Datum Conflict & Watershed Model

12-14-2018 07:55 AM
New Contributor III

Thanks and maybe there are two questions in this situation. I am not sure if the datum conflict is causing the unexpected watershed model or if there is something else I am doing wrong in the process.

In the process of creating a watershed for a lake, see images, I received the datum conflict warning 001003. Reading notes on the warning, it would appear that it can be ignored but...is that always or only in some processing situation? Ignoring the warning, I continued the watershed modeling steps and on two occasions the watershed extends beyond/downstream of the pour point I placed at the lake outlet; see image. 

What is causing the watershed to extend beyond the pour point? My understanding is that all the watershed should define a surface that flows into the lake and therefore all the water in the watershed would flow out the lake at the downstream point/pour point.

0 Kudos
2 Replies
MVP Frequent Contributor

First, check your data to see if it is in the same projection/coordinate system.  It is usually best to place all the data in the same projection/coordinate system before processing it.  In theory, one could ignore the warning, but it may be that the warning is relevant.  Given the uncertainty, reduce that potential by putting all your data into the same system before processing.  Data prep on the front end goes a long ways to head off later issues.

001003: Datum conflict between input and output.—Help | ArcGIS Desktop 

As to why the watershed may be derived "downstream" of the pour point, to troubleshoot this I would start by looking at the output surface and checking the elevation values.  If the elevation values "downstream" of the pour point are the same value as above, that usually indicates a data issue.  Check the DEM used as the input to to the Watershed process.  Maybe it is too coarse for what is being attempted?  For example, the cellsize is some large value like 1 mile when what is needed is a resolution of 10 feet?   Or a Fill process run before the Watershed geoprocessing tool caused the area "downstream" of the pour point to be "leveled out".  Just examples of what can go awry.  You will have to check the data.

Along with this, confirm that the pour point is really in the correct and relevant location.  Is it located right at the top of the spilllway of the lake or instead in the lake many feet away?  If the latter, that will likely be the issue.  Any chance the Snap Pour Point process didn't work out that well?  Review the data and see if its location makes sense given other reference information.

Chris Donohue, GISP

New Contributor III

Thanks Chris.

All good points. 
I will implement your suggestions one at a time to see which if any make a difference and share results for the GeoNet community.


0 Kudos