.L. "Applications of 3D seismic data to exploration and production " 1996 published by the American Association of Petroleum Geologists. The above references are incoφorated herein by reference.

SUMMARY OF THE INVENTION

Accordingly, there is disclosed herein a method for allowing a user to investigate data sets and have the attributes presented in some combination of visual, audible, and haptic forms. In one embodiment, a user is provided with a 3D visual display of at least one data set attribute (of one or more data sets) and an explorer that may be directed within the visual display. The explorer has an associated analysis region that encompasses some group of the data (e.g. data samples enclosed within a volume, surface, or polyline). The explorer processes the data values enclosed by the volume to generate one or more output values, which are then conveyed to the user as corresponding values of an audible, visible, or haptic signal attribute. A sound may be generated having a frequency, amplitude, beat frequency, sounding pattern, or some other attribute whose value is adjusted to correspond to the explorer output. The user control may be vibrated or altered (e.g. given a greater motion resistance) in proportion to the explorer output. The color, transparency and texture of the explorer on the screen may similarly be modified to reflect one or more output values of the explorer. Within the audible signal itself, various explorer output parameters may be mapped to different "instruments" in an "orchestra" to provide a user with simultaneous access to a large number of parameters. It is expected that this method will greatly ease inteφretation and understanding of the data sets. BRIEF DESCRIPTION OF THE DRAWINGS

A better understanding of the present invention can be obtained when the following detailed description of the preferred embodiment is considered in conjunction with the following drawings, in which: Fig. 1 is an example of one seismic section visual display;

Fig. 2 is a second example of a seismic section visual display;

Fig. 3 is an inteφreted seismic section;

Fig. 4 is an example of a seismic section visual display employing color to assist inteφretation; Fig. 5 is an example of a visual display for integrated seismic and well data;

Fig. 6 shows one immersive system embodiment;

Fig. 7 shows exemplary explorer sample region types;

Fig. 8 shows various two-dimensional interface shapes;

Fig. 9 is a table of volumetric seismic shapes; Fig. 10 is a table of two-dimensional seismic shapes;

Fig. 11 is an example of a smooth seismic horizon; and

Fig. 12 is an example of an irregular seismic horizon having the same general shape as the horizon shown in Fig. 11.

While the invention is susceptible to various modifications and alternative forms, specific embodiments thereof are shown by way of example in the drawings and will herein be described in detail. It should be understood, however, that the drawings and detailed description thereto are not intended to limit the invention to the particular form disclosed, but on the contrary, the intention is to cover all modifications, equivalents and alternatives falling within the spirit and scope of the present invention. TERMINOLOGY AND NOMENCLATURE

As used herein, a data attribute is a measured (or otherwise valued) characteristic such as location, density, porosity, resistivity, temperature, etc. A data value is the quantity of the characteristic such as 3.2 meters, 789 kg/m³, 85%, 54 ohms/m, 85 °C, etc. A data set is a collection of data values associated with the attributes they puφort to measure. A data volume is a multidimensional volume that, having the data attributes as Cartesian axes, encloses the full range of data values associated with each axis. A data field is a reduced-dimensional representation of the data volume.

Experiential data attributes are herein defined to be those characteristics that the human senses can perceive directly. This includes color, texture, sound, smell, taste, and spatial location (position). This definition also encompasses virtual worlds where the data values for these data attributes are fictional. Experiential data attributes can be stored in compressed or translated, non-experiential forms. For example, color may be stored in terms of intensity (Red+Green+Blue), and color difference signals (Red-Green, Red-Blue). The color difference signals are not characteristics that human senses can directly perceive. The data attributes listed in Tables 1-3 are non-experiential.

As used herein, the term explorer relates to a tool used to examine an area of interest in enhanced detail. Other suitable terms could be a "feeler", a "sound", a "meter", a "gauge", an "investigator", or more generically, an "analysis device". DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

In its raw form, a data set is a tabulation of alphanumerics. This form is adequate for small data sets and for data sets reflecting simple interrelationships between a limited number of attributes. However, for large and complex data sets, the sheer volume of data tends to conceal the information of interest. For this reason, it is desirable to use many senses to examine and ultimately comprehend the data. This is accomplished by presenting the data set in a form that allows the user to interact with it. It is believed that, as long as the data set is well- presented, greater levels of interaction will induce greater levels of understanding in the user. Patterns naturally emerge in a well presented data set.

