Outputs with no data cells


I ran the SDR model, but the outputs (sed-retention, sed-exported) are with some no data cells.
I checked the intermediate files the (ic-factor, ic-factor bare soil, sdr factor, sdr factor bare soil, sed retention) had the same problem.

Can you help me to solve this problem

Thank you
Sana Bouguerra 


  • RichRich Administrator, NatCap Staff
    It might be related to the DEM being unable to route because of hydrological pits.  Can you look at the flow accumulation raster and see if you see nodata holes there too?  If so, it'd suspect that's the problem and you'll need to fill those pits with some GIS tool.  If not, we might need to look at your datastack which you can dropbox to me at richsharp@stanford.edu
  • Hi Rich!
    Thank you for the answer;

    the flow accumulation and the stream don't have a nodata cells. but, the rivers are scattered and not compatibles with the hydrographic network.
    do you think that's related to the DEM?
    I will dropbox the datastack to you.

  • RichRich Administrator, NatCap Staff
    Thanks Sana, I got your datastack, and yes the issue is that your DEM has hydrological pits in it.  You can see these by looking at the flow accumulation output of InVEST and clicking around the pixels on the DEM where you see an internal flow accumulation raster cut off.  

    As an example, I clicked around an internal flow accumulation pixel that terminated at a pixel at 79 meters.  The surrounding pixels were at 80 or higher.  This creates a discontinuity in the drainage layer which ultimately creates swaths of nodata regions in the SDR output.

    For the moment, InVEST requires that the DEM be free of hydrological sinks.  I don't have a lot of experience with this, but you can apparently fill these with ArcHydro in the Terrain Preprocessing -> DEM Manipulation -> Sink Evaluation and Fill Sinks.  Can you give that a try?

  • Thanks Rich, 

    it's works

This discussion has been closed.