Select to view content in your preferred language

semivariogram on grid data

2673
4
12-22-2011 05:48 PM
porntipphontusang
New Contributor
Hi

I'm new about Geostatistics. I have some problem about semivariogram. I collected soil in grid sampling and I use semivariogram to analyse with the data. The problem is when semivariogram plot it looks strange there are many point in  1 distance (figure).Is it possible to fit or not? what does it mean from this semivariogram? anybody can help me ? thank you
0 Kudos
4 Replies
EricKrause
Esri Regular Contributor
Kriging can work fine on data sampled on a grid. 

However, your semivariogram does look a bit strange.  The biggest problem I see is that the semivariogram is basically flat, meaning that there is little spatial correlation in the data.  If the data is not spatially correlated, it does not make sense to interpolate at all.

Try using this geoprocessing tool on your data:
http://webhelp.esri.com/arcgiSDEsktop/9.3/index.cfm?TopicName=Spatial_Autocorrelation_%28Morans_I%29...

Find the p-value from the output.  The lower the p-value, the more spatially correlated your data is.  You'd like to see something below 0.05.
0 Kudos
JeffreyEvans
Frequent Contributor
It does not look like you have any spatial structure in your data. As such, there is little you can say about the spatial process and interpolation is not supported. Perhaps your sampling distance is too coarse and you missed the spatial process you were after. This is why it is so important to apply a sample design before you start collecting data. You may want to consider collecting more data and sampling in a different grid pattern. I have found that hexagonal sampling works quite well for things like soil compaction. Based on the semivariogram, your data looks somewhat categorical. Is this the case? If so you need to read up on methods that support discrete data (not kriging).     

Given your current sample you can model the expected semivariance and design a secondary sample, that does capture the spatial structure, to augment your current data. There are a some very good papers on dual-phase sampling that may allow you to salvage your study. A good starting point is:

Delmelle E. and P. Goovaerts (2009). Second-Phase Sampling Designs for Non- Stationary Spatial Variables. Geoderma, vol. 153: 205-216.
0 Kudos
EricKrause
Esri Regular Contributor
Based on the semivariogram, your data looks somewhat categorical.


I think this is just due to the small major range (look at the semivariogram surface on the bottom-left).  I would suggest increasing the range, but if there is no spatial correlation, increasing the range won't fix the underlying problem.
0 Kudos
porntipphontusang
New Contributor
Thank you for your all suggestion.It's very a big help. ^^
0 Kudos