There are a myriad of ways of presenting and allowing interaction with a data set. In a preferred embodiment, the data set is presented in an immersive 3D visual environment in which the user is able to move and turn at will in any direction. This interaction mode is preferably augmented by presenting non-experiential data to non-visual senses, and by the use of a configurable data explorer which permits localized data analysis.

Figure 6 shows one immersive presentation system embodiment. A computer 602 is coupled to a virtual reality display 604, one or more speakers 606, one or more multiple- degrees-of- freedom input devices 608, and a keyboard 610. A user viewing virtual reality display 604 will be able to perceive a 3D image presented to him by the computer 602. The display 604 may include position sensors that enable the computer to determine the user's viewing direction, and thereby allow the computer to alter the image in accordance with the viewing direction to simulate the viewing of a physical scene. The one or more speakers 606 are preferably arranged to present a 3D sound field to the user. In such a configuration the speakers 606 can operate cooperatively to create the illusion of a sound source at any desired position relative to the user. The input devices 608 and keyboard 610 are used to interact with the immersive environment presented by the computer as described further below. It is noted that the input devices 608 may be "force-feedback" devices designed to vibrate or exert a dynamic and programmable force on the user's hands as the user grips the devices. Novel device designs which present other sensations to the user may also be used in connection with the preferred embodiment, e.g. heat-conveying surfaces, programmable texture surfaces, inflatable chambers for changing grip pressure on enclosed digits, etc.

Before proceeding, a note is made regarding the use of four or more speakers (physical sound sources) to present a 3D sound field. The speakers are arranged as the vertices of a volume (e.g. a tetrahedron or a cube) enclosing the user's head. A virtual sound source located within the volume is simulated by each of the speakers producing a portion of the sound energy at an amplitude and phase that is determined by the position and loudness of the virtual sound source relative to each of the speakers and relative to the user. A virtual sound source located outside the volume is similarly simulated by each of the speakers around one face of the volume.

The loudness of the simulated sound source is (within limits) preferably scaled according to a power law such as, for example, the inverse square distance relationship. Other distance relationships may be desired, including non-power law relationships. Such relationships may improve the user's ability to discern distant events or distinguish between closely located events.

Computer 602 executes a software application that presents a 3D image to the user via display 604. The software application preferably provides the user with a virtual explorer that analyzes data associated with the 3D image. The ensuing discussion focuses on the explorer type, explorer function, explorer operation, and explorer output. It is noted that support is preferably provided for the concurrent use of multiple explorers, as this may permit multiple people to interact with the data set at the same time, or may allow one user to perform multiple analysis tasks at once.

Herein, "explorer type" refers to the selected data attributes and the selected sample region associated with the explorer. By way of example, an explorer may analyze spatial attributes such as measured depth, true depth, and off-axis distance; logged attributes such as density, resistivity, and porosity; attribute relationships such as differences, standard deviations, and ratios; and processed attributes such as weighted averages, scaled distributions, and products. The shape and orientation of the sample region is preferably adjustable. Fig. 7 shows illustrative options for sample region shapes. They include a point, a line, a surface, a volume, and a directional volume. For explanatory puφoses, each of them is shown relative to a proposed well track.

The point sample region simply captures a single representative data value. Where necessary, this data value may be determined by inteφolation from the nearest measured data values. The line sample region captures an odd number of representative data values located along a line such as a well track. In this and other sample regions where multiple data values are captured, the data values are combined to form a single output data value. This combining may take the form of simple or weighted averaging, maximum or minimum value, range or standard deviation of values, among other forms. The orientation of the sample regions may be dynamically controlled by the user, or alternatively may be set to a programmed orientation or constrained relative to an object such as a well track or a surface.

The surface sample region may be an ellipse or rectangle. It may be oriented peφendicular to a linear object or may conform to a surface such as an interface. The size and aspect ratio may be varied. Explorers having this sample region shape may be useful for identifying formation strata. The volume sample region may be a cylinder, ellipsoid, parallelogram, or other shape. These sample regions may be useful for identifying producibility. Directed volume regions include pie-shaped wedges and cones. These may be distinguished from normal volumes by the placement of the reference point. Normal volumes have their center as a reference point, whereas directional volumes have a vertex or edge as a reference point. These may be useful for identifying production zones.

Herein, "explorer function" refers to the way the captured data values in the sample region are processed to determine an output value. The processing may include scaling, simple averaging, filtering, thresholding, derivative and integral calculations. Other potential processing includes root mean square, average absolute differences, median, trend, instantaneous frequency, instantaneous phase and coherence. These explorer functions are preferably available as pre-programmed explorer options, but it is expected that these functions will also be programmable.

