Update home for Field Geophysical Software Suite that accompanies
Introduction to Applied Geophysics: Exploring the Shallow Subsurface,
by Burger, Sheehan, and Jones
Mac OS 10.7 (Lion) users: The original software distributed on the CD-ROMs will not run under this OS (it was Power-PC code originally). However, if you update that version using the updaters below, you should be able to then run the software under this OS.
With versions of the compiler since late 2011, support for PPC code was dropped. Because compiler updates usually improve compatibility with the newer systems, I am not attempting to maintain PPC updates after the early 2012 releases (should some critical problem emerge, I might revisit this--I do have a copy of the last PPC-capable compiler).
Windows users: In general, the updated versions of the programs may not run on pre-XP versions of Windows.
View the help topics for the software (includes
some FAQs; also lists some changes for updated versions of applications)
Patches for Windows version of GravMag, Refract, Resist:
Patches for Mac versions of Field Geophysical Software Suite:
Known issues:
All:
- On occasion files saved from a code cannot be reopened. Workaround:
Add .txt to the filename. This should force the file to be recognized by
all FGSS codes. If you edit a file, be sure it is saved as plain text (not
.rtf, or rich text, or MS Word .doc).
- Preferences are not saved. This was originally by design as use in classrooms
or from CDs was envisioned. All updated versions after 9/11 will save preferences if possible.
- Windows Vista seems to have some issues with some of the codes, but mainly with GravMag. I am hoping this is compiler dependent and newer versions of the compiler will get past these problems. There are fewer problems seen in XP. Windows 7 64-bit OS has produced problems with Resist and seems likely to have problems with older versions of the code. Again, update produced in late 2010 and especially Sept. 2011 should remove this problem.
- Updated Windows software (from April 2010) will not access the Help Guide correctly owing to the extra directory layer added in the new versions of the code (caused by the addition of the .dll files with the software). This is corrected in later updates.
- If some fonts were missing (notably Monaco), some parts of windows would not display. Corrected in Sept. 2011 releases.
- On Mac, some non-English keyboards might not allow numbers to be typed in some edit fields. Switching to an English keyboard allows this.
GravMag (v 1.0; most fixed in 1.0.14):
- Huge error in all versions to at least 1.0.16: if points are listed in counterclockwise order, the calculated magnetic anomaly is incorrect.
- Windows (from CD): When running GravMag from the CD with
the text, problems often emerge in changing sizes in the section window and
entering data in the data window. The program becomes sluggish or stalls. Workaround:
Launch the program through Windows Explorer instead of using the launcher
application (right click on the Geophysics CD icon and select "Explore,"
then open the Windows folder and then double-click on GravMag). OR Copy
the program to the hard drive and run from there.
- Erroneous behavior when changing from meters to kilometers if section width
was unchanged from opening. Workaround: Just change the
width of the section (even a trivial change) before changing the units the
first time.
- Exporting while units are km and not m results in a file saved
as though km were m.Workaround:
Set units to meters before saving files.
- Occasionally GravMag will lock up in an infinite loop because of a bug
in part of the code determining where to place the label in the section view. Workaround: none.
Instructors who have adopted the book may contact me directly for a new version.
Mac users should obtain the updater (above).
- Elevations of magnetics points are negative of where they should be. Workaround:
Assume that the z column in the Magnetic Data window
is positive down.
- Changing the preference to set the magnetic field manually produces a crash
if the Magnetics Data window is not open. Workaround: Open
the data window before changing this preference.
- Changing the preference to show a larger number of significant digits has
no immediate effect. Workaround: set this preference prior
to working with data files (model values will update as soon as the model
is changed or recalculated)
- Clicking in the bottom pane of the Model window when there is no body, and then clicking on "Add points" crashes GravMag. Workaround: add a body before trying to add any points.
- Changing the units while drawing a body in the section window causes unusual problems (usually a distorted shape). Workaround: Do not change the units while drawing or editing a body in the section window.
- Crash if adding first body after opening a Section Window after a data window is open and before defining any bodies in the Model Window. Workaround: Avoid this particular sequence of events (bug fixed 1.0.20).
- Section pane of Section Window is blank if opening Section Window with no bodies and a data window is open with all points at the same elevation. Workaround: Avoid doing this. If this happens, can change the elevation of a data point or close the Section Window, add a body, and reopen Section Window. Fixed v. 1.0.20.
