Jump to content
GMS, SMS, and WMS User Forum

Diverting a Drainage Path


Recommended Posts


I'm a GIS Tech and have been tasked with producing a number of hydrographs for some culvert installations within a forested water supply catchment.

I'm using a 2m LiDAR DEM (converted to ASCII for import) and for the most part the TOPAZ-derived drainage network is representational of reality in the field.  For those areas where WMS is sending the drainage down an incorrect path I need to make adjustments to both the drainage and its associated basin.  I can't figure out how to do this!!  I saw the tutorial video on how to divert runoff along a topographic barrier (the example used was a railway), and while this is close to my scenario, it differs in that the landscape barrier - in my case a roadway - IS being recognised by WMS.  It just isn't seeing the cut I added to allow water to pass through.

I need to be able to re-route a stream over what must be a small 'bump' that still exists in the DEM somewhere.  Is that even possible?  I've tried selecting the path and moving a vertex over to the other side of the catchment edge and it just resets back when I run the Delineate Basins Wizard.

See attached image for help interpreting my problem.

Thanks in advance,



Link to comment
Share on other sites


This is exactly what I ran into the other day for a project I was working on.

One possible workflow would be to edit the DEM using a feature arc. This is done by creating a new coverage, digitizing an arc through the center of the culvert (terminating it at the toes of the roadway embankment, or just beyond), editing the underlying DEM cells along the arc, and then rerunning TOPAZ. You can read more about this process in the online documentation: http://www.xmswiki.com/wiki/WMS:Editing_DEM_Elevations.

An easy way to get the "Edit DEM Elevations" to pop up is to select the arc running through the culvert and then right click on it and select "Edit DEM Elevations". In the "Edit DEM Elevations" window, you can multi-select the first and last points in the plot by holding down shift and then you can press the "interpolate" button to linearly interpolate the elevations between the two selected points. When you select OK, it will edit the DEM elevations accordingly. Don't forget to rerun TOPAZ to pick up the new drainage path and then re-delineate your basins.

If after doing the above, TOPAZ still does not pick up on the new drainage path further editing could be required such as widening the gap by repeating the process with a parallel arc or extending your culvert centerline further upstream/downstream.

Let us know how it goes.


Link to comment
Share on other sites


What Cody recommends is a good solution.  I would add that sometimes TOPAZ won't change the flow path even if you edit the elevations, so you might need to change the elevations of several cells around where you want the flow path to change to make it obvious to TOPAZ where you want the flow to go.  You can edit individual DEM cell elevations by double-clicking on the DEM points as needed.  But I'd try Cody's approach of adding a single arc first and see what TOPAZ gives you.


Link to comment
Share on other sites

Thanks guys. Very useful; I'll give those ideas a try and report back.  

It was a fine balance with the LiDAR DEM when I originally exported to ASCII for WMS.  I didn't want to make the culvert 'holes' through roads too large or they would start impacting the surrounding terrain unrealistically.  

My basic process was to run raster math on the original DEM.  As mentioned it was a 2m dataset.  First, I made the roads 10m tall to ensure they were a barrier, then cut 2m holes in them where each culvert was situated.  Then I tried both a 5m buffer depression and a 10m buffer depression surrounding the culvert.  Attached is a screen capture of the DEM surface after cutting the culvert hole and using a 5m depression.  Brown = higher, Blue = lower...


Link to comment
Share on other sites


Just reporting back on my flow pattern issue.

Editing the DEM as described worked perfectly, transferring the water across the road, but in the end I had to come at the problem a different way.  Cody's method was convenient and straightforward but in my dataset I just had too many 'misdirections' to make it feasible.

Instead I experimented with widening the gap in my roads.  Initially the culvert location was a single pixel hole in the road's berm.  I created a depression around this gap (trying various diameters up to 22 metres wide) to pull the water toward the culvert, it just wouldn't go through when TOPAZ ran.  However, when I widened the gap in the road to be 10m (5 pixels) and kept the surrounding collection zone to 10m, then TOPAZ was able to send the water through without issue.

This approach will make it easy to repeat the calculation in a consistent fashion should I ever have to go back and adjust values for climate change, etc.

Below is a screen cap of the widened gap and 10 metre depression.

Thanks again Gentlemen.




Link to comment
Share on other sites

By the way - a couple other important points to this topic that others might find useful:

1.  When we were running WMS on a 10 metre pixel DEM there were NONE of these issues.  Culverts, flow patterns and drainages all came out exactly as expected.  It wasn't until we tried to refine the data by tightening up the resolution that the problems appeared.

2.  When I used the ArcGIS Hydrology toolset to derive the Flow Direction and Accumulation surfaces there were no problems with the result at either 2m or 10m.  The flow went through at culverts just the way you would expect... and originally, when running the data at 10m, I had used this approach rather than using TOPAZ to calculate Direction/Accumulation within WMS.  No real reason why I chose that - it was just due to my familiarity as a GIS Tech with the ESRI dialog.

However, when I switch up to 2m resolution, and ran the Direction and Accumulation in ArcGIS, they wouldn't import correctly into WMS.  No matter what dataset I was working on - if it was 2m resolution - the WMS result from import was incomplete.  The drainage patterns would look fine for about 1/2 of the area in question and then would just stop.  Almost as if it was processing from top-down and hit a maximum number of values, or was tanking out on memory or computations...

Below is a screen cap of the stopping line.  You can see the hanging flow lines left when the import seems to have just 'stopped'.

Hope it's useful,



Link to comment
Share on other sites

  • 3 months later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.


  • Create New...