Kriging to estimate vegetation cover - simple, ordinary or universal?

[TessAndersson]

Hi!
I'm trying to do a prediction map of vegetation cover. I've estimated moss cover in a number of plots in a forest stand and wish to predict the cover where data hasn't been collected. I figured kriging could work, but I'm quite a rookie and if someone would point me in the right direction I would be very grateful! I've read that Kriging performed on data that have no geographic trend is called ??????simple kriging?????? if the mean is known or ??????ordinary kriging?????? if the mean is unknown but I don't fully understand the meaning. I tried to do both simple and ordinary kriging (by just clicking "next" all the time...) and got quite different results. With simple Kriging I got a bulls eye effect but not with ordinary kriging, that map was much smoother. Does anyone have an idea on what method is most suitable for my purpose?

Thanks in advance!

[EricKrause]

It's hard to answer this question without looking at your data, but your best bet is to use Empirical Bayesian Kriging if you aren't sure which type of kriging to use.  It's available in the Geostatistical Wizard and as a geoprocessing tool.

[MarcoBoeringa]

Although Eric is right that answering these questions is difficult without looking at the data, I know vegetation cover for certain species or vegetation types tends to vary erratically and discontinuously. This violates the basic assumption of continuity of the interpolated variable that interpolation methods tend to have.

Data like this may be better processed using a generic statistical package like SPSS etc. E.g. to estimate an average "moss cover" percentage beneath different types of forest, or on different soils, and see if there are statistically significant differences between classes.

If you than know that forest type A on soil B tends to have a certain "moss cover" value C, you might use that to classify an existing map of forest or soil types to get some estimate of "moss cover" in unsampled regions.