- Crash possible after adding a body in Section Window using the "Done" button with the Magnetics Data Window open. Workaround: End interactive adding of a body by double-clicking or clicking on first point instead of using "Done" button. Fixed 1.0.20
- Windows only: Typing a density with a leading decimal
(e.g, ".7" instead of "0.7") or omitting a decimal (e.g.,
"2" instead of "2.0") can crash the program. Oddly, this
seems to depend on the service pack or installation
of a library. Workaround: Use
a zero before decimals and a decimal point after whole numbers.
- v1.0.18: Opening the Section Window after a data window is open results in the bottom of the section pane (part showing the model) extending off the screen. Workaround: Toggle the "View Gravity" or "View Magnetics" option. Fixed in 1.0.20
Refract: (most fixed in v1.1)
- (bug only in v 1.0.2 - 1.0.5): Data saved with uncertainties will be read in incorrectly: the errors for the forward shot will be swapped with the observations for the reversed shot. Fixed in 1.0.6. Workaround: If you have a saved data file, say Data1, read it in and then save it as Data2. When you read Data2, everything will be correct.
- If a reversed time pick is present but the forward pick is absent, then
changing the time units will not change the value for the reversed time pick. Workaround:
Set the time units before entering all your values (or make sure that you
have picks for the forward shot at each geophone picked for the reverse shot).
- Changing the number of geophones when the data window is open will lead
to a crash. Workaround: Set your number of geophones before
opening the Data window.
- Changing the shot location does not result in updated data points on the
Plot Window. Workaround: Editing anything in the data window
will force an update in the polt window.
- Hiding Model Points in the Plot Window doesn't stick (points show up again
after changing the model. No workaround other than rehiding the model points
when you really need them to be hidden.
- Model files overwrite shot offsets, etc., without user confirmation. Workaround:
When reloading exisiting models and data, load the model first and then the
data.
- Data files are saved in the currently used units, and read in the currently
used units, so data could be misread if different units were chosen between
the save and open commands. Workaround: When opening a data
file, set the time units to match what was used when the data file was saved.
- Cannot click on the angle of the top layer to change it after reading in
a model of more than 2 layers when previously only had one layer. Workaround: Select
another parameter and use tabs or shift-tabs to navigate to the angle.
- Changing the number of the first geophone in the preferences leads to a
crash. No workaround.
- If the Model from Plot operation makes a nonsense structure (e.g., with
layer boundaries crossing one another), there is no warning in the upper
right of the Model Window. Workaround: Aside from recognizing
such a mess in the Section Window, Updating Times and Plots will show the
warning.
- Negative values for shot offset produce aphysical answers for some arrivals. Workaround: Don't do this, or just ignore the obviously aphysical arrivals.
- RMS misfit is not updated when loading a model or data. Workaround: Force
the model times to be recalculated (usually just selecting a model parameter
and tabbing out will enable the button)
- Rare nonresponsive behavior on drawing a section profile. Workaround: None
other than leaving the section window closed. This is exceptionally rare
in Refract (equivalent bug in GravMag was more serious)
- Two layers with the same velocity will crash the program. Workaround: Avoid
entering the same values for 2 layers.
- Lines in the plot window will often be strange when a low-velocity layer is present. (The model points, in constrast, should be fine). Workaround: If you have to have a low-velocity layer, turn off the lines from the View menu (Choose to "Hide Line Fits"). As you cannot invert line fits for a velocity model with a low velocity zone anyways (such layers do not produce a refraction), this isn't much of a hardship.
- (Present through 1.1, fixed after 1.1.2) Data points with no data when saved to a file show a value of "0" when data file is opened. This is most annoying in that the RMS value will be greatly affected. Workaround: manually delete all the zero times after opening the data file.
- (bug introduced v 1.0.7, fixed 1.1.5): Refract crashes if opening a model file with additional geophones while Data Window is open. Workaround: set the maximum number of geophones before opening a Data Window.
- Cannot select a line in the Plot Window if it is vertical or horizontal. Workaround: Don't make lines vertical or horizontal (kind of aphysical anyways). Fixed, 1.1.5.
- Shot offsets and geophone spacing rounded to nearest integer when opening a model file. Workaround: Manually change to correct values after opening the model file. Fixed 1.1.5.
