This section provides a Select Sesimic... button which will allow you to pop-up a seismic selector dialog. This section also allows you to specify the 3D Survey Area to be used to retrieve trace data for Coloured Inversion analysis.

Figure 4.24. Select Input Data dialog (Input Seismic tab)

The figure above shows the top portion of the Input Seismic tab where seismic data volume has been pre-selected. This is the recommended method when running SciQt as an OpendTect plug-in. Whilst it is highly recommended, it is not necessary to pre-select the input "stored cube" via the OpendTect "Attribute Set" dialog prior to launching the SciQt plug-in. If the selected seismic is not pre-selected then you can always select the "stored cube" by clicking the Select Seismic... button.

With the seismic data selected, it is usually desirable to define a sub-area. This is achieved by clicking the Sub button from the 3D Survey Area radio button group. Following this, the user would specify the sub-area using the Start In-line, End In-line, Start X-line and End X-line edit boxes. Finally, the number of traces is specified and the Load Seismic button is clicked.

With the seismic data volume selected you can now load a set of traces to be analysed. Here you can optionally choose to select traces from the full 3D survey area or sub 3D survey area. This is achieved using the 3D Survey Area radio button group as follows:

Typically you would select a single sub-area. Specifying a sub-area in this manner enables you to include an area of the survey where the trace data is known to be of good quality and is representative of the survey area as a whole. With sub-area selected it is simply a case of specifying the number of traces to load (default 40) via the Num. Traces spinbox widget. This is then followed by you clicking the Load Seismic button to load a random set of traces. A list of the randomly loaded traces is then displayed in the Input tab beneath the Load Seismic button. At the same time an alternative set of traces is also loaded. These alternative traces are listed in the QC tab and are used with the derived Coloured Inversion operator for QC purposes. The software also enables you to select trace data from multiple sub-areas. This is achieved by specifying another In-line/X-line Start and End range and clicking the Load Seismic button again. The goal is to generate seismic trace spectra which is a good representation of the area as a whole. If you attempt to specify an area which is outside of the available area then a warning message will be issued when you attempt to load seismic traces by clicking the Load Seismic button. You will also get an error message if you set an area which would not allow enough unique traces for the Num. Traces requested.

Note: the following image illustrates the Landmark R5000 version of the Select input Data dialog which includes a Select Survey... button and text area.

Figure 4.25. Loading Seismic Data

The image in the figure above shows the SciQt Main Window with Select Input Data Dialog superimposed. The Raw Seismic spectra resulting from clicking the Load Seismic button on the "Select Input Data Dialog" (above left) with Num. Traces set to 40 and a 3D Survey Area is defined by:

In-line: 1400 - 1800

X-line: 4300 - 5400

By default, all loaded traces are used in the spectra calculation and contribute to the Seismic Mean Spectrum. However, you can remove a given trace from the SciQt analysis by deselecting it via the Input list. Deselection of a trace is simply achieved by unchecking the checkbox within the list. If you want to completely remove the deselected trace you click the Clear Unselected Traces button. Alongside each trace on the Input list, are various properties associated with the trace. These include Trace Id (or trace label), In-line, X-line, RMS Amp, RMS Err, Horz Time. The RMS Amp property is derived from the input seismic trace, whereas the RMS Err property is derived from the trace spectrum. The latter can be considered as a "goodness of fit" parameter as it compares trace spectrum with the smooth mean trace spectrum. So for RMS Err, the smaller the number the better the fit. Loaded traces, by default, are sorted by Trace Id. However, you can sort the list of traces using one of the other properties. This is easily achieved by clicking the column header on the Input list. So, for example, if you want to sort by RMS Amp in ascending order, you click the RMS Amp header label. To sort in descending order you click the RMS Amp header label again. You can also change the position of the property columns. This is simply achieved by clicking and holding on a property header then dragging the property header to its new position.

The image shows the result of clicking the Load Seismic push button. Here 40 traces have been randomly loaded and immediately displayed in the "Raw Seismic" chart within the SciQt Main Window. Also visible is a portion of the "Select Input Data Dialog" showing the trace list on the "Input" tab sorted by ascending RMS Amplitude values.

Figure 4.26. Clearing Seismic Data

The image in the figure above shows the result of clicking the Clear Traces push button. Clear traces removes all traces previously loaded to memory and has the immediate effect of removing this data from the "Raw Seismic" chart within the SciQt Main Window.

