Jump to content
GMS, SMS, and WMS User Forum


  • Content Count

  • Joined

  • Last visited

Everything posted by mmamobile

  1. Please consider adding support for drawdown observation data - particularly as it would pertain to parameter estimation simulations. For MODFLOW-2000/2005/NWT where the observation process is being used, this change would be relatively easy to implement via a change to the ITT flag value. For MODFLOW-USG, the change may be even easier to implement through an additional call to usgmod2obs that extracts relevant simulated drawdown values. It would also be useful to provide an option to pause PEST execution following creation of the input files - the control file, in particular. -Mike
  2. Hi, I'm experimenting with MODFLOW-USG and local quadtree grid refinement (versus traditional refinement and MODFLOW-2000/2005). My USG simulation converges; however, when I attempt to import the results, GMS crashes. I've traced it back to having parameters specified using key values - when I directly specify the parameters, the results import correctly. Is there a workaround / fix in the works? PEST is a must for my project work. I'm fine with going back to MODFLOW-2000/2005, if necessary - but I was looking forward to using USG for two upcoming models. -Michael
  3. mmamobile

    Zone Budget

    Select your MODFLOW solution and use the Plot Wizard -> Flow budget vs. time option pick your zone(s) and export from there.
  4. The conceptual model pumping coverage has an attribute table constructed as you describe (step format) - the problem is, while the time intervals match the stress periods in the transient MODFLOW model, the pumping data are not converted correctly. For example, if I have a rate specified as 5.0 cfd at time = 10 days, after converting to MODFLOW, I would expect to see a rate of 5.0 cfd corresponding to the transient stress period spanning day 10 to day 11. Instead, I see a rate that has been adjusted by some form of interpolation.
  5. Hi, Just a suggestion regarding MT3D's transient observation (TOB) package: Implementing the additional options available within the latest version of this package would be very useful (such as the ability to turn on adjacent cell interpolation and weighted multi-layer concentration weighting). Most of the options are simple flag values or line additions to the TOB input file, so I've been doing this manually, but adding these options to the GMS GUI would be nice. Also: adding the ability for MNW recognition within the Source/Sink Mixing (SSM) package would be very helpful - I've been doing this manually by creating a dummy file using the WEL package option, then replacing the flag value (2) with the MNW flag (27). Much appreciated, Mike
  6. I'm having some difficulty with the conceptual model translation when using the WEL or MNW packages. I have a transient pumping rate dataset, and my model's stress period setup matches the intervals exactly (e.g., rate specifications at the start of each stress period). After converting the data from the conceptual model to MODFLOW, I notice the rates have been altered with respect to the inputs to the coverage attribute table. That is, interpolation is being performed (even though I'm not getting a message). Am I missing something in terms of how GMS interprets transient data in the conceptual model versus a transient MODFLOW stress period setup? Again, this occurs when using either the WEL or MNW packages to represent transient pumping rates. Any help would be greatly appreciated, Mike
  7. Hi, I'm working with a single layer transient MODFLOW model, and I'm trying to convert the heads to boundary conditions for a local-scale sub-model. When I attempt to convert the MODFLOW-generated heads to 2D scatter points (using a coverage defining the local-model region), GMS creates an empty folder containing no head data. I've gone through the tutorial on transient conversion and that seems to work fine (with the example file), so I'm a bit confused as to why I'm running into this issue with my model. Working with GMS V8.1.4 64-bit Any help would be greatly appreciated, -Mike
  8. I actually found the issue was with NAPL source cells being co-located with MODFLOW constant head cells. Once I limited my cell selection to CHD-free cells, everything worked correctly.
  9. Hi, I'm trying to do some simple hypothetical modeling using the SEAM3D NAPL dissolution package, and I'm encountering a memory issue when initializing NAPL source cells through the Point Source/Sink option. After highlighting the cells and entering the appropriate source information, I receive the following message: "This machine does not have enough memory to continue this operation. Save your data and exit. (mtCellPointSourcSinkDialog)" I'm using the 64-bit version of GMS V8.1 on a machine with 4GB of memory installed, and I seem to get this error regardless of the number of cells selected. With that in mind, I have to doubt it's a hardware issue. Any thoughts? Much appreciated.
