Why the interpolation result of Spline with Tension in the Spatial Analyst is different from the that in the Geostatistical Analyst (both in ArcGIS 10), including their minimum values, maximum values, and isopach?

Just as the maps below, with the same original date and corresponding parameter setting, fig. 1 is done by Spline with Tension in Spatial Analyst, fig. 2 is done by Spline in Geostatistical Analyst, and they are totally different.

What does the prediction map in Geostatistical Analyst exactly mean?

As we know, Spline is an exact interpolation method, but in Geostatistical Analyst, it seems not. Just as the yellow dots in the fig. 2, the original value of them are all 0, but why they appear in different color bands in the prediction map, which means that their predicted value are different with their original value.

Why? I'm really confused about that.

Some further explanations:

  • the yellow and red dots=0, and the values of the green dots are labeled in the figs, all of them are used to do the interpolation.
  • Fig. 1 is understandable to me, which conforms with definition of Spline, and all the dots appear in the correct color areas that include their original data.
  • I cann't understand fig. 2.

Why does the same interpolation method in different tools lead to very different results?

Forgive my poor English.

enter image description here

  1. interpolation result of Spline with Tension in the Spatial Analyst

enter image description here

  1. Prediction Map in the Geostatistical Analyst

closed as too broad by PolyGeo Oct 3 '16 at 6:35

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. 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.

  • 1
    Answer to this question is not very simple but I'll refer to read this link webapps.fundp.ac.be/geotp/SIG/interpolating.pdf I hope you find usefull – Predator X Dec 19 '15 at 7:35
  • 1
    Which points are the data here? The labelled green ones, the red and yellow coloured ones? All of them (with yellow=zero, red=what)? Applying the same colour mapping to the two smoothings would help assess the difference also. The first one is stretched by that huge red blob of a peak at the top... – Spacedman Dec 19 '15 at 16:39
  • 1
    This is a very complex answer, hopefully it is on the board already if not maybe Bill will jump in, if not Ill try to get to it in the next day or two. – If you do not know- just GIS Dec 19 '15 at 23:45
  • As per the Tour there should be only one question asked per question. – PolyGeo Sep 16 '16 at 23:37
  • 1
    @AndreSilva It is too broad due to there being two questions. If the question focused on just one then it'd possibly be OK – Midavalo Sep 19 '16 at 18:41

You can find Comparing interpolation methods in below link. [link]http://pro.arcgis.com/en/pro-app/tool-reference/3d-analyst/comparing-interpolation-methods.html I hope this figure and below description can be helpfully [Comparing figure for IDW, Kriging and Spline] [enter image description here]1 IDW and Spline are two deterministic methods that create surfaces from samples based on the extent of similarity or degree of smoothing. However, while a spline surface passes exactly through each sample point, an IDW will pass through none of the points. Kriging is a geostatistical method that uses a powerful statistical technique for predicting values derived from the measure of relationship in samples and employs sophisticated weighted average techniques.

  • 1
    Thank you. But what I want to know is that the same interpolation method performs quite different in different analyst tools. Your link doesn't say something about that. – qing Dec 19 '15 at 14:31
  • Did you see my link in comment also? – Predator X Dec 19 '15 at 15:41
  • 1
    I think you misunderstand my question. I used the same interpolation method (in different tools), but got different results. – qing Dec 21 '15 at 2:54
  • I see, I'll say in my opinion they aren't same, they have many differences, it back to parameter and algorithm run behind each tools. – Predator X Dec 21 '15 at 3:54

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