Jump to content
GMS, SMS, and WMS User Forum

Sean Czarniecki

GMS Experts
  • Posts

    264
  • Joined

  • Last visited

  • Days Won

    2

Everything posted by Sean Czarniecki

  1. My understanding is that Top is the top of the stream bed and Bottom is the bottom of the stream bed. So typically, you know the Top of the stream bed. Then, you will often estimate the thickness of the stream bed (unless you have field measurements) and can subtract that thickness from the Top to put in the bottom value.
  2. I usually use the STR2 package, so I had to look up what may be happening to you. When I looked up the STR workflow in the GMS wiki page, it showed me that you may have your arc direction going the wrong way (who knew?). Under display options, you can turn on arc direction arrows (at least in the latest build of GMS). To change the direction, select the arc with the problem and right mouse click on it. There will be an option to change the arc direction.
  3. Ooh....I *might* have come up with the answer. I recall something like this happening in the past for a transient run. I think the first stress period was okay, but then the rest weren't. You know how we use the box to check off using the previous stress period? I think that you might look at the button for "Edit All Use Previous" in the ET dialog box. If you see that they aren't all "1", then you may have some stress periods where you didn't enter data and it sees no ET. Look into that. Curious - what version are you using? I think I brought this issue up in the past, but I'm not sure it is fully corrected.
  4. My point on the New Conceptual Model is that you were asking why you couldn't add a Coverage. You can't make a coverage without first having a conceptual model (it doesn't have to have anything in it except your new coverage. I will have to look into it, but I'm wondering if you would be able to visually show the drain symbols and then export them as a shape file, pull them in as points, and then convert them to drains in a coverage....just brainstorming here. You would probably lose the attributes of the drain though. You might be able to export/copy them to Excel at some point, but tying them together could be tough. Every model is different in getting through things like this. Regarding evaluating individual drains, set up a plot that shows the flows over time for whatever cell you pick and you can right mouse click on the plot and export the flows for that drain. I'm not sure if that's what you are looking for or not.
  5. If you have a well activated that doesn't have any flow, you can just leave it. This is a warning, not an error that will be a problem. If you aren't supposed to have any wells without flow, then add a flow to them.
  6. Does your ET depth actually reach the groundwater? That's all I can come up with quickly.
  7. Based on the information you have provided, it sounds like you may have a confining layer where your lower aquifer has a higher head than the upper aquifer (upward gradient). If you were measuring this with wells out in the field, you would see the shallow overburden wells with a lower water elevation than the deeper overburden (or bedrock). In this scenario, the actual water table (which you would look at for nearby effects) is typically the water elevation of the shallow overburden, unless something allows water to get past the confining layer (like a well that is screened across all intervals.....or digging through the confining layer).
  8. I've never assigned stage using a TIN, so I'm just guessing here. It sounds like your polygon manages the bed elevation and conductance, so that is working properly. However, when you use the TIN to assign transient stage data, does it only assign stage data where there are nodes? If a node doesn't land in the partially covered cell, then stage data would remain zero. Again, I'm just brainstorming.....
  9. First - you won't see existing drains in the Project Manager if you are pulling in an existing model. You will see them graphically, but can only change them using the menu system. Second - You have to make a new Conceptual Model and then add a new Conceptual Coverage to that Model.
  10. My first thought is that you are running this as a steady-state model. That would set up a situation where your starting heads and where the solution wants to be is too far apart. The solution immediately oscillates significantly and can't converge. I suggest that you make it transient. That should help.
  11. Check to make sure you don't have 2 river nodes in the same cell. That can sometimes happen if drawing rivers with arcs and there is an overlap of lines. If that isn't the issue and you don't mind me looking at your model, I'd be up for it.
  12. I have Build 10.4.10 installed and I see the Horizons -> 3D Mesh option come up when I look for it. My only guess is that your license doesn't include the 3D Mesh module. The pop-up box you show on your screen shot actually covers up the view of the available module icons, so I can't answer that by looking at your screen. However, the location where the 2D Mesh module would normally be (between the solids module and the 2D grid module) does not show the icon for the 2D Mesh module, so I'm guessing that you don't have either of the 2D Mesh or 3D Mesh modules. You can figure this out by looking at your license registration and the modules included.
