Software FAQ

If you currently subscribe to the Waterloo Hydrogeologic software maintenance plan, you are eligible for a free upgrade. If you do not subscribe to the Waterloo Hydrogeologic software maintenance plan, please contact our sales team sales@waterloohydrogeologic.com to obtain an upgrade quote.

• What is MODFLOW?

Yes, with Visual MODFLOW Flex 2013.1 and later you can run contaminant transport models with MT3DMS. Click here to learn more.

Similar to Recharge, the Visual MODFLOW Flex interface provides an option to apply “Evapotranspiration” to different layers during the simulation

• To the top layer
• To the top-most active layer

When defining a lake boundary condition, there is now an option to define horizontal leakance explicitly; this is important when there is both horizontal and vertical flow between the lake and aquifer, which is common in quarries or mine tailing ponds that have steep walls.

In the Output, there is a new Lake Budget option for Graphs that allow for showing time series plots of the lake parameters: stage, precipitation, evaporation, runoff, volume changes. You can chart results from a single model or load multiple model results on the same plot, for comparing different model scenarios.

MODFLOW-USG is a new version of the MODFLOW groundwater flow simulation code developed by the United States Geological Survey (USGS) which uses unstructured grids instead of rectilinear grids to discretize the model domain.

An unstructured grid is made up of triangular, rectangular or other polygon shaped geometries individually or in combination to discretize the groundwater model domain. This is in contrast to previous versions of MODFLOW, including MODFLOW-2005 which uses structured, rectilinear grids. Unstructured grids require the use of the finite volume method whereas traditional structured grids are used with the finite difference method.

Structured finite difference MODFLOW-2005 grid (left) and an UnStructured grid with Voronoi Grid (middle) and Quadtree Grid (right) for MODFLOW-USG

Although MODFLOW-USG can be used with any grid geometry – triangles, rectangles, polygons – Visual MODFLOW Flex supports unstructured grids made up of polygonal Voronoi Cells or rectangular Quadtree cells.

A common problem when refining finite difference grids is high aspect ratio cells. Cells have a high aspect ratio when the ratio between their height and width is far away from 1. By default, the “Edit grid” window will highlight cells that share a face with an aspect ratio greater than 1.50 pink:

These cells can – but not always! – cause stability and convergence problems when MODFLOW runs. Because refinements to finite difference grids are carried out all the way to the model edges, it’s common to have bands of high aspect ratio cells extending orthogonally from your refined areas of interest.

How to deal with high aspect ratio cells:

Option 1: Is it a problem?

As mentioned above, having high aspect ratio cells is not always a problem, particularly when hydraulic gradients are not excessively steep in the area. It’s worth running the model to see if the model converges and produces sensible results. If so, you might simply ignore them.

Option 2: Can you refine the entire model?

Partial refinement of the model is a method to increase resolution in areas of interest while minimizing the resultant increase in computation times. However, for many models the computation times aren’t that long to begin with. If feasible, you should consider uniformly increasing the grid resolution.

Option 3: Unstructured grids, MODFLOW-LGR

For some highly refined model domains (e.g. regional-scale models with site-scale areas of interest), the above options won’t work. Thankfully, Flex includes plenty of tools to efficiently discretize these kinds of model domains.

One option is unstructured grids, available with MODFLOW 6 and MODFLOW-USG. These grids don’t discretize space in the same way as finite difference grids and can lead to much more efficient local refinement for some models. See the primer on our website for more information.

Another option is local grid refinement, or MODFLOW-LGR. Using MODFLOW-LGR, you can embed MODFLOW models (called “child grids”) inside of a larger one (the “parent grid”). See the Flex documentation for more information and resources on MODFLOW-LGR.

Option 4: Graduated refinement

If none of the above options work for you, the final option is to refine your model on a gradient, so that your chosen maximum aspect ratio is never violated. An example for the above grid is shown below, using a maximum aspect ratio of 1.5:

To do this method, simply refine the outer highlighted edges by the largest ratio you can to satisfy the criteria, and work outwards until the highlighted cells disappear. In the above example, the interior cells were originally refined by three times. The rows and columns next to them were refined by two, and then the rows and columns next to those were refined by 3/2 (exactly satisfying the 1.5 max ratio threshold).

As a consequence, however, this grid refinement is not very efficient. After increasing the resolution of a 2-by-2 cell area by 3 and conducting graduated refinement, our model area has grown from being 20 cells by 20 cells to 28 by 28, an increase of nearly double.

Voronoi Cells

One of the types of unstructured grid supported in Visual MODFLOW Flex. Voronoi Cells are a special type of polygon that honors the Control Volume Finite Difference (CVFD) constraints required by MODFLOW-USG. Using this grid geometry with MODFLOW-USG results in an efficient and robust numerical grid that can handle complex hydrogeological geometries and boundaries with ease – similar to triangles, but with a higher degree of accuracy.
VMODFlex_ImportingVmodProjects_Tutorial

Learn more about Voronoi Cells.

