I'm trying to use QGIS (2.6.1-Brighton) to aggregate polygons according to an ID attribute.

I'm using the shapefile of school district polygons in Michigan found here:


MI Geographic Framework School Districts (v14a) is the name of the shapefile I'm working with.

I'm trying to create a shapefile of Intermediate School Districts, which in Michigan are a level of aggregation above school districts.

I should be able to do this using the dissolve tool, as there is an ID variable in the v14a shapefile (called ISD).

However, upon doing so, the resulting polygon layer appears to have been re-projected and rotated.

Here's the input polygons: enter image description here

The CRS of this layer is strange, but here's the definition:

+proj=omerc +lat_0=45.30916666666666 +lonc=-86 +alpha=337.25556
+k=0.9996 +x_0=2546731.496 +y_0=-4354009.816 +no_uoff +ellps=GRS80
+towgs84=0,0,0,0,0,0,0 +units=m +no_defs

And here's the result of the dissolve: enter image description here

With CRS:

+proj=omerc +lat_0=45.30916666666666 +lonc=-86 +alpha=337.25556
+k=0.9996 +x_0=2546731.496 +y_0=-4354009.816 +gamma=0 +ellps=GRS80
+towgs84=0,0,0,0,0,0,0 +units=m +no_defs

Any idea what's going on here??

I would be particularly grateful if there was a way to reproject the rotated layer so as to undo the rotation, as I had already done a fair amount of stuff by hand to add new attributes to the merged layer (so that starting from scratch would require a lot of repeated manual labor).

  • 1
    what's the CRS of the input layer and what's the CRS of the output layer?
    – underdark
    Jan 3, 2015 at 11:22
  • See edited question; seems clear now that what happened is +no_uoff became +gamma=0 (fixing this solves the issue), but what's unclear is why this would happen... Jan 4, 2015 at 4:01
  • eventually just settled for using SAGA to dissolve the polygons. I'm beginning to think I should just use SAGA for all geoprocessing... Jan 4, 2015 at 5:06
  • 1
    I'm pretty sure that the difference between the two is the +gamma=0 in the second definition. Try taking that out, OR try setting the gamma value equal to the alpha value. The coordinate ref system is called Michigan GeoRef.
    – mkennedy
    Jan 4, 2015 at 14:26
  • So you ended up using SAGA for dissolving the data, but what about your manual edits on the merged layer? Jan 4, 2015 at 14:57

1 Answer 1


Assuming you don't use the latest QGIS version (you don't mention any version in the original post), upgrade your QGIS version to 2.6.1-Brighton. I've followed the work flow you described (same data, same dissolve base field) and got the dissolved data properly located.

As to the attributes you've already appended to the result, perform a table join (find the steps here) to transfer all your edits to your properly located dissolved data. Then use "Save as..." on your joined layer to get a new layer that includes both your properly located dissolved data and your manual edits.

If there is something unclear, please tell me.

  • Good point about using table join. I'm running QGIS 2.4-Chugiak and can't get it to update for the life of me. Jan 3, 2015 at 21:11
  • What operating system are you working on? Jan 3, 2015 at 21:12
  • Linux Mint 16 Petra; I've tried adding the unstable ppa (via sudo apt-add-repository ppa:ubuntugis/ubuntugis-unstable) but to no avail--only 2.4 is coming back. Jan 3, 2015 at 21:31
  • From QGIS website: "Default Ubuntu software repositories often hold older versions of QGIS." I work on Linux Mint 13 Maya and don't need to add any specialized repository for QGIS v.2.6.1. Jan 3, 2015 at 21:39
  • I think the problem is that Petra corresponds to Ubuntu's saucy distro; for some reason saucy is stuck at 2.4: launchpad.net/~ubuntugis/+archive/ubuntu/ubuntugis-unstable/… Jan 3, 2015 at 23:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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