  13. If I have this right, multipliers for Recharge in MODFLOW are not applied on a cell-specific basis, but rather by stress period across the entire grid. Therefore, if you want different multipliers for each stress period, you can do this in a spreadsheet (one column), copy the cells, and drop them into the GMS recharge input box. If you need to multiply different areas of your grid, you will have to do that manually, but creating data sets (one for each recharge and one for the multipliers) and multiplying the data sets before pulling them into the recharge rate input. Right now, that's the only way I can think of, but maybe others have different ideas.
  14. If your river stage is below the river bottom, then you definitely want to switch to one of the stream packages. The stream packages allow the river/stream to go up and down based on the surrounding groundwater elevation and flows in the stream.
  15. Okay - so the way your first post was worded, it sounded like you were getting different head results in the NWT than the USG runs. Your latest post suggests that what you are comparing are "observation" results (which is probably why the topic is titled as it is). Can you confirm that? Can you also confirm that by manually checking cells in the NWT and USG simulations that the heads are similar? I know that I had issues with observations in my USG simulations (and seem to recall asking about it as well - there might even be a post on this board about it), so I managed it in other ways (select specific cells and get their results manually - not much fun, but I didn't have time to do anything else).
  16. I think the difference may be in the MODFLOW build itself. While working with MODFLOW-USG a couple of years ago (which, if I understand correctly, would use similar code to MODFLOW-NWT), an updated version came out which changed how calculations were performed for "dry" cells (due to users who didn't agree with how the solver handled those cells). It changed my model results and would have negated the work I had done for half a year. Working with the GMS developers, we solved it by using the older MODFLOW executable during my model runs. It is easy to do this by directing GMS to run the older executable from another location. I would check the year that the output file says the NWT results were produced and then look at the USG output for the version there. I think that the new version of USG came out in 2018. I have had to use the previous version (2015) to run my model.
  17. Well, based on what I'm seeing, you have one side of your model that has a constant head boundary condition and the other 3 are No flow boundaries. If this is the case, the only way for the water (which you are adding through recharge) to leave the model is through the constant head boundary. Therefore, the groundwater elevation has to get high enough for the gradient to want to flow towards the constant head. At a minimum, it will need to be at the elevation of the constant head boundary. If the conductivity of the peat layer is too low, this will create quite a mound. I'm thinking that you may not want a No Flow boundary in all locations, but I don't know everything about your model.
  18. Sorry that my suggestion hasn't worked out - I haven't used UGrid before, so I don't have an answer for you.
  19. An excellent question. I don't know if the GMS tools will work to do this or not with the Ugrid (I haven't tried it in your situation). In other cases, I would take a side view of the grid (general mode), select the layer with the layer select tool, right mouse button, redistribute from 1 to 2 layers.
  20. To simulate it exactly as you show it, you would have to split Layer 1 into 2 layers, so that the HFB goes up to the elevation you choose. The other way to do it (without splitting layers) could be to allow flow through the HFB that is equivalent to the fraction of layer thickness that is open (the HFB is across the full thickness, but not totally impermeable).
  21. The 3 sides with rock don't need anything. The 4th side needs some condition to allow water to leave (or enter), such as a constant or general head boundary...river/stream, etc.
  22. Tough to describe here, but I'll try to do it briefly. You actually have to set it up to have 2 values for the same time. The first value would be zero and then second will be one. Then when you want to end having it be one, you make another entry for that same time and make it zero. It ensures that there is no averaging from one time step to the next.
  23. Interesting. I don't really have time to look into it further, so I'll brainstorm with you. I'm wondering if setting IPHDRY to a positive number could be causing the problem, depending on what you are setting HDRY at. Try putting IPHDRY to 0 to see what happens. The cell would then clearly not be available for particles. " IPHDRY is a flag that indicates whether groundwater head will be set to HDRY when the groundwater head is less than 1×10-4 above the cell bottom (units defined by LENUNI). • If IPHDRY=0, then head will not be set to HDRY. • If IPHDRY>0, then head will be set to HDRY. If the head solution from one simulation will be used as starting heads for a subsequent simulation, or if the Observation Process is used (Harbaugh and others, 2000), then HDRY should not be printed to the output file for dry cells (that is, the UPW Package input variable should be set as IPHDRY=0).
×
×
  • Create New...