Quadtree Grids

Quadtree Grids are one of the types of unstructured grid supported in Visual MODFLOW Flex. Quadtree grids are based on locally refining rectilinear grids by recursively subdividing cells into four equal quadrants.  This allows you to locally refine the grid around non-linear features such as streams and lakes without having to carry those refinements laterally across the entire model domain as occurs when using fully structured rectilinear grids.

Using this grid geometry with MODFLOW-USG results in an efficient numerical grid that can handle complex hydrogeological geometries and boundaries with ease. However, the use of Quadtrees can introduce some errors as the geometry of Quadtree grids does not adhere to the assumption inherent in the Control Volume Finite Difference approach utilized in MODFLOW-USG that shared cell interfaces occur at the perpendicular bisector of the respective cell centroids (see discussion below on Ghost Node Correction [GNC]).

Ghost Node Correction is a technique that is applied to some unstructured grid types (e.g., Nested or Quadtree Grids), whose geometries do not inherently satisfy the requirements of MODFLOW-USG.   One of the geometric constraints for the Control Volume Finite Difference approach used in MODFLOW-USG is that the line formed by connecting the centroids of two adjacent Finite Volume cells must perpendicularly bisect the shared cell interface. While Voronoi Tessellation by definition fits this criteria, other grid types (such as Delaunay Triangulations and nested or Quadtree Grids) do not strictly conform to this constraint.

In order to permit the mathematical solution of the Finite Volume Method certain grids, the USG package employs a technique called Ghost Node Correction (GNC) in which a temporary ‘ghost’ node is created within a cell such that it meets the geometric constraints and the equations may be solved. This solution at the ghost node is then interpolated (corrected) back to the actual cell center point, resulting in a Ghost Node Correction within each cell. (http://pubs.usgs.gov/tm/06/a45/pdf/tm6-A45.pdf)

The ghost node correction package is generally not required when using Voronoi Grids because by definition they satisfy the geometric conditions required for the Control Volume Finite Difference approach used in MODFLOW-USG.

The use of Quadtrees may introduce errors in the simulation results as the geometry of Quadtree grids does not adhere to the assumption inherent in the Control Volume Finite Difference approach utilized in MODFLOW-USG: that shared cell interfaces occur at the perpendicular bisector of the respective cell centroids.  The interfaces between neighboring cells with different degrees of refinement will occur at an angle that departs from perpendicularity and the distance between the centroids will become increasingly asymmetric as the relative degree of refinement between the cells increases. In cases where the model is smooth (that is the model properties, cell elevations/thicknesses, aspect ratio between adjacent cells is small [2:1]) and/or the non-conforming cells are located far from the area of interest, the resulting error may fall within tolerances that are acceptable to the modeler.  For cases where this error may be unacceptable, Visual MODFLOW Flex supports the MODFLOW-USG Ghost Node Correction (GNC) package which corrects for this potential source of error. Note that the use of the GNC package is likely to increase model run times.

Increased Simulation Accuracy

Unstructured grids permit local grid refinement along model features such as wells and boundary conditions, in addition to supporting fully discontinuous layers, both of which deliver high resolution detail and accuracy around areas that matter the most.

Reduced Simulation Runtimes

Traditional MODFLOW grid refinement often results in an excessive number of grid cells outside the areas of interest which leads to an inefficient numerical model with unnecessary long run times. With unstructured grids, the grid refinement is focused only to the areas of interest, resulting in much fewer grid cells, more stable grid structures, and therefore reduced run times.

Further Reading:
Case Study: MODFLOW-USG Yields 90% Reduction in Run Times for Regional Groundwater Model

Superior Modeling of Complex Geology

Unstructured grids are flexible and conform easily to the natural variability found in surface features and subsurface stratigraphy. Unstructured grid layers permit discontinuous layers that pinch-out to zero thickness, providing a more realistic representation of complex hydrogeological conditions, and improved model stability.

Increased Numerical Stability

Unstructured Voronoi grids in Visual MODFLOW Flex are composed of a special type of polygon called Voronoi Cells that are generally superior to other unstructured grid geometries (nested grids, Quadtree grids) as they completely honor the Control Volume Finite Difference (CVFD) constraints, and therefore do not require the Ghost Node Correction (GNC) package. In addition, cell centroids will coincide exactly with conceptual wells and boundary condition feature vertices. This results in an efficient, robust and dependable numerical grid, with improved convergence and accuracy.

MODFLOW-USG can be used to simulate groundwater flow for virtually any application. However, It’s ability to handle complex geology makes it ideal for applications in mining where steep slope geology is often encountered. Flexible grid refinement capabilities also make it suitable for large regional and watershed scale models.

