Jump to content
GMS, SMS, and WMS User Forum

Chris Blount

  • Content Count

  • Joined

  • Last visited

  • Days Won


Everything posted by Chris Blount

  1. I think I found a minor but annoying bug in SMS 10.1 with the 2D mesh polygon properties. I'm wanting to interpolate a different dataset onto part of my grid so I created a polygon to isolate the area of interest and used the existing nodes mesh option. I've done this plenty of times in the past. I recently used the Truncate Values option in the Interpolation dialog box and it worked fine. However, every time I tried a reinteroplation after that the values were truncated with the previously specified values even without the box being checked. I even tried changing the truncation values to be well outside the values of my dataset so it wouldn't affect the calculations, but the original truncation values were still used. I was able to solve this by changing the coverage type to something other than ADCIRC and then changing back to ADCIRC. That seemed to clear everything out so could reset the options. Everything works now, but I thought I would give everyone a heads up. Chris
  2. Is there a way to change the orientation of contour colorbars to horizontal? If not, I'd suggest this for the next release with more options for the labels as well (ie., show only every other or third label, etc.). Being able to export objects in an EPS format as well would be helpful for creating better figures in other software like Adobe Illustrator. Thanks, Chris
  3. Chris, I've had a similar problem when I tried merging meshes. However, try going to Nodes -> Options and check "Retriangulate voids when deleting" if it isn't already. Please let me know if merging with this option on works. I haven't tried it yet, but I was going to try it myself when I needed to merge meshes again. Chris
  4. The fort.22 is the generic wind input file and the format depends on the NWS setting in the control file (fort.15). See the fort.22 page on the ADCIRC site for descriptions. You'll have to process the netCDF files to match the format you pick. Chris
  5. For the fort.19 error, it means you have an elevation specified boundary in your grid (fort.14) but since you don't have the boundary forcing specified in the fort.15 (NTIF and NBRF) it's looking for the fort.19 with non-periodic forcing data. If you're wanting to use tides, then use the tidal forcing tab in the model control to set up the boundary forcing. Otherwise, you'll need to supply a fort.19 file. I'm not sure if you can set the fort.19 up in SMS or you have to do it yourself because I've never used it. The second part about tau0 is not a fatal error, but it is important for accuracy. From the ADCIRC webpage: "TAU0 = Generalized Wave-Continuity Equation (GWCE) weighting factor that weights the relative contribution of the primitive and wave portions of the GWCE. If "primitive_weighting_in_continuity_equation" is specified as a nodal attribute in the fort.15 file above, this line will be read in but ignored. If a nodal attribute file is not used or "primitive_weighting_in_continuity_equation" is in the nodal attribute (fort.13) file, but not specified in the fort.15 file this TAU0 parameter will be used. = 0, the GWCE is a pure wave equation. > 1, the GWCE behaves like a pure primitive continuity equation. A good rule of thumb for setting TAU0 is to set it equal to the largest value of an equivalent linear friction factor (e.g, for linear friction TAU0 = TAU; for quadratic friction TAU0 = maximum (speed*CF/depth). Typical values for TAU0 are in the range of 0.001 - 0.01. = -1, the TAU0 is spatially varying but constant in time; it is calculated according to depth as follows: If the depth is >=10 TAU0 is set to 0.005, if the depth is < 10, TAU0 is set to 0.020. = -2, the TAU0 is spatially varying but constant in time; it is calculated according to depth as follows: if the depth is >=200 TAU0 is set to 0.005, if the depth is < 200 but > 1, then TAU0 is set to 1/depth, and if depth < 1, TAU0 is set to 1.0. = -3, the TAU0 varies spatially. In high resolution areas, it also varies in time, and is dependent on the local friction and a tau0 "base" value (set by the user in the fort.13 file). High resolution areas are designated by a user by setting the tau0 in those areas to a value greater than 0.025 then Tau0 is calculated by Tau0 = Tau0base+1.5*Tk. The ceiling for the resulting tau0 in the calculation is 0.2. If Tau0base < 0.25 then Tau0 = Tau0base. = -5, the TAU0 varies spatially and in time, and is dependent on the local friction; it is limited to a range specified by Tau0FullDomainMin and Tau0FullDomainMax. If the above values have 0.1 added to them (i.e., resulting in 3.1, 5.1, etc), then the tau0 is also written out to a fort.tau0 file in the same format and with the same timing as the fort.63 file. The purpose of this file is simply to allow the user to verify that the tau0 values are being calculated and set as intended." I've been using tau0=-5, but you also have to manually add another line with Tau0FullDomainMin and Tau0FullDomainMax (see the fort.15 page). Chris
  6. I didn't notice anything wrong other than a zero WTIMINC. This is the time interval between wind inputs and has to be greater than zero. Chris
  7. I'm not in the loop in terms of development, but I have noticed a few slight changes between v47 and v48 in the input format. This may not have been fixed in the preproscessing done by SMS. Can you post your fort.15? Chris
  8. It means that something isn't formatted right in the fort.15. Did you manually change anything in the fort.15 after SMS produced it? What version are you using (should say in the fort.16 output header)? Chris
  9. It might help to know the intent of the model. If you want to force the entire model with a single wind speed then NWS=3 or 5 might work. You'll have to preprocess the data to the appropriate format and name the file "fort.22" for use by ADCIRC. The ADCIRC homepage has an example for NWS=3 (APES test case) that might be useful to look at. Chris
