0

I'm using QGis and I have a set of points with a depth attribute. I'm trying to make an interpolation using the Raster interpolation plugin (Raster -> interpolation -> interpolation). But when I do so I get a misplacement on the produced raster. It's positioned too far north. Do you have any ideas of what may be causing this?

To make it clear what I am doing and what my problem is I've created a short video where I'm showing the issue: http://jimutt.com/interpolationProblem.swf

As you can see in the video I'm using the same projection in both the viewport and for the point layer.

Please tell me if you can't display the video.

  • How were the points created? Are you sure they were actually in that CRS and datum, and not just defined as such? Are they being reprojected on the fly to the dataview's CRS even though they're actually in a different CRS? – Chris W Jul 1 '14 at 21:38
  • The points are from a Garmin gpsmap 421 chart-plotter device. I've checked the settings and it says that the projection used in the device is WGS 84. And so I think it should be correct. Worth mentioning is that when I use Raster -> Analysis -> Grid (Interpolation) it gets positioned correctly. Though I'm having a hard time when using the grid (interpolation) to get a result as good as when using the interpolation plugin. – jimutt Jul 2 '14 at 8:52
  • I still don't have an exact solution for you, but I noticed one thing in your video and had another general obvservation to point out. First, interpolation is usually best done in a projected coordinate system rather than geographic as you're using. You might want to project to the appropriate UTM zone before interpolating. Also, I notice in your video that the output cell size x and y values are not the same - .00006 and .00002 respectively. I'm wondering if either of these might be the cause of the y shift in the raster. – Chris W Jul 2 '14 at 21:55
  • Thank you, I will definately give it a go and see if any of your suggestions can help me out! – jimutt Jul 3 '14 at 7:11
  • Looks like it was the cell size values that caused my issues. Thank's for the help Chris! If I only could mark your comment as an answer. – jimutt Jul 3 '14 at 19:41
0

I notice in your video that the output cell size x and y values are not the same - .00006 and .00002 respectively. I would expect this to cause a compression rather than just a shift (or offset) in the y axis results, but then it's kind of hard to tell from the video whether it's just one or the other.

Also, interpolation is usually best done in a projected coordinate system rather than geographic as you're using. You might want to project to the appropriate UTM zone before interpolating.

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.