Latest Contributions by EricKrause

If you're comparing ordinary cokriging to universal kriging, you aren't getting a fair comparison.  You need to compare the predictions using the same type of kriging. Restricted maximum likelihood is the most accurate method for determining variography parameters; however, it doesn't scale well.  For large datasets, the method quickly becomes computationally infeasible.  I don't know what algorithm SAS uses, but if they really are using REML, it will take an incredibly long time to process when there are thousands or millions of points.  Our weighted least-squares algorithm, however, is able to efficiently handle datasets into the billions. ... View more

The cokriging equations are well-known, and you can reference them in Cressie (1993), for example.  Any differences between SAS and Geostatistical Analyst are almost certainly due to different variogram parameters, different search neighborhoods, and/or different preprocessing techniques (transformations, detrending, declustering, for example). What criteria are you using to say that SAS is more accurate?  Are you using the default model and comparing crossvalidation results? ... View more

GGS is limited to Simple Kriging layers. ... View more

Thanks for the paper.  I'll give it a read.  You're correct that we don't have a way of incorporating heterogeneous measurement error directly into the kriging.  The only other thing I can suggest is to condition on a subset of the data and use large number of simulations.  But even then, GGS has a limit on the size of the raster it can output, and a 30m grid on a 500km x 500km area is well beyond that limit. ... View more

Bill is right.  Interpolators are good at, well, interpolating.  Unless you have an expert understanding of the physics involved with the process (and build your model around the physics), extrapolated predictions will be unreliable. You can force the interpolators to make predictions as far outside the data extent as you want, but this is generally a very bad idea.  If you build the model correctly, you can trust the predictions within the data extent, but you should be very careful about any conclusions you draw outside that extent. ... View more

After looking at your data, I don't think there's much you can do.  The biggest problems are that your data shows no spatial autocorrelation, and you only have 26 points.  The data is also heavily skewed with at least one outlier, and the data is zero-inflated. Sorry to say, but no interpolation method is going to work with this data. ... View more

You probably want to apply a Normal Score transformation.  In the Geostatistical Wizard, use Simple Kriging, and the default is a normal score transformation.  Try changing the approximation method to Gaussian kernels. ... View more

We can do this with the Gaussian Geostatistical Simulations tool. First, create a kriging model in the Geostatistical Wizard, and don't worry about the measurement error.  Then, use this kriging layer as input to the GGS tool.  Condition on the field you want to interpolate, and specify the conditioning measurement error field.  Do a few dozen simulations, and save the MEAN raster.  This workflow will result in a raster that takes the non-constant measurement error into account. Let me know if you need any clarifications or have any questions. ... View more

Even if it only happens with one dataset, we'd still like to know.  If it's happening with one, it will happen with others. ... View more

We have Win 7 and XP machines.  I tested both and didn't get the crash. ... View more

Yeah, I thought of that a bit after I posted.  It's not a perfect solution, but I don't think there's a better way to do it with ArcGIS tools. And remember that kriging doesn't require spatially balanced points for the math to work correctly.  But the closer the points are to being spatially balanced, the lower the standard errors.  So I don't think you'll be losing much using the technique I suggested. ... View more

This functionality isn't built into the tool, but you can do it pretty easily.  Let's say that you want 100 points total, 50 points in each strata.  First, use the Create Spatially Balanced Points tool with your binary raster, and create 50 points.  Then, use the Raster Calculator in Spatial Analyst to switch the 0's and 1's of your raster; the code will be (1 - "value").  Then create 50 more points with the new raster.  You'll now have two point feature classes, each with 50 points, and they'll each be within a different strata.  You can then combine them into a single feature class using the Append tool. Let me know if you have any questions. ... View more

I can't seem to reproduce any of your crashes.  Can you send your data to ekrause@esri.com?  It may be something in your particular data.  Thanks. ... View more

You definitely want to take a subset of your data.  The Trend Analysis tool is not really designed to be used with so many points; it will just look like a huge blob, and it will be very difficult to actually see any trend.  Also keep in mind that when you rotate the graph, it's trying to dynamically move hundreds of thousands of points; you will definitely get lags and delays.  In geostatistics, several hundred points is more than enough for building the interpolation model.  Giving so many points is unnecessary when looking at any of the ESDA tools. ... View more

I just sent the workflow. ... View more