  10. Abhishek, Setting NOLICAT to zero turns off the time-derivative terms, but I don't think this is what you'd want. I'm not sure exactly what you're referring to without knowing the reference, but I'm guessing you might want to look at the harmonic analysis features of ADCIRC. From this, you can get the steady component of water surface elevation (fort.53) and velocity (fort.54). I hope this helps. Chris
  11. I don't know about the SMS problems, but the ADCIRC error is due to a change is how than changed the input for submergence state is given from v45 to v46. In v45 a fort.12 was used to specify the submergence state and now a fort.13 file contains input nodal attributes (variable friction, the submergence state, etc.). See the ADCIRC documentation. Version 48 isn't up there yet, but I don't think anything has changed between 47 and 48 with regards to the nodal attributes. Chris
  12. Aha! Consider this my bonehead question of the week. I didn't know you could edit the file name during the import! It just never occurred to me to try it there. Thanks, Chris
  13. Rusty, Is there a way to rename mesh datasets in 10.1? For example, when I import a maxele.63 file or multiple fort.63 files from different runs with ADCIRC the default name is "Water Surface Elevation (63)". I used to be able to change this in 10.0 by right clicking on the name. I can still do this with scatter datsets. Also, I've never been able to get the mesh datasets' new names to be saved for later sessions. For each new session I had to reload the files and rename them to keep track of everything. Both of these functions would be extremely useful especially since the new SWAN+ADCIRC program outputs files in unit 63 (H, T, and rad. stress) and 64 (direction) formats and they all end up with the default names when opened in SMS. Thanks, Chris
  14. Amy, Only the serial, pre-compiled for Windows version comes with SMS. I had the same problem so I can tell you you'll have to get the code from the development group and compile it on your own. The code doesn't cost anything for university research. If you know Jason Fleming at UNC he is the one that gave me v47.27 so he could set you up. Otherwise, I'd suggest contacting Drs. Luettich or Westerink via the addresses on the ADCIRC homepage (http://www.adcirc.org/development_group.html). Chris
  15. Sorry, I don't have the other tutorials. I kept that from when someone else asked about it. The only tutorials I have are the ones that come with versions 9.0+. Chris
  16. For the SMS team, For several of the constituents (2N2, L2, MU2, NU2, and T2), the amplitudes for the tide potential forcing in the ADCIRC model control is zero and the reduction factors are 0.069 (should be 0.69). Why is this? Is it because the others just weren't included in the ADCIRC report? The reference used for those are Reid (1990). Where can I get these amplitudes and reduction factors? Thanks, Chris
  17. Ian, this is probably what you're talking about. Chris 14_Lesson04.pdf
  18. John, My raster is going to be too big to load as a scatter into SMS. I'm not really wanting to filter the data because my grid and raster resolutions will be comparable. So I'm thinking I'll have to work with this in ArcGIS. Is there a more direct way of doing this other than converting the grid to scatter, importing to ArcGIS, extracting the raster to those points, and then importing the points back into SMS? Also, when using the "existing nodes" option for paving, does that just reinterpolate the elevation? Thanks, Chris
  19. John, The latest update fixed the problem with the colors. Thanks for your work. Chris
  20. I'm relatively new at using ArcGIS so I'm not sure how to do all of the data manipulation, file conversions, etc. However, I'm going to be using pretty large datasets for topography, landcover, etc. for an ADCIRC grid. So how do I go about interpolating the data onto the mesh? For example, I have 90 m topo in GRID format which I can open in SMS when I enable ArcObjects, but how do I actually interpolate it to the mesh? I've also tried creating a TIN from the mesh to use in ArcMap, but it doesn't recognize the file. Any help is appreciated. Chris
  21. It works for me as well. Thanks for the quick work! Chris
  22. John, Please let me know when the new version is ready. Thanks, Chris
  23. John, I just installed the beta again and now it crashes every time I try to open it. I even tried an uninstall and reinstall and it still doesn't work. Chris
  24. I can't seem to get the nodestrings for ADCIRC to display right in the 10.1 beta. It seems like the different boundary conditions don't show up as the representative colors. I thought maybe it was stuck on the default color, saved as part of the project and I couldn't figure out how to change it back. So I tried just opening the grid I was working on in a new session on 10.1 and it still didn't work right. So I tried it in 10.0 and it works okay. Also, when I tried to renumber the grid I got an error saying something about part of the grid being disjoint (which was true) and it was highlighted in red (which it wasn't). So that's another display problem to check. While I'm posting about this, I have two other general comments about nodestrings. 1.) I think the instructions that display on the bottom about going CW and CCW are reversed. 2.) The nodestring direction depends on whether it's an internal or external boundary so it would be helpful if there was a way you could incoporate a check or automatic correction of the direction when the boundary condition is assigned. I'm working with large grids with lots of islands (100+) in some instances and I've had problems with the nodestring problem before. And this is a rather tedious thing to check. Chris
  25. There is a lot of literature on this subject, but usually you want tau0 to be 1-10 times your bottom friction if memory serves me correctly. ADCIRC can calculate a spatially (and temporally if desired) variable tau0 automatically in v47. This removes the necessity to repeatedly tweak tau0. SMS doesn't support this but you can manually change the tau0 value in the fort.15 file to a specified negative number for different settings. Please see the manual at http://www.adcirc.org for the option settings. Chris
  • Create New...