Mosaic blend function

[DivanVermeulen]

I'm having a problem when mosaicing raster DEMs together using the 'Mosaic' or the 'Mosaic to new raster tools'. We only have this problem with ArcMap 10.1. When mosaicing the rasters, the areas where the overlap occurs does not "BLEND" as it should (a line or zone is left within the raster which clearly shows where the overlap occurs). I found when running the script manually, it provides the same output as say the "FIRST" or "LAST" Mosaic Operator. Running it in an arcpy script solves the problem in ArcMap 10.0 SP4, but in ArcMap 10.1 the problem still persists. We never had this problem in ArcMap 9.3 and as I said, not in ArcMap 10.0 SP 4 (when scripted). The DEMs which we need to mosaic needs to be given to clients of ours, but if they are using the latest version of ArcMap 10 they won't be able to mosaic the DEMs correctly.

What I would like to know is if someone has run into the same problem I have, and if there is any known solutions or if ESRI is aware of the problem? Is there maybe any updates to ArcMap 10.1 which fixes the problem, or is there a script which someone maybe set up to solve the issue? Any other suggestions will also be appreciated!

Thanks,
Divan

[GayathriAlallasundaram]

Hi,

It seems to me that you are running into this bug in version 10.1.1 NIM086790(The blend mosaic operator in the Mosaic To New Raster tool generates unexpected results.).

Thank You,
Gayathri A.

[JenniferMcCall4]

Hi Divan,
I think I might have had the same issue as you.  I had two rasters and they blended along the seamline and produced an out-of-place line along where the seamline was.  This only happened to me where there was another raster underneath the other two rasters that were not blending properly.  To resolve this, I created a new mosaic dataset and added just the two rasters that i wanted as the top level of my mosaic dataset.   I did not apply a blend to this dataset.
Next, I exported the mosaic to a tiff (I found it worked better exporting to a geodatabase than just a file though).  Then I created another mosaic dataset with the new tiff and the rest of the rasters I wanted in the dataset.  This removed the blending issue between the two problematic rasters, and the ones that were not problematic could be blended to look more seamless.