2

I need to interpolate water quality data in a rather small area with peninsulas, islands and developed coastline that will act as barriers between my testing/sampling points.

I understand I have to use non euclidean method to do so, but does anyone know the difference between "Diffusion Interpolation with Barriers" and "Kernel interpolation with barriers"?

closed as unclear what you're asking by PolyGeo, Chad Cooper, Brad Nesom, Mapperz Jun 18 '14 at 14:24

Please clarify your specific problem or add additional details to highlight exactly what you need. As it's currently written, it’s hard to tell exactly what you're asking. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • Can you supply references for these two terms? – whuber May 30 '13 at 21:27
  • Hi. Not sure what you mean by "references". The two methods I am asking about are in the Geostatistical toolbox, and both seem to interpolate data around borders. I used the "polynomial 5" parameter for the kernel method which seems to be the most appropriate parameter for what I want to do based on my research and readings. I also read that kernel gets better results but I would like to really understand the difference between both to make the best selection. So far it seems that both use non euclidean to go around borders. Thanks. – timo May 30 '13 at 21:54
  • Thanks. These aren't standard terms: it was essential that you tell us where they came from. Now we know you are using Geostatistical Analyst. The answers to your questions therefore must come from the GA documentation. You could help readers by providing links to the particular documentation you have been reading: those are the "references." – whuber May 30 '13 at 22:24
  • For the "diffusion interpolation" method I found this but it has some inconsistencies and lacks the details needed to offer any kind of authoritative answer. The help page for "KI with barriers" is even less comprehensible--it reads as if translated directly from Russian, meaning the words it contains are appropriate for the context but when put all together don't tell us anything specific. It refers to a 2010 paper but access is restricted. – whuber May 30 '13 at 22:48
  • Thanks whuber! I had looked at the ArcGIS resource center/help menu and had found the same link. As you said it lacks some clear information/explanation and was hoping to find more info on this forum. I also found the article you're mentioning but it is very abstract. Thanks anyway. – timo May 31 '13 at 1:21
1

I am using the R package ipdw to do something very similar. You can find it at http://www.cran.r-project.org/package=ipdw

  • Welcome to GIS SE! I think the Question you have answered could do with some editing to make clear whether it required an ArcGIS for Desktop solution. Given that ambiguity and the passage of a year with no Answers, I am happy to upvote your R answer to what I think is an ArcGIS for Desktop question. – PolyGeo Jun 18 '14 at 0:59

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