6 replies to this topic

[François Goulet]

Hi!

We have data taken from a boat across a river. On the picture, the river bed is shown in brown and the water level is the thin black line.

As you can see from the classified points, the phenomenon we are interested in (here, water salinity) is very linear.

We normally use a natural neighbor interpolation because we have a lot of points, but here, the river is roughly 500 wide so the distance between the ups and downs is about 100 m and the distance between each points about 50 cm.

If we choose a large aggregation distance, everything is averaged... if we take a smaller one, you can see on the first image that it's not linear at all as it should be (we are not trying to prove that it's linear, we're just looking for the depth of the saltwater front).

We try with a TIN, but even with setting large triangles, the interpolation doesn't do it quite right. It's closer, but not it!

So, my question, it there a method to interpolate data by forcing it to "look" at a particular angle? Could we interpolate our data by asking to whatever application, to interpolate with data of the same depth first and if not, the closest one from that angle?

Thanks folks!

Attached Files

•  NaturalNeighbor.jpg   103.67K   46 downloads
•  TIN.jpg   115.95K   39 downloads

[GISRox]

What software are you using to do the interpolation? How much control do you have over your search radius and/or anisotropy?

[François Goulet]

What software are you using to do the interpolation? How much control do you have over your search radius and/or anisotropy?

We use Vertical Mapper and/or ArcGIS. For the rest, we have all the control we want, if it will display the data correctly

[Tim Baigent]

François,

This is crude but it might work ideally well.

This assumes use of Manifold.

If your sample data are points, paste them as a surface, choosing the target resolution and no interpolation. (This surface will run vertically, but there's nothing wrong with that.)

Next run InterpolateRowHeight over the surface. This will only look for data along the same row, so in this case at (roughly) the same depth. Interpolation is linear from the nearest pair of values (or if there's only one, that value).

Nulls will remain where no data could be found at the same depth. One way to avoid these would be to choose a coarser resolution, but a better way would be to search adjacent rows (up to some limit). I could throw some SQL together if it's useful.

Tim

Edited by Tim Baigent, 16 November 2008 - 08:32 PM.

[François Goulet]

This assumes use of Manifold.

Thanks! It's not that crude, considering that our methodology to import Doppler data into GIS is 3 pages and involve a dozen temporary files...

We don't have Manifold, unfortunately. I've been talking about it at work and it didn't seems to have interested anyone, since we can do pretty much what we want with ArcGIS, MaPublisher, MapInfo and Vertical Mapper...

I've meant to try Manifold for a while (thanks to Hans! ) and it might be the chance but it will be too late for this job. I have 36 of those interpolation to do and it has been decided by the project managers to do it by hand (or that I'll be doing them manually...) since we need them this week. It will not be that hard, just long and repetitive.

I'll keep your solution in mind and see if we could find an similar solution.

Thanks again!

[Tim Baigent]

This assumes use of Manifold.

I've meant to try Manifold for a while (thanks to Hans! ;) ) and it might be the chance but it will be too late for this job. I have 36 of those interpolation to do...

If you want to see the result in Manifold, you could send me one of the 36 and I'd be only too happy to send it back.

[François Goulet]

This assumes use of Manifold.

I've meant to try Manifold for a while (thanks to Hans! ) and it might be the chance but it will be too late for this job. I have 36 of those interpolation to do...

If you want to see the result in Manifold, you could send me one of the 36 and I'd be only too happy to send it back.

And I'll be only too happy to see what it looks like! I'll see what I can do tomorrow!

Thanks!