Specifying the time range (gate) is an important step. Ideally, you should choose a time gate in the range of 500 - 1000 msec which should be over the zone of interest. You can set the Time Range mode in one of three ways via the "Range" radio button.

Of the three options the Horizon Range mode is the preferred mode since it allows the gate to follow the geology and should, with a well interpreted horizon in the target zone give, in theory, trace spectra which are more closely matched.

Note: the following image illustrates the Landmark R5000 version of the Select input Data dialog which includes a Select Survey... button and text area.

Figure 4.27. Modifying Seismic Time Range (Full)

The image in the figure above shows the SciQt Main Window with Select Input Data Dialog superimposed. Here, "Full" (full trace range) has been selected for the Range radio button. This mode is rarely used since our goal is to design a Coloured Inversion operator over the zone of interest. This will optimise the process over that zone. However, it does allow you, although insensitive, to see the start and end times of the entire trace.

Figure 4.28. Modifying Seismic Time Range (Sub)

The image in the figure above shows the SciQt Main Window with Select Input Data Dialog superimposed. Here, "Sub" (sub trace range) has been selected for the Range radio button. In this mode the Absolute Start time and the Absolute End time Line Edit fields will be sensitised. This allows you to specify a time gate for the zone of interest. Typically, you would specify a time gate which has a time range between 500msec and 1000msec.

Figure 4.29. Modifying Seismic Time Range (Horizon).

The image in the figure above shows the SciQt Main Window with Select Input Data Dialog superimposed. Here, "Horizon" (horizon relative trace range) has been selected for the Range radio button. In this mode the Relative Start time and the Relative End time Line Edit fields will be sensitised. This allows you to specify a time gate for the zone of interest. Typically you would specify a time gate which would give you a time range between 500msec and 1000msec. Specifying the Start and End times here are relative to a geologic horizon. A negative number represents time above the horizon whereas a positive number is a time below the horizon. The time gate can be either above, below or span the horizon.

Occasionally, it is necessary to remove one or more poor spectral traces which contribute to the generation of the mean spectra. This is easily achieved by identifying problem spectral traces from the chart and deselecting these traces from within the Select Input Data Dialog.

Figure 4.30. Removing Bad Traces

If one or more seismic trace spectra displayed on the Raw Seismic plot were anomalous then it is possible to remove those traces from the Seismic Mean Spectrum calculation.

Let us consider the blue trace (which appears to have a particularly low amp (dB) at around 60Hz). Now, if we wish to exclude this trace from the mean calculation, we must first identify it. This is achieved by pointing close to one of the vertexes of this spectrum and clicking on it. We need to be within few a pixels of one of vertexes to be able to identify it. If we are within range when we click then the spectra named, together with the Frequency and Amplitude of the vertex, is displayed in the Status Area of the Main Window. In this case, the raw seismic spectra is identified as "TRACE_003". We can now to go to the Input Trace list and click the checkbox to the left of entry "TRACE_003". This causes the trace to be flagged as not currently selected and the Raw Seismic plot on the Main Window is updated accordingly. The Seismic Mean spectrum (black) on both displays changes as a consequence. An alternative method, which works well in this case, is to look at the RMS Err property. You will note that the RMS Err of TRACE_003 is significantly larger than that of the RMS Err values of the other trace spectra. So, if you sort by RMS Err, you will easily be able to identify any trace spectra with an abnormally large RMS Err value. You should note that once you toggled off a trace, it may loose its RMS Err value as the spectrum is not calculated.

The Select Volume dialog provides you with a 3D seismic volume selector. From here simply select the seismic volume you require for your SciQt analysis and operator design. A volume filter is provided to reduce the list to a manageable size for projects with many volumes.

Figure 4.31. Select Volume

...

The images in the figure above show examples of the "Select Volume" dialog for selection of a seismic volume. The LH image shows a seismic volume list without a seismic volume filter. The RH image shows the use of a filter to create a shorter list. As can be seen, the wild card character (*) can be used in the Filter field.

The Select 2D Seismic dialog provides you with a 2D seismic line selector. From here you select the 2D lines you require for your SciQt analysis and operator design. A 2D line filter is provided to reduce the list to a manageable size for projects with many 2D seismic lines.