Herein "explorer operation" refers to the way the sample region is moved and oriented. The explorer sample region may be "snapped" to grid points, attached to objects, and/or otherwise constrained, e.g. conformed to object surfaces or oriented peφendicular to object axes. The explorer may be connected to the user's point of view, i.e. moving along with the user, or it may be external to the user, i.e. movable within the data set separately from the user's point of view. In addition, the motion may be automated (controlled by the computer in a predetermined pattern) or manually guided (dynamically controlled by the user). Herein "explorer output" refers to the way that the output value determined by the explorer function is communicated to the user. It is expected that the explorer will generate multiple output values, either because it is operating on multiple data attributes, or because multiple values associated with a given attribute are generated from the values captured in the sample region. The explorer output values may be communicated by sight, sound, or feel. Given the large numbers of different types of information that may be extracted from seismic data, inteφreted seismic data, and well information, it is desired to use multimedia means to assist the geophysical inteφreter in understanding different aspects of the information simultaneously. For example, if the user is inspecting certain well information visually, he can be made aware of the information contained in the seismic data at the well location at the same time as he "inteφrets" the well data through the use of his other senses. It is noted that the information content of two data types or data sets may be different. Understanding the differences in combination may assist in determination of the true changes in the subsurface geology. The integrated "display" preferably employs multimedia techniques to assist the inteφreter by allowing the inteφreter to simultaneously examine additional data types. As the inteφreter examines a visual display, the inteφreter can touch, point at, or otherwise select a portion of the display, the selected explorer may exhibit changes to multiple visual properties in order to convey output values when the sample region is located at the region of interest. For example, the explorer may have the appearance of a sphere that exhibits changes in color, texture, transparency, brightness, size, and shape. Each of these explorer attributes may reflect the value of a respective output value corresponding to a respective data attribute in the data set.

In another embodiment, sound is employed to communicate additional information to the inteφreter. As the inteφreter examines a visual display, the inteφreter can touch, point at, or otherwise select a portion of the display, and a sound is provided to communicate the additional information corresponding to that point. A single note, or "voice", can have multiple attributes including frequency, waveform, amplitude, pulse repetition rate, pulse duty cycle, beat rate, and beat frequency. To convey even more information, multiple voices may be used. These voices may correspond to familiar instruments such as trumpets, clarinets, violins, guitars and percussion instruments, among others.

In another embodiment, haptic sensation is used to communicate additional information to the user. As the user identifies a region of interest, parameters of the user's control device may be modified to increase motion resistance, induce vibration, alter the texture, and/or induce temperature or pressure changes. Of course, the visual, audio, and haptic attributes may also be used in any combination to convey the explorer output values to the user.

It may be advantageous to use sound for conveying the following attribute types: Scale, Time, Topography, Volumes, Process, Shape, Uncertainty, and Data Quality. All geologic attributes can be thought of as being on some "scale". Scales can be large, for example general basin shape, or small, for example, the characteristics of a thin section which might be examined under a microscope. It is often difficult to understand the information which might be conveyed by a thin section, while (for example) looking at data from a well log. Sound could be used to characterize the properties of the rock matrix as revealed on a microscopic scale, while the well log was being inspected. In other words, sound can convey information available at different scales from the one being observed visually. In one embodiment, sound amplitude may be used to communicate porosity.

In this context "time" has three specific and different meanings. One is related to the travel time of seismic data. Another is related to the geologic time associated with particular rocks or depositions. A third is related to the time in which human activity occurs. It is often not easy to understand the geologic time of a seismic event at a particular seismic time. However, if geologic time were associated with a particular sound scale, it would be easy to understand what geologic time was relevant when inspecting a particular seismic event at a particular seismic time. This would potentially have great value to the inteφreter. In one embodiment, sound frequency may be used to communicate travel time or geologic age or production history.

