AnsweredAssumed Answered

Create Geostatistical Layer and output both Prediction and SE in Model Builder

Question asked by cotton.rockwood on Oct 6, 2016
Latest reply on Oct 6, 2016 by cotton.rockwood


I have built a model in Model Builder that iterates over point Features in a file geodatabase, selects a subset of each point feature based on attributes and then uses the 'Create Geostatistical Layer' tool to implement Ordinary kriging with optimization via crossvalidation.  I am using an xml file as a template.  I attached a screenshot of the model for reference.

I would like the output to be 1) a raster of predictions and 2) a raster of Standard Errors.  I have used the 'GA Layer To Grid' tool to output the prediction raster.  However, the type of output from the 'Create Geostatistical Layer' tool is specified in the xml file.  Do I need to re-run the tool with a new xml template in order to output the SE raster or is there a simpler way to do this. (This is simple to do when a GA Layer is created with Geostatistical Analyst by right clicking the layer and selecting 'Change output of Standard Error').

I am using ArcGIS 10.3.1 for Desktop with the Advanced license on Windows 10.  I also have 64-bit Background Geoprocessing installed in case that is relevant.

I used some of the information from this (Automating geostatistical interpolation using template layers | ArcGIS Blog) post by Eric Krause and this (Need Help Automation of Kriging using Model builder or python - Reg. ) to get started.