Figure 4.32. Select 2D Lines

...

The images in the figure above show examples of the "Select 2D Seismic" dialog for selection of 2D seismic lines. The LH image shows seismic lines listed without a 2D line filter. The RH image shows the use of a filter to create a shorter list. As can be seen, the wild card character (*) can be used in the Filter field. From the list of available 2D lines (optionally filtered) you can select one 2D line to be included by clicking the unchecked check box to the left of a 2D line name. Multiple 2D lines can be selected by repeating this step. All 2D lines can be selected by clicking the Select All push button. De-selection of a single 2D line is achieved by clicking a checked check box to the left of a 2D line name. All 2D lines can be de-selected by clicking the Select None push button. With the 2D lines selected you must then select a class. Optionally you can skip traces at the start and end of each 2D line. This is achieved by specifying the Traces to skip via the From Start and From End line edit fields.

The Select SEGY dialog provides you with a 3D SEGY file selector. From here you select the 3D SEGY file you require for your SciQt analysis and operator design. Unfortunately, selecting a 3D SEGY file for input here is not the end of the story. You need to also either select or build an offset mapping template for use with the selected SEGY file. Fortunately, within the software this is a straightforward task.

Figure 4.33. Select SEGY

...

The images in the figure above show examples of the "Select SEGY" dialog for selection of SEGY files and offset mapping. The LH image shows the dialog prior to opening a SEGY file. To open a SEGY file click the Open File... push button which will pop-up a file selector dialog allowing you to traverse the file system to select and open a SEGY file. The RH image shows that a SEGY file has been opened and has been matched with an offset Template. In this case, it was matched with the "Other 1" template. This dialog attempts to automatically match the opened SEGY file with one of the available templates. If the match is successful then the matched template will be highlighted and Min/Max Inline/Xline together with Start/End Time obtained from the Opened SEGY file will be displayed in the "Survey Area and Range Info" tab within the "Segy Summary" block. Similarly the opened SEGY file's EBCDIC header information will be displayed in the "Text Header" tab with the SEGY Binary header information being displayed in the "Binary Header" tab.

Now, if no template match is found then it will be necessary to create a new template. Fortunately, this is a relatively simple process and is achieved in the following manner. Select an existing template, it doesn't really matter which one you use, then click the Copy push button. Now select the new template which will appear at the end of the list of templates and will have a default name similar to "New Template 1". Finally click Edit... push button to pop up "Edit Template" dialog.

The Edit Template dialog provides you with a mechanism to edit SEGY templates. The original SEGY standard (rev 0) has been used extensively within the oil and gas industry as a convenient way to exchange seismic data between various companies. This standard was published in 1975 for storage of 2D line seismic data on magnetic tapes. Since this original publication there have been many advancements in geophysical data acquisition including disc recording and 3D seismic. The SEGY standard was subsequently used in ways that were never originally envisioned. Inevitably, this led to inconsistent use of the trace header attributes by different companies for storage of 3D seismic on disc. Whilst the SEGY standard (rev 1) published in 2002 was designed to address such issues, there is still a need to map (set up offsets) various trace header attributes as there is software and old SEGY files in existence that don't adhere to the new SEGY rev 1 standard.

Figure 4.34. Edit Template

The figure above shows the "Edit Template" dialog. This dialog allows you to assign a unique name and add a brief description. The main purpose of this editor is to allow you to set up the Template Offsets for mapping various trace attributes.

OK, let's look at how we set up the various Template Offsets. This dialog works with the opened SEGY file. Within the "Select Trace" block there is a slider and a spin box object. Both allow you to move to different traces within the SEGY file. Using these objects you can position the trace reader to any trace within the SEGY file. So for any trace you will be able to display the trace header data in the right panel beneath "Trace Header" label. By scanning the trace header for a given SEGY trace you should be able to determine where in the header various pieces of information are stored. So to edit the template offset we do the following:

Whilst ARK CLS recommends that when creating a new Template all Data items are mapped, it may be possible to generate a Template without mapping every Data item in the Template Offset list. We consider the most important items are "Inline", "Xline", "X Coord", "Y Coord", "Coord Scaler", "Start Time", "Number Of Sample", "Sample Interval" and "Trace ID Code". These "Data Items" in the "Template Offsets" list should always be set.