Surfaces, such as those derived from seismic, well, or other geophysical or geological data, may be characterized as having different sound properties. Seismic horizons represent geologic interfaces in the earth. Surfaces separating wind speed define the Jet Stream. The properties of the surface often have to be inspected in detail in order to gain a specific understanding. In two dimensional earth models, localized areas tend to have specific horizon topographies such as those shown in Fig. 8: flat (includes sloped geometries), curved (includes wavy geometries), and faulted. Each of these topographies may be characterized by a corresponding sound. In one embodiment, these topographies may be associated with "sound bites" such as white noise, a "twang", and a crunching noise, respectively. In another embodiment, these topographies may be associated with a beat frequency, where the frequency indicates the shaφness of the curve or offset of the fracture, and the beat amplitude indicates the local slope of the curve. It is easy to gain an understanding of the type of surface that is being examined by listening to the sounds. Geologic volumes of the subsurface may be characterized with sound. For example, in some geologic settings, the presence of hydrocarbon can be indicated by a "bright spot" or high amplitude seismic reflection. This is distinct from the shape of the topography, but related to the amplitudes of the seismic reflections in the volumes. In one embodiment, these bright spots may be associated with sound source distance and direction. In a 3D virtual reality environment it would be possible to immerse the inteφreter in the earth, and allow him to listen to the bright spots (or other features) which surround him. By mapping the location and magnitude of a bright spot as a sound source, it will be possible to listen to and locate the bright spots in the earth as the inteφreter moves through the virtual earth model. This application may require the use of quadraphonic (or octaphonic) sound projection around the inteφreter.

Geologic Processes - such as sedimentary deposition, carbonate reef growth, salt movement, and tectonics - are naturally presented with sound. Just as we can imagine the speed of wind by an appropriate sound, we can see how sound can define the differences in grain size, amount of deposition and constituent components of deposition in a landlocked lake verses the mouth of a major river system verses a large deepwater turbidite system. The time aspects and physical characteristics of other geological processes can be similarly tagged and presented in a multi-media environment. Like music, which requires sound across time, geological processes require sound to understand how the process changes across time.

While similar to topography, Shape is intended to characterize the volumetric "shapes" of different depositional environments and even different reservoir styles. It can be thought of as an extension to topography. Some characteristic geologic shapes are shown in Figs. 9 and 10. By mapping the shapes to particular tones or sounds it will be possible for the inteφreter to understand the large scale picture (of what type of geologic volume he is examining) while inteφreting the small scale picture of local features within the earth. In one embodiment, the shapes may be associated with wave shapes (harmonic frequency amplitudes and phases).

Not only are the exemplary sound characteristics used to describe the various information types distinct, variations of tone, beat, timbre, instrument, pitch and other characteristics can be used. For instance attributes of the seismic or well data may be mapped to several characteristics of a sound. For example, a single frequency sound wave may be changed by altering the frequency of the sound (pitch), and the magnitude of the sound (loudness), among other things. A single frequency sound wave may be modified with a second frequency to set up a "beat". This means that the sound can be modified in two additional ways: beat frequency, and beat magnitude. Any method that allows conversion of one or more sets of attribute data to sound could be used. Additionally it is not necessary to distinguish the primary frequency. One attribute could be matched to "middle C" with the harmonics associated with a piano, and another mapped to "middle C" with the harmonics associated with a trumpet. Each "instrument" can be assigned a playing or "sounding" pattern, so that, e.g., the trumpet plays a sustained whole note while the piano sounds alternating quarter and sixteenth notes. The sounds could be distinguished easily. It is not difficult to see how multiple data attributes might be mapped to an orchestra.

Understanding different attributes of geophysical and geological data at the same time is made easier by allowing the inteφreter to look at information from one attribute and listen to information from one or more other attributes. For example, it would be possible to "scan" a cursor down data from a well log and observe the changing character of the well data, at the same time as listening to sound which relays information about the seismic data at the same location. The sound signal information could be as simple as a monotonic frequency, which was modulated in amplitude, based on the amplitude of the signal in the seismic data. More that one attribute of the seismic signal could be translated into a sound signature, for example, by mapping amplitude to magnitude of the signal, and mapping data frequency to the frequency of the sound. If different attributes were mapped to different tonal qualities of a signal, information could be understood much as the listener of music is able to separate out the different instruments, which are playing concurrently. In this way many attributes of the geotechnical data can be examined simultaneously.

The potential for multi attribute analysis can be easily seen (and heard). Imagine looking at a well log, and interrogating the log for information. For example pointing at a particular position on the log graph, and seeing a numeric value associated with the log value at the point which was identified. It would be possible to map (translate) the values of the log attribute to sound in several different ways. For example:

Log value maps to amplitude of sound wave of fixed frequency

Log value maps to frequency of sound wave of fixed amplitude

Log value maps to other sound attribute of sound wave. For example pitch or "beat" of a sound sequence or musical segment.