The following processes and packages are supported in MODFLOW-USG:

• BAS, DIS, BCF, LPF, HFB
• CHD, FHB
• WEL, DRN, GHB, RIV
• STR, SFR, LAK, GAG
• SUB
• New Packages: DISU, CLN, GNC, SMS

MODFLOW-USG & Visual MODFLOW Flex

MODFLOW-USG is only available in the Professional and Premium editions of Visual MODFLOW Flex.

This usually means your model may have non-convergence or stability issues. Please refer to our Blog Post about common troubleshooting steps to resolve these issues.

Yes. If you currently use Visual MODFLOW Flex, you will already have a 3D conceptual model that defines the geologic layers, properties and boundary conditions independent of the grid. To run your model with MODFLOW-USG, simply auto-generate the unstructured grid and then run the simulation. Visual MODFLOW Flex’s multi-model environment allows you to manage both your MODFLOW-2005 and MODFLOW-USG models within the same project, and both models can be easily compared side-by-side.

If your model currently exists in Visual MODFLOW Classic, you will need to import your model into Visual MODFLOW Flex before you can run the model using MODFLOW-USG.

Yes, you can leverage the power of 64-bit versions of the MODFLOW engines during running of your models to speed up your run time, if you are running 64-bit version of Windows. This support is available for all supported Engines in the Professional and Premium versions of Visual MODFLOW Flex.

Parallel Processing

If your computer contains multiple processors, or dual core processors, you can leverage the power of parallel processing for your flow simulations using Visual MODFLOW Flex simulations using PCG, WHS, or SAMG solvers can be solved over multiple processors reducing the simulation run time. Note that the parallel/multi-core version of the SAMG solver is only available in the Premium Edition of Visual MODFLOW Flex.

SAMG

The latest version of Visual MODFLOW Flex supports SAMG v4. This version of the SAMG solver is dramatically faster than its predecessors, and is ideally suited for multi-layered models with heterogeneous properties. The single core-version is available with Professional Edition of Visual MODFLOW Flex, while multi-core version is available with the Premium Edition of Visual MODFLOW Flex.

Yes, the Visual MODFLOW Flex interface supports MODFLOW-NWT package, a version of MODFLOW which provides enhanced capabilities for solving problems involving drying and rewetting non-linearities of the unconfined groundwater flow equation.

Hydro GeoAnalyst can be used to build the conceptual model layers for a Visual MODFLOW Flex model. You may also use the data in Hydro GeoAnalyst to import observation wells into Visual MODFLOW’s Classic Interface.

Creating Conceptual Model Layers:

In Hydro GeoAnalyst, you can query the database to determine layers with different lithology and export this data for use in Visual MODFLOW Flex.

For example, the HGA demonstration database can be queried to determine the bedrock elevation. To query the bedrock surface, we will need to do the following calculation:

• Ground Surface (station.elevation) – Top of Bedrock (depth to shale or limestone). In the Query Builder this would be represented by:
• Select Max Function and expression, station.name= lithology.from_ with the Alias name Top of Bedrock
• A condition may then be applied for soil_type = limestone
• You may need to adjust the query according to your project fields/data. You may then execute the query. Once the database has been queried to determine the bedrock elevation, a gridded surface can be created and used in Visual MODFLOW. In HGA, load the query into the Map Manager and create a gridded surface of the data (.grd file). For additional information on creating gridded data please view the online help documentation.
• Export Model Layers. You can choose which layers to export and from which cross sections. The more cross sections you define, the more points you have for defining the layer interfaces. The model layers will be exported as a text file with x, y and z values. For additional details on creating your geological layer interpretations, please view the online help documentation.

In Hydro GeoAnalyst, you can create a calibration point file by creating a query with the following columns, for example, to import observation wells:

Once you have created the appropriate query, you may export the data to an .xls or .txt file to import into the Visual MODFLOW Classic Interface.

In Visual MODFLOW, you will need to set up the model with the appropriate model or world coordinates so that the files imported from HGA will correspond with these coordinates. Create the model and import the model layers and calibration points from HGA into Visual MODFLOW.

Delete the listed parameters and then re-enter/re-select the parameters. You will notice the column contents now match the headers and the report should be able to run.

The PHREEQC database files distributed with the program are installed under My Documents AquaChem PHREEQC.

You can find the most up to date versions of the PHREEQC Graphical User Interface from the USGS website by selecting File>Preferences> PHREEQC, and clicking on the PHREEQC Website button.

