DEM Reconditioning error

5039
5
12-19-2017 10:22 AM
Labels (1)
KatieNance
New Contributor
I work for tech support at a university. A professor has an assignment that has students use DEM Reconditioning through HEC-GeoHMS 10.3, but he is having an error with the assignment. The basic process is as follows:
  1. In C:\temp, create a folder for the project
  2. In ArcMap, navigate there, create a default geodatabase, add a toolbox to the geodatabase
  3. Import map grid data
  4. Data Management Tools > Projections and Transformations > Project Raster
  5. Spatial Analyst Tools > Surface > Slope
  6. Import flowline data
  7. HEC-GeoHMS > Preprocessing > DEM Reconditioning
Step 7 causes the error. Additionally, running DEM Reconditioning through Arc Hydro Tools 10.3.0.170 instead of HEC-GeoHMS also comes up with the error.
In previous versions of ArcMap, Arc Hydro Tools, and HEC-GeoHMS, we have gotten various errors every single year with this same assignment and dataset that we have been able to solve. A few years ago, we were getting this exact same error message. At the time, we uninstalled the three programs and reinstalled them and it worked. Hoping that would be the case again, I tried it and still got the error.
To rule out software interactions, I ran the assignment on a computer with nothing but those three programs installed and it still failed.
I also tried doing the basic assignment with fresh data obtained through USGS, thinking maybe the data got corrupted or was otherwise faulty, and still got the error, so it is not bad data.
I tried the assignment with default generated layer names and locations, still didn't work.
We have followed every possible suggestion listed online, like short file names, working locally, clearing the temp directory, etc in order to have the maximum chance of success. I have also tried setting the targets, no luck.
The full error output from HEC-GeoHMS is attached. I get the same output if I run DEM Reconditioning from the Arc Hydro Tools toolbox
I need to figure this out before spring semester begins and I'm at a loss. Any ideas on how to fix this?
Let me know if you need anything else. I'm not very knowledgeable about ArcGIS terminology etc so bare with me.
EDIT: Attached data
0 Kudos
5 Replies
MarkBoucher
Occasional Contributor III

Make sure all of the data is in the same coordinate system and the processing extents include the work area.

Mark Boucher

Sent from my iPhone

0 Kudos
KatieNance
New Contributor

All of the data is in the same coordinate system and the processing extents are correct.

Nothing about this has changed from previous years, where after a little troubleshooting (wrong versions of component programs, incorrect directory permissions, etc) it worked.

0 Kudos
KatieNance
New Contributor

Apparently DEM reconditioning works with the data from the tutorial here. I noticed that the stream data for that one is not in a geodatabase, instead it is a .shp file. To confirm this, I tried running DEM Reconditioning on USGS data for about the same area as the data attached to my first post that used .shp files for the stream data instead of geodatabases and it succeeded.

In my previous testing with fresh data that failed DEM Reconditioning, I used geodatabase format for the stream data. to match the formats the professor was using.So it looks like Arc Hydro Tools/HEC-GeoHMS has an issue with the data being in a geodatabase. Perhaps this is what has been contributing to the errors in previous years.

Is there any way to solve the issue with using geodatabases?

0 Kudos
MarkBoucher
Occasional Contributor III

Katie,

I downloaded your data and went straight to DEM reconditioning and had now problems. I am using 10.4.1 and can't use 10.3. I ran the reconditioning and it worked fine.

I also ran my typical process on the original DEM (not reconditioned) fill sinks-flow direction-flow accumulation-stream definition-stream links-catchment grid delineation-catchment poly processing-drainage lines. Everything worked fine. 

As for having the flowline date in a geodatabase, I've always found geodatabases to be more stable and my practice it to put the data in a geodatabase as a matter of course. So, your finding conflicts with mine and I'll have to keep a heads up.

I had problems in the past with versions of HEC-GeoHMS and ArcHydro not being based on the same version. When I "mixed" the versions, I would run into problems.

One thing I do is ignore streams and rely on the DEM to define the streams except when there is a problem area like a bridge or embankment that the DEM "doesn't know about". I set lines and burn through those. I do burn in pipes especially in the flat lands.

Attached you can see a version of the DEM with the streams burned in (reconditioned) with the drainage lines from the non-reconditioned (non-burned) DEM. The drainage lines do not always follow the NHDFlowlines. This tells me that the flow lines are likely manually created or are out of date (ie. based on an older DEM).

Best,

Mark

KatieNance
New Contributor

Thanks for getting back to me, and sorry for the delay. The university was closed for Christmas.

Your findings are interesting. I attempted to run DEM reconditioning with nothing but the base map and flowline data and HEC-GeoHMS says "Please add data with appropriate projection to the map, save and try again." Running it with Arc Hydro Tools does not produce this error, and instead runs DEM Reconditioning successfully.

Since the professor wanted to run HEC-GeoHMS DEM reconditioning, I ran project raster on the base map to satisfy the error message. Then, if I run DEM Reconditioning through HEC-GeoHMS with the project raster layer as the Input Raw DEM, I get the same error as originally. Same with Arc Hydro Tools.

Here's what confuses me. If I create the project raster level and run HEC-GeoHMS with the base map as the Input Raw DEM instead of the project raster level, it succeeds.

I personally am satisfied with that result, but I am not sure if the DEM reconditioning and subsequent steps (the professor has students run fill sinks, flow direction, flow accumulation, stream definition, stream segmentation, catchment grid delineation, drainage line processing, adjoint catchment processing) would be affected by the data not being based on the project raster layer. I ran the subsequent steps and to my untrained eye it looks OK. Can you think of any reason this would be unacceptable?

Thank you so much for looking into this for me.

0 Kudos