The Horizon Input Dialog provides you with a horizon selector. From the Horizon Data tab, simply select the horizon you require for your SciQt analysis and operator design. A Horizon filter is provided to reduce the list to a manageable size for projects with many horizons.

Figure 4.35. Select Horizon

...

The images in the figure above show examples of the "Select Horizon" dialog for selection of a seismic horizon. The LH image shows a seismic horizon list without a horizon list filter. The RH image shows the use of a filter to create a shorter list. As can be seen, the wild card character (*) can be used in the Filter field.

Loading acoustic impedance data from the OpendTect database is straightforward. The details are explained in the table beneath. Essentially, you click the Load Wells push button which will auto select all the acoustic impedance log curves in the OpendTect data repository.

Figure 4.37. Database Well Log Loading

....

The two images in the figure above show the top portion of the Input Well Log tab used to load OpendTect well log AI data. The left image shows the state when the "Input Well Logs" tab is first selected by you following display of the "Select Input Data" dialog.

Click the Load Wells push button which will auto select all the acoustic impedance log curves in the OpendTect data repository. It is assumed that acoustic impedance logs will have the characters "AI" (upper or lower case) or "IMP" (upper or lower case) somewhere within the log curve name. If, for a given well, there are no such AI log curves or the log curve doesn't meet the assumed naming convention, then the user can still select or generate AI log curves manually.

The image to the right above shows, in the central area, that two AI log curves, one each from two wells, have been automatically loaded. In OpendTect, log curves are stored in depth. Coloured Inversion requires the AI log curves to be in time. During the loading process the AI data which is stored in depth within OpendTect is depth to time converted using the associated time-depth model associated with a given well. It is then resampled to be regular in time. The sample interval used for this purpose can be changed via the "Advanced Controls" dialog, "Time domain" tab.

We can also manually select or generate AI well log. This is achieved by right clicking on well name which will pop-up a menu with Select AI Log... and Generate AI Log... menu items. Selecting the first item will pop-up the "Select Database Logs" dialog which lists the available logs allowing you to choose an AI log curve however it's named. Selecting the second item will pop-up the "Generate AI Log" dialog which displays two lists. The left hand list allows you to select a sonic log with right hand list allowing you to optionally select a density log. If no density is available, then you can select User for Density Source radio button and supplying a User Density Value. Here the AI log will be generated from the supplied sonic and density.

The push buttons Clear Wells and Clear Unselected Wells are self explanatory. The Reshow Wells push button will re-display any of the wells/logs cleared via the Clear Wells and Clear Unselected Wells push buttons. Note the Load Wells push button will change to a Reload Wells push button after the initial automatic selection. The Reload Wells push button will reload from the database all the currently available wells. In this way any new data loaded to the database will be displayed.

3.2.1.1. Select Database Logs dialog

This dialog allows you to select the Acoustic Impedance Well Log data that is to be used from the OpendTect data repository. The figure below shows the "Select Database Logs" dialog in OpendTect.

Figure 4.38. Select Database Logs (OpendTect)

This dialog provides a wild card filter field which operates on the list of logs to provide a facility to shorten the list of logs to a manageable size should this be necessary.

3.2.1.2. Generate AI Log dialog

This dialog allows you to generate an Acoustic Impedance Well Log curve from the sonic and optionally, density logs that are stored in the OpendTect data repository. The figure below shows the "Generate AI Log" dialog in OpendTect.

Figure 4.39. Generate AI Log (OpendTect)

This dialog provides wild card filter fields which operate on the two list of logs to provide a facility to shorten the list of logs to a manageable size should this be necessary.

While as sonic source the user always needs to select a Well Log, as density source there are 3 different types of source that you can use:

1. Log, which requires the user user to select a density log from the list provided below.

2. User Value, which requires the user to set a density value on the User Densityvalue field so that it will be used as constant density.

3. Gardners Equation, which requires the user to set two constants (A and B) so that the density can be calculated by applying that equation on the sonic log, Vp.

Loading Acoustic Impedance data from ASCII is straightforward. Simply click on the Load ASCII Wells... push button to pop up a file selector. Within the File Selector you can traverse the file system to locate Acoustic Impedance data in ASCII format. The File Selector allows one or more ASCII files to be open at once. Use the <Shift> key modifier to select consecutive files and the <Ctrl> key modifier to select non-consecutive files from within the File Selector dialog. You can also MB1 and drag the mouse over a number of ASCII files.

