topic Re: Interpolation techniques in ArcGIS Spatial Analyst Questions

Interpolation techniques

PeterMelle — Wed, 02 Mar 2011 00:00:15 GMT

hello all,

I am trying to generate a depth to groundwater map using available data that I have. The data that I'm using is primarily well water data. I am unsure as to which method I should use, Inverse Distance Weighted, splining, or kriging. If I do choose kriging, should I use universal, ordinary, simple, etc.?

Re: Interpolation techniques

KylePurdon — Wed, 02 Mar 2011 00:49:01 GMT

All methods will work. Inverse Distance Weighted method is a bit more straight forward but if you want an interpolation method with a lot of variability in the model used go with Kriging. I typically dont use spline unless there is a very specific need for it. As far as Kriging goes universal considers a percentage of local v. global points used (easier understood as close versus far points) and ordinary is just as it states a more ordinary interpolation method. I would recommended if you have the availability use the "Geostatistical Wizard" for Kriging interpolation as you can create a much more custom model for interpolation. If your goal is a relatively quick, simple, and easy to understand interpolation IDW is the way to go but if you want a customizable very accurate interpolation Ordinary Kriging using the Geostatistical Wizard is your best bet.

Re: Interpolation techniques

PeterMelle — Thu, 03 Mar 2011 15:21:32 GMT

Thank you both for your input, it was very helpful. The map I am generating is a groundwater elevation map for Luzerne County in Northeastern Pennsylvania. The geology of the area is very unique, as there are both confined and unconfined aquifers within the county. I will consider both techniques and consult a local hydrogeologist ang geologist about the accuracy of the model.

Re: Interpolation techniques

PeterMelle — Fri, 04 Mar 2011 18:39:28 GMT

If I am understanding this correctly, you mean creating groundwater elevation models for smaller units for the layer I am working in and then combining them together as a whole. If that is correct, I have considered that possibility, but am unsure as to how to proceed in that direction. regarding "breaklines", I could separate points on the map that are within watershed sub basins or elevations to attempt this, however I am unsure if this would be reflective of the aquifers. Any suggestions?