# What's new in ArcGIS Geostatistical Analyst 10.1

The ArcGIS Geostatistical Analyst extension provides a broad range of powerful spatial modeling and analysis capabilities. With version 10.1, the Geostatistical Analyst extension provides two new interpolation methods, a new transformation, and two new ArcPy classes.

## New interpolation methods in the ArcGIS Geostatistical Analyst extension

There are two new interpolation methods for Geostatistical Analyst extension in ArcGIS 10.1.

### Areal interpolation

Areal interpolation is a geostatistical interpolation technique, available in the Geostatistical Wizard, that extends kriging theory to data averaged or aggregated over polygons. Other kriging methods are only valid for continuous Gaussian data, but areal interpolation additionally accepts counts or probabilities. In conjunction with the Areal Interpolation Layer To Polygons tool, predictions can be made from one set of polygons to another set of polygons, such as predicting obesity rates in census blocks from known obesity rates in school zones. The smooth prediction surface is created in the Geostatistical Wizard, and the reaggregation to new polygons is done with the geoprocessing tool, as shown in the graphic below.

 Polygon-to-polygon predictions using areal interpolation

### Empirical Bayesian Kriging

Empirical Bayesian Kriging is a kriging method that uses repeated simulations to account for the error introduced by estimating the semivariogram. Because the method does not require interactive semivariogram modeling, it is offered through the Empirical Bayesian Kriging tool and in the Geostatistical Wizard.

 Empirical Bayesian Kriging

## New normal score transformation

The new Multiplicative Skewing approximation method for normal score transformation has been included for version 10.1. It is now the default transformation for simple kriging, and it comes with a choice of five base distributions: Student's t, Lognormal, Gamma, Empirical, and Log Empirical. Using Lognormal, Gamma, or Log Empirical base distributions guarantees that predictions will never be negative, which is often appropriate for environment variables such as rainfall.

## New ArcPy classes

Because Empirical Bayesian Kriging does not support elliptical search neighborhoods, two new ArcPy classes have been added for version 10.1. The first is a standard circular neighborhood, and the second is a smooth circular neighborhood.

## New default kriging method

Simple kriging is now the default kriging method; in previous versions, the default was Ordinary kriging. The change was made because of the flexibility of the new Multiplicative Skewing normal score transformation.