Listening to two logs or portions of two logs simultaneously and then moving one log up or down in depth, relative to the other log, allows an auditory correlation of the same geologic response in different wells. This is an important advance, simply because humans have orders of magnitude more dynamic range in our hearing than we do visually. It is also possible to map different data attributes to different sound attributes. For example, it would be possible to look at the S.P log, to inteφret information on permeability, while listening to a sound signal which had a frequency related to the gamma log, and amplitude related to the resistivity log. This means that on a single inteφretation the geoscientist would have information about permeability, sand/shale content and water presence conveyed at the same time. It is possible to utilize sound in such a way that conditions which are favorable for hydrocarbon presence may be easily and instantly identified.

Multi-attribute analysis of seismic data has big potential for the geophysical inteφreter. The ability to look at the amplitude information (for example) and listen to the instantaneous phase (for example) permits the geoscientist to gain knowledge about two or more different attributes of the data at the same time. For example, seismic reflectors often have large amplitudes (bright spots) and phase reversals at the edge of the direct hydrocarbon indicator.

Looking at well data while listening to seismic data at the well location at the same time is similar to performing multi attribute analysis of different well logs. The seismic data, or attribute(s) of the seismic data can be mapped to sound while the well logs are inspected visually.

It is possible to characterize (inteφret) the whole of a seismic volume by use of sound, and therefore allow the seismic inteφreter to gain an understanding of the volume complexity from a single inspection point. This will assist in gaining information quickly. For example, if the geoscientist is examining visually a seismic trace, or series of traces, or a well log, at a point that the data coincides with an inteφreted seismic horizon, an audible analysis of the event in 3 dimension can be supplied from a single point. An event can be characterized (for example) by: -Average time of the event represented by the frequency of the sound. -Average amplitude of the event represented by the amplitude of the sound wave. -Continuity of the event represented as a beat frequency. Continuity may be defined as the "spread" of second derivatives at any point on the surface, or perhaps the standard deviation of the times from the mean. This characterization would (for example) make it easy to distinguish between the surface displayed in Figs. 11 and 12. In this example it is likely that the average time of the event is unchanged, and the amplitude is unchanged, but the standard deviation of the time of the event is very different. The two surfaces might be characterized by entirely different waveforms. In one embodiment, the multimedia data presentation method provides the following features: i. Using sound for conveying seismic data information to an inteφreter. ii. Using sound for conveying well information to an inteφreter. iii. Using sound for conveying seismic inteφretation information to an inteφreter. iv. Using sound for conveying seismic attribute information to an inteφreter. v. Using sound for conveying seismic information to an inteφreter at the same time as information is conveyed visually, vi. Using sound for conveying well information to an inteφreter at the same time as information is conveyed visually. vii. Using sound for conveying seismic inteφretation information to an inteφreter at the same time as other information is conveyed visually, viii. Using sound for conveying seismic attribute information to an inteφreter at the same time as other information is conveyed visually. ix. Using sound for conveying seismic information to an inteφreter at the same time as other geophysical or geological or engineering data is conveyed visually. x. Using sound for conveying well information to an inteφreter at the same time as other geophysical or geological or engineering data is conveyed visually. xi. Using sound for conveying seismic inteφretation information to an inteφreter at the same time other geophysical or geological or engineering data is conveyed visually, xii. Using sound for conveying seismic attribute information to an inteφreter at the same other geophysical data is conveyed visually. While there are many contemplated methods of mapping sound to other attributes, even with the simple examples described it is clear that shape to sound mapping can be done. The following advantages may be realized by some embodiments:

1. The same information can be conveyed to an inteφreter in two different ways: visually and audibly. 2. Two or more different attributes of a data set can be conveyed at the same time.

3. Two or more different data information sets can be conveyed at the same time.

4. Areal information can be conveyed while evaluating local information.

5. The technique can be implemented very effectively on any computer that is able to process and present sound since the various sound qualities can be easily computed. 6. The technique only adds information, current methods do not require any change.

7. Data from different geophysical data sets can be inteφreted together, without requirement to "look" at each data set separately.

8. Better intuitive understanding of the relationship between different attributes of geophysical data results from the ability to inteφret several attributes at the same time.

It is contemplated to employ the disclosed method in Virtual Reality (VR) systems for the inteφretation and understanding of geophysical data (either seismic data or well data). The method employs the use of senses beyond simple visual examination of the data to provide a significant benefit in understanding and inteφretation of the information. In particular, sound is used to assist the inteφreter in gaining a better understand of multiple geophysical attributes. The inteφreter can (for example) look at the information from a single well log, and listen to the information from another type of well log at the same time. Seismic data can also be mapped to sound, either on a trace by trace basis, or as a mapping of seismic "events" to particular tones or rhythms. A musical representation of a 3D seismic volume, as well as multiple time-lapse 3D seismic volumes, is contemplated.