- Data values will often be lost when changing the number of geophones (even if increasing number of geophones). Workaround: Set number of geophones before entering data. Warning and cancel option added, 1.1.6.
- Fractional geophone or shot offsets can be difficult to enter. (Generallly, will get a beep and will have text replaced if trying to start to type with a 0 or a decimal point). Workaround: First type the first non-zero digit, then put the entry cursor before that digit and type in the decimal and any zeroes. Fixed 1.1.6.
- Changing the number of the first geophone in the Preferences Window causes a crash if the Model Window is in front of the Data Window. Workaround: bring the Data Window in front of the Model Window or have the Data Window be closed when changing the first geophone number. Fixed in 1.1.7.
- Changing the array geometry when showing layer thicknesses or depths at both ends will result in a model that differs from one in model window. Workaround: Shift back to layer thicknesses and dips prior to changing the array geometry, or, after changing the array geometry, toggle between the mode showing dips and the one desired. Fixed in 1.1.8.
Resist:
- Bad RMS values, occasional spontaneous changes to numbers when run in Windows 7, 64-bit OS. I have not seen this on other platforms. Hopefully fixed in post 1.0.2 versions as new compiler more aware of 64-bit OS?
- Reading a model file with more layers than the current model results in one layer having the wrong resisitivity. Workaround: re-read the same model file a second time. The correct resistivities will be displayed. (Fixed after 1.0.4)
- Inversion often fails to do anything with some parameters fixed (most frequently, fixing a layer thickness and trying to invert for a resistivity). This has to do with some choices made in constructing the inversion. A somewhat more relaxed inversion setup is tried in versions from 1.0.3 on; this might prove to be too relaxed and occasionally result in RMS increasing, though I haven't managed to make this happen, relaxing the conditions on the inversion does result in the inversion calculating RMS during iteration incorrectly. Workaround: If trying to invert with some parameters fixed and the inversion seems not to do anything even with a poor fit, try manually altering the values or swapping fixing thickness for fixing resistivity.
- Inversion occasionally produces a model with junk values in Model Window that don't reflect what is shown in the Plot Window. Workaround: recognizing these values, can recalculate from model to see what these would really yield. Fixed after v 1.0.3.
- Importing a data file with a different starting spacing after calculating a model will result in the Calculated Values in the Model Window and the calculated curve in the Plot Window showing incorrect values. Workaround: Press "calculate from model" button after importing the data. (from 1.0.6 on, model is recalculated if importing data changes the starting separation).
- Changing the starting electrode spacing does not keep the data at their original spacings, but instead keeps these values at the same slot relative to the first sounding (this is by design; it is unclear how to reassign values unless the new starting spacing was consistent with the old spacings). Workaround: set the starting spacing before doing anything else.
- Opening a data file with more soundings than currently shown results in the new soundings being filled with junk. Workaround: Import the data a second time. Once the extra soundings are in place, the data values in the file will be correctly inserted. Fixed v 1.0.6.
- Opening a model file which has more layers than the current model while a layer is selected in the Model Window results in incorrect values for layer thicknesess and resistivities. Workaround: Make sure nothing is selected in the Model Window when opening a model file. If this happens, simply deselect anything in the model window (or select the bottom layer) and reopen the file. Fixed in v 1.0.7.
- Windows 7: Have seen Resist fail to update the Plot Window despite repeated changes in the Model WIndow. Have not been able to track this down to a source. Workaround: Probably quit and restart. If you see this and can make it happen repeatedly, please get in touch so I can kill this bug.