Figure 4.40. ASCII Well Log Selection

The picture in the figure above shows the Input Well Log tab, where well log data has been loaded. In this case the well log data was loaded from ASCII files. The central list area shows four impedance curves have been loaded. One each for four wells. Normally, well log ASCII data files do not contain inline, xline or horizon data. So, in such cases, it is not possible to select a time range relative to an interpreted horizon. However, SciQt recognises certain XML style meta tags, which can be used by you to provide additional data. Below are the five XML style meta tags recognised by SciQt.

<inline>inline data here</inline>

<xline>xline data here</xline>

<horztime>horizon time data here</horztime>

<xcoord>x coordinate here</xcoord>

<ycoord>y coordinate here</ycoord>

These XML meta tags are normally inserted on one line within the ASCII file. For example, in Landmark style ASCII files these would normally be placed on the second line (which is blank). It is not necessary to include all the meta tags. Normally, only the <inline> and <xline> meta tags are supplied as there is then sufficient information for horizon data to be extracted from the supplied horizon. In this example, the horizon information for well_4.dat, well_3.dat and well_3.dat can be obtained from the supplied horizon (Horz Scr = horz) using the In-line and X-line data supplied via the meta tags <inline> and <xline>. For well_1.dat however, the horizon information comes from the <horztime> meta tag (Horz Scr = well) since no In-line or X-line data has been supplied.

This section describes how you can define the format for ASCII file reading of well AI log data. Clicking the Define Format... push button will pop up the "Define ASCII File Format" dialog.

Figure 4.41. Define ASCII File Format

The figure above shows the "Define ASCII File Format" dialog. This dialog has three groups: "Select ASCII Format", "Record and Column Information" and "File Time Column Units".

Select ASCII Format: This group contains a radio button with two options Landmark ASCII and User defined. In Landmark ASCII mode (default) the other two groups are desensitised. In User defined mode you can supply additional information in the "Record and Column Information" and "File Time Column Units" groups. Although no explicit LAS input facility is provided, you can normally read LAS files in User defined mode.

Record and Column Information: If your ASCII has header records before the wireline curve data you should skip these by specifying a value in the No. of Recs to Skip input field. You should also specify which column the time data and the impedance data can be found. By default, these are set to 1 and 2 respectively.

File Time Column Units: This radio button item allows you to specify whether the time data is in milleseconds or seconds.

Specifying the time range (gate) is an important step. Ideally, you should choose a time gate in the range of 500 - 1000 msec which should be over the zone of interest. You can set the Time Range mode in one of three ways via the "Range" radio button.

Of the three options the Horizon Range mode is the preferred mode since it allows the gate to follow the geology.

Figure 4.42. Modify Well Log Time Range (Full)

The image in the figure above shows the SciQt Main Window with Select Input Data Dialog superimposed. Here, "Full" (full trace range) has been selected for the Range radio button. This mode is rarely used since our goal is to design a Coloured Inversion operator over the zone of interest. This will optimise the process over that zone. However, it does allow you, although insensitive, to see the start and end times of the entire trace.

Figure 4.43. Modify Well Log Time Range (Sub)

The image in the figure above shows the SciQt Main Window with Select Input Data Dialog superimposed. Here, "Sub" (sub trace range) has been selected for the Range radio button. In this mode the Absolute Start time and the Absolute End time Line Edit fields will be sensitised. This allows you to specify a time gate for the zone of interest. Typically, you would specify a time gate which has a time range between 500msec and 1000msec.

Figure 4.44. Modify Well Log Time Range (Horizon)

The image in the figure above shows the SciQt Main Window with Select Input Data Dialog superimposed. Here, "Horizon" (horizon relative trace range) has been selected for the Range radio button. In this mode the Relative Start time and the Relative End time Line Edit fields will be sensitised. This allows you to specify a time gate for the zone of interest. Typically, you would specify a time gate which would give you a time range between 500msec and 1000msec. Specifying the Start and End times here are relative to a geologic horizon. A negative number represents time above the horizon whereas a positive number is a time below the horizon. The time gate can be either above, below or span the horizon. Clicking the Select Horizon... push button will pop up the Horizon Input Dialog.