0

I recently posted about a problem in ArcGIS using the Goode Homolosine (Land) projection with the Project Raster tool HERE.

The problem was related to a bug with that tool, and a workaround solution was proposed.

I have since tried to re-project different data to the Goode Homolosine projection, but the workaround solution does not work for these data. I find the proposed solution to work for some data but not others.

Here is an example of the proposed solution that produces an Inconsistent Extent warning:

  1. I downloaded this Crop Suitability Index raster data.
  2. I open these data in ArcMap and set the data frame's coordinate system to projected coordinate systems, world, Goode's Homolosine (Land).
  3. In the table of contents I right click the data and select Data, Export Data.
  4. When exporting I set the Spatial Reference to the Data Frame, and save as a .tif file.
  5. When I open the saved .tif file in a blank map I receive an Inconsistent Extent warning message.

I do not understand why this workaround solution isn't a general solution, i.e., why this solution worked with the Malaria Ecology Index raster data I originally posted about, but not these Crop Suitability Index raster data?

2

1 Answer 1

0

Well, we won't know for sure until a developer debugs the tool to figure out what's happening.

In the earlier dataset, the Malaria Ecology Index, the version that triggered the warning had one extra row and column of cells, so it could be a slightly different rounding algorithm when the projected raster's extent is calculated.

Without looking at the new dataset, it may have two columns or rows, enough that export data tool in ArcMap still has a row or column extra and the output triggers the warning.

Please note that the warning is just that--a warning. It shouldn't affect display or most work.

3
  • Thank you for the follow up. As you note that my problem relates only to a warning, I am curious then how to assess whether the inconsistent extent will actually have any effect on my work? Or perhaps you are aware of documentation I could read on this matter? I wish to calculate average raster values within polygons, but hesitate to move forward until I am sure the inconsistent extent is not affecting my data work. Thank you.
    – acd
    Commented Aug 31, 2015 at 16:25
  • You could trim the raster to remove the offending row/column. But if you're checking values within a polygon, then the existence of an extra row or column at the edge of the raster shouldn't matter.
    – mkennedy
    Commented Aug 31, 2015 at 18:51
  • None of my polygons are near the edge of the raster, so I should be okay. Thanks for the clarification.
    – acd
    Commented Aug 31, 2015 at 19:07

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.