Reflect:
- Opening a second model file when the data window open and reflections checked in the Model Window will crash the program. Workaround: Save data before opening a new model file and close the data window. (fixed after 1.0.1)
- Opening a model file from Refract will reveal any dipping interfaces in the Section Window below the first layer. Workaround: Save model from Reflect and then open it from the Open... menu. (fixed after 1.0.1)
- Two-way travel times listed in info pane at bottom of fit window have wrong units (fixed in 1.0.3)
- Opening a model would reset units to defaults without updating menus or display Workaround: change units from Units menu (might have to first select a checked option to reset) (fixed 1.0.4)
- Data saved in last used units (so will not open properly with different units) Workaround: In pre 1.0.4 versions, make sure that the units have been set to the settings used when saving the model or data file. In 1.0.4 and later versions, when reading a pre-1.0.4 data or model file, the old file will be read assuming it was saved in meters, m/s, and ms (fixed 1.0.4 so correct numbers always read in)
- If using km/s, new velocities were made into Mm/s instead of km/s Workaround: best to use m/s units. (fixed 1.0.4)
- "Add layer" would produce new layer with velocity shown in m/s regardless of units chosen Workaround: If using km/s for units, change value to desired value (fixed 1.0.4)
- Header on data window did not change if using seconds as time unit. Workaround: ignore units in label (units on RMS correct) (fixed 1.0.4)
- Shift-tabbing in ModelWindow's model listbox shifts everything off to the left Workaround: don't use shift-tab; if in this situation, using tab will get to all the thickness and velocity fields; can expand window width to see everything (fixed 1.0.4)
- New layers were always 10 m thick even though appeared 10 km it km were display units Workaround: change the thickness to some value other than 0.01 if using km for units. (fixed 1.0.4)
- If only a halfspace and First refraction checked, "show all arrivals" goes out to infinity Workaround: manually set the upper time limit to something desired, or uncheck First refraction box in Model Window(fixed 1.0.4)
- Time axis labelled distance.Workaround: ignore. This would go away once time units were changed (which sometimes was also an issue) (fixed 1.0.4)
- Changing time to seconds doesn't work Workaround: have to select ms from Units menu even if checked and then select s) (fixed 1.0.4)
- v1.0.3 only-program crashes if distance unit was not default when last used. Workaround: ideally almost nobody has 1.0.3, but if you do, then make sure units are m when you close. If you already have closed with km as unit, then delete the ReflectPrefs file. (fixed 1.0.4)
- Adding layer with the Data Window open results in crash when changing distance units before updating times and plots. Workaround: Only change units after updating times and plots. (fixed 1.0.4).
- RMS layer thickness in Fit Statistics window is 2x too large. Workaround: Divide by 2. This does not affect the model that would be derived frmo the fit. Fixed v 1.0.5
- Model generated from fits in Fit Window have thicknesses and velocities 1000x too small if km is the current length unit. Workaround: Make sure that meters is the current length unit before creating a model from fits. Fixed v. 1.0.5.
Diffract:
- Note that the calculation of diffraction hyperbolae is approximate for multiple layers and when the source-receiver distance is near the depth of the diffractor.
- Depth labels on section window occasionally vanish. Workaround: adding a layer boundary with a non-integer depth will usually get rid of this. Fixed from 1.0.3 on.
- Plot window messed up if using cm length scale. Workaround: use meter scale. Fixed from v. 1.0.3 on.
- Extra diffractors appear when opening a model file with fewer diffractors than were already in the Model Window. Workaround: Do not edit or delete the extra diffractors. Probably best to quit the program and restart and load the model file. Can delete diffractors before opening a model file. Fixed in v 1.0.4
- Opening a data file in different units from when it was saved results in the wrong values Workaround: Change the units to what they were when the file was saved and then reopen it. After 1.0.4, files should be opened correctly regardless of units chosen; pre 1.0.4 files are assumed to have had time units of ns.
- Cannot delete a single diffractor if only one diffractor when changing to the diffractor pane. Workaround: Add a second diffractor and then, without changing the focus from the diffractor pane, delete one and then the other diffractor. Fixed 1.0.4.
- Creating a new diffractor when units used are cm results in the diffractor being placed 100x deeper than the display initially shows. Workarounds: can use m scale, can toggle from cm to m and back to cm to see true depth of new diffractor, can simply edit diffractor depth to desired value. Fixed in 1.0.4.
- Data file that has uncertainties loses the uncertainties if opened when preferences are set to not use uncertainties. Workaround: be sure that your preferences are set to use uncertainties before opening such a data file. In 1.0.4 and later, user is asked if he or she wants the preferences changed when opening such a data file when uncertainties are not allowed.
The geophysical software suite is cross platform, behaving the same on Windows,
Macintosh, and Linux OSes.
Please send mail to
cjones@cires.colorado.edu if
you encounter any problems or have suggestions.
C.
H. Jones | CIRES | Dept.
of Geological Sciences | Univ.
of Colorado at Boulder
Last modified at
January 10, 2012 7:29 PM