  10. Hi, I'm trying to do two things: 1. calculate mass flux (or mass discharge) across a cross-sectional area using a transient dataset 2. export a subset of these data (i.e., access the transient data for a single column) The mass flux/discharge calculations are relatively straight-forward (I think) - I've converted the MODFLOW cell by cell flows to individual face flows then multiplied the transient concentrations by these values, which gives me mass discharge in mass/time. However, this dataset is transient and three-dimensional (and relatively large), which limits my options as far as exporting the data. I've tried to convert to a 2D grid, but that will not work because it would need to be vertically-oriented, and GMS does not provide this option. Any suggestions would be greatly appreciated.
  11. Alan, Yes, I'm still having the issue with the TOB package - I'm thinking the problem boils down to an imperfect match between the model-calculated time step (model time) and the observation intervals. The TOB package (OCN output file) provides model-calculated concentrations at each location at the end of each time step, but it does not interpolate the observed concentrations or take the observation in closest temporal proximity, which results in the OCN file having no "observed" data. Does that make sense? I don't mind working outside of GMS to compare model predicted and observed concentrations, but it would be nice to use the TOB package to do the work. Thanks again.
  12. Apparently, you have to do a full installation - there is no incremental patch available to update from 7.1.5 to 7.1.6.
  13. Thanks Alan - would replacing the .exe or redirecting to the 5.3 version resolve this issue? I imagine I'd have some manual manipulation of the SSM package to do... Also, any thoughts on the lack of observed concentration data in the .OCN file?
  14. It sounds like the conceptual model hasn't been initialized with the transport option set to active. If you right click on the conceptual model and go to properties, the "Transport" toggle should be checked - also, "Leachate" should be defined under the "Species" button.
  15. Following up on this issue: I've been able to get MT3DMS to run with the MNW package, but there seems to be no way to represent an MNW well in the SSM package.
  16. I've generated a flow model using the MNW package to represent several pumping wells - the flow model seems to be producing reasonable results, but I'm not able to run MT3DMS. The error message generated in the MT3DMS window is "ERROR: MAXIMUM NUMBER OF SINKS/SOURCES ALLOWED [MXSS] = 1674 INCREASE VALUE OF [MXSS] IN [sSM] PACKAGE INPUT FILE". When I convert to the WEL package, I have no problems. Am I missing something? I know MT3DMS version 5 was updated to provide compatibility with the MNW package, and from the look of it, GMS is using version 5.2. I've also tried to manually alter the SSM input file, but MT3DMS fails if I do this. If I can't use the MNW package directly with MT3DMS, is there a simple operation to convert the WEL package generated by the MNW package to something that can be recognized by GMS? More specifically, the WEL package is in standard MODFLOW format, which differs from the WEL package format used by GMS-based MODFLOW simulations. Also, I'm attempting to use the transient observation (TOB) package in MT3DMS to compare model results to tracer breakthrough at several observation wells. I've gotten to the point where the .OCN file is providing the simulated concentration at each time step (at each observation location), but the file is not showing the observed concentration. Is this due to the fact that my observation times do not exactly match the time steps produced by the solver? Any help on any of these issues would be greatly appreciated! Thanks!
  17. If the model(s) is/are transient, make sure you've selected output for each time step under "output control".
  18. Are there any plans to extend PEST to the transport modeling side of GMS (e.g., optimization of MT3D/SEAM3D transport parameters)? If not, are there any suggestions as to using PEST in its current form within GMS to set up PEST input files that include transport parameters (such as NAPL dissolution rate, etc.)? Thank you - any information is greatly appreciated.
  19. I get *similar* errors moving back and forth between 7.0.4 and 7.1.1 - sometimes the GMS project will not open at all, and sometimes I get a MODFLOW h5 error.
  20. The solids were the problem - thank you.
  21. Hi, I'm trying to export materials from a relatively large MODFLOW/MT3D model to an ArcHydro GW Geodatabase, but the export routine seems to continuously hang at the following step: "Writing out GeoVolume feature class". Is my model too large for this translation? Thanks.
  22. Hi, I've completed a PEST simulation using the pilot points approach to adjust K values within a zone in my model. I would like to export the resulting conductivity field, but the grid values are only held in array form as the key values - that is, the negative key values used to translate the zones to PEST are the only information available. Is there any way to export these data? I'd even be satisfied with converting to a 2D dataset that could be exported and reimported to a different model (tried this....didn't work). Thanks!
  23. I'm having trouble with Zonebudget as well - when I convert zones created using the conceptual model approach to MODFLOW, the entire Zonebudget zone matrix is removed from the grid menu entirely. I've tried to manually copy + paste, but the matrix seems to reset itself to a uniform value of 1 even after saving. This is version 7.0.2.
  • Create New...