runoff graph is jagged

Question

Has anyone had a jagged graph for a catchment? I was playing around with creating time depth curves from IDF curves and my catchment graph looks like this. The data table shows a bunch of zeros. 
 
 /cfs-file/__key/communityserver-discussions-components-files/5925/hydro-stsw.zip

WPsquared · Answer

Brian, 
 What solver are you using? GVF-Convex, SWMM, or Implicit? I would suggest first trying a smaller time step to see if that helps a little. 
 Regards, 
 Mark

WPsquared · Answer

Try a one minute time step and a 30 second time step to see if that helps. You can also try changing the output increment. In this case if you're not running just the hydrology and running the entire model then you'll also want to make certain that you have a good continuity error, usually under 5%, but it depends on the complexity and size of the model (the closer you can get to 0% the better). If you need help troubleshooting the continuity error or haven't yet tried to achieve a better continuity error have a look at this wiki . To better understand what the continuity error is this wiki will explain it in more detail. If you're not locked in to using the implicit engine you might want to look into using the SWMM engine. Sometimes it can provide more stable results depending on the model setup and input. 
 Mark

Scott Kampa · Answer

Hello Brian, 
 First, you currently don't have a conduit connected to the outfall. Based on the red user notification, this doesn't appear to be a valid setup. Is there something that the catchment drains too, like a ditch or conduit? If so, modeling this is recommended, since you will likely not be able see results at the outfall. Anytime you see a red user notification, this indicates an issue that needs to be fixed before you have valid results. 
 As for the appearance of the hydrograph, this appears to be related to the storm data. If you look at the curves in the Storm Data dialog, the graphs are not smooth. This is likely the cause of the appearance of your hydrograph. As a test, I created a new TypeII SCS storm with a depth of 3 inches and computed your model with that. The hydrograph appeared smooth. 
 Regards, Scott