MS Access 2007 is a prerequisite for the installation of AquaChem (version 2011.1 and higher).
If this dependency is missing, you will receive the template message and a “[Warning!] ActiveX component can’t create object” error will occur. To correct this issue you will need to install the MS Access 2007 Runtime Engine directly from Microsoft Website: http://www.microsoft.com/download/en/details.aspx?id=4438

The issue can be due to a missing system table in the database if the project was created using the advanced template. This can be fixed by following these steps:

• Create a new project in AquaChem.
• Open the project in MS Access.
• Copy (ctrl+C) the table M_LinkedObjects from the list of tables.
• Open the client project in MS Access.
• Paste (ctrl+V) the table into the list of tables.
• Save the changes in MS Access and open the project in AquaChem.

To create a new water quality standard, follow the steps below:

• Open the Database options dialogue (File > Preferences), and select the Standards-Import tab.
• Click the [+] Add button to create a new standard.
• A blank Water Quality Standards dialog will load. Here you can: enter the name for this new set of standards, enter a reference for the new set of standards, enter a URL for the set of standards, enter the year that the standards were approved, and enter guideline levels for the new set of standards.
• There are two ways to enter parameters and assign guideline levels: the values can be entered manually or the data can be imported from a file. Either way, data is entered for one guideline at a time.
• Once finished, click the [Close] button.

To display the data on these plots, values must be present for all of the ions that you are trying to show on the plot. If even one of the ions does not have any values then your information will not be plotted. Right click on you parameter from the Sample Details screen, and double check if all parameters are available.

Please ensure that the plot symbols have been assigned correctly. This means that a separate symbol will represent each station in the project.

• You can confirm how the symbols are assigned by selecting Plots>Define Symbol or Line from the top toolbar. Ensure that Station is selected as the symbol grouping.
• If this is not available, you can add a new grouping for station id by selecting the […] button and adding a new grouping. Select this new grouping as the Symbol Group and then select Auto to automatically assign a symbol based on the station id.
• Alternatively, you can add only the stations you would like to view on a plot to the symbol grouping list.
• Once the symbol grouping has been selected and applied to the project, recreate your plot and select the stations you wish to view on the plot.

New feature in Pro: Use derivative “smoothing” to reduce noise in the dataset. AquiferTest provides three methods for derivative smoothing: Bourdet Derviate (BOURDET 1989), Standard (HORNE 1995) and Regressive (SPANE & WURSTNER 1993). For more information on these methods, please refer to the original texts.

• To open the Derivative Setting dialogue,go to the Analysis tab, and select Analysis> Derivative… , from the main menu. You can apply derivative smoothing to all datasets in the analysis by selecting the Use sample setting for all data option.
• To assign different methods to different data sets, select the ‘Set’ each data set separately option.

New feature: Click on the curve, generated from your data, and drag it to your desired best fit location using your mouse. This method provides more flexibility and improved accuracy.

New feature in Pro: The Clonts & Ramey solution is suited for determining hydraulic properties, i.e., transmissivity, storativity and conductivity anisotropy ratio, from horizontal well pumping test data from confined aquifers.

This message indicates that given the data set, the automatic fit was unable to match to the data points. If the automatic fit did not succeed you may consider adjusting the starting parameters by using the Parameter Controls.

• Select the parameter control icon above the analysis graph and use the slider bars to modify the parameter values and to provide a better starting point for the automatic fit.
• Once these values have been adjusted you can perform the automatic fit.

Additionally, if you continue to experience difficulties with the automatic fit, you may consider increasing the number of iterations used for the fit by selecting Tools>Options and browsing to the Constants tab. The automatic fit options are available under this tab.

Once in reports, please go to >Maine menu, >Tools from the top menu bar and choose “Options”. In the Appearance tab, you can modify the colors and symbols used.

Please note that currently only projects created in AquiferTest 4.x onward can be opened in the current version of AquiferTest . To open 3.x AquiferTest projects in the new version please follow these steps:

• Using version 4.x of AquiferTest, open your project and save. Then, using the current AquiferTest version, open your desired project, creating a back up if prompted.
• It is recommended that you ALWAYS create a backup copy of any project files before you open them in the new version. Specifically, ensure that you back up your original MS Access database (.MDB), which contains all project data.

Your printing issue could be the result of a number of settings within AquiferTest Pro that have not been selected. Please consider the following:

Printer Setup

• Click File and select Printer Setup from the main menu
• Select the appropriate printer from the drop down menu since the default may not always be correct.
• Click OK and attempt to print the reports again

*This option will also allow you to select PDFCreator as your printer which will export your reports as PDF documents. You could then open each .pdf file and print them from there.

Printout Selection

• From the Reports tab ensure there is a check mark in the boxes beside each printout you wish to be generated. AquiferTest will allow you to view your reports by highlighting each name, however they will only print when the check mark is showing.

Printer Drivers

• Be sure to update the drivers on your printer, this can be done by selecting your desired printer from the Control Panel on your computer.