Claims

WHAT IS CLAIMED IS:

1. A method of presenting data, wherein the method comprises: determining a region of interest within a data field having one or more data attributes; and generating a sound having a perceptible attribute indicative of one or more data values associated with a data attribute in the region of interest.

2. The method of claim 1, wherein the perceptible attribute is indicative of data values associated with only one data attribute.

3. The method of claim 1, wherein the sound has a second, different perceptible attribute indicative of one or more data values associated with a second, different data attribute in the region of interest.

4. The method of claim 1, further comprising: providing a visual image indicative of data values associated with one or more data attributes of the data field.

5. The method of claim 4, wherein the visual image is indicative of data values associated with experiential data attributes of the data field.

6. The method of claim 5, wherein the experiential data attributes include spatial location.

7. The method of claim 1, wherein the perceptible attribute is a sound frequency.

8. The method of claim 1 , wherein the perceptible attribute is a sound volume.

9. The method of claim 1 , wherein the perceptible attribute is a beat frequency.

10. The method of claim 1 , wherein the perceptible attribute is a beat amplitude.

11. The method of claim 1 , wherein the perceptible attribute is a waveform shape.

12. The method of claim 1, wherein the perceptible attribute is a sound pulse repetition frequency.

13. The method of claim 1 , wherein the perceptible attribute is a sound pulse duty cycle.

14. A method of presenting data, wherein the method comprises: determining a region of interest within a data field having one or more non-experiential data attributes; and generating a tactile sensation having a perceptible attribute indicative of one or more data values associated with a non-experiential data attribute in the region of interest.

15. The method of claim 14, wherein the perceptible attribute is indicative of data values associated with only one data attribute.

16. The method of claim 14, wherein the sound has a second, different perceptible attribute indicative of one or more data values associated with a second, different data attribute in the region of interest.

17. The method of claim 14, further comprising: providing a visual image indicative of data values associated with one or more data attributes of the data field.

18. The method of claim 17, wherein the visual image is indicative of data values associated with experiential data attributes of the data field.

19. The method of claim 18, wherein the experiential data attributes include spatial location.

20. The method of claim 14, wherein the perceptible attribute is a vibration frequency.

21. The method of claim 14, wherein the perceptible attribute is a vibration amplitude.

22. The method of claim 14, wherein the perceptible attribute is a surface temperature.

23. The method of claim 14, wherein the perceptible attribute is motion resistance.

24. The method of claim 14, wherein the perceptible attribute is a surface texture.

25. The method of claim 14, wherein the perceptible attribute is a grip pressure.

26. The method of claim 14, wherein the perceptible attribute is an air velocity.

27. An information carrier medium configured to communicate a software tool to a computer, wherein the software tool comprises: a user interface configured to display a graphical representation of a data set, configured to receive user input indicative of a desired explorer having an associated analysis region, and further configured to position the explorer in response to user input; and an analysis module configured to process data values in the analysis region to determine a property of the data set in the analysis region.

28. A method of examining information, wherein the method comprises: selecting an explorer having an associated analysis region; and moving the explorer in a data field to analyze a region of interest.

29. The method of claim 28, wherein the analysis region is one of a set of analysis regions consisting of a point, a line, a surface, and a volume.

30. The method of claim 28, wherein the motion is unrestricted along display dimensions of the data field.

31. The method of claim 28, wherein the motion is constrained to motion along an interface in the data field.

32. The method of claim 28, wherein the motion is constrained to motion along an axis of an object in the data field.

33. An information carrier medium configured to communicate a software tool to a computer, wherein the software tool comprises: a user interface configured to display a graphical representation of a data set, configured to receive user input indicative of a desired explorer having an associated analysis region, and further configured to position the explorer in response to user input; and a sensory output generator configured to generate a sound having a perceptible attribute indicative of all data values associated with a non-experiential data attribute in the analysis region.

34. The medium of claim 33, wherein the sensory output generators combines a plurality of sound sources to form an output sound.

35. The medium of claim 33, wherein the volume of the sound source is determined according to a distance from the explorer.

36. The medium of claim 33, wherein the computer includes multiple speakers, and wherein each speaker is provided with a sound source intensity based on the user-source distance, the source-speaker distance, and based on the user-speaker distance.