Performing Spatial Interpolation Using ArcGIS 10

Description

For some applications, it is impossible to collect data for every point in an area of interest. Using ArcGIS Geostatistical Analyst tools and sample data, however, you can predict, or interpolate, values at every location across a surface. This course focuses on the kriging geostatistical interpolation method and teaches how to create a kriging model that is optimized for your sample data. Many factors impact the reliability of your interpolation results. You will learn how to use the kriging model to produce a statistically valid prediction surface, generate confidence in your analysis results, and improve decision making.

Who Should Attend

GIS analysts and specialists, students, and researchers and professionals working in industries such as public health, meteorology, precision agriculture, environmental management, mining, and energy production.

Learn How To

• Understand the data assumptions inherent to the kriging interpolation model and find out if your sample data matches those assumptions.
• Define an optimal semivariogram model for your data in preparation for kriging.
• Apply techniques to modify your sample dataset so that it meets kriging model criteria.
• Cross-validate a prediction surface to see how predicted values compare with known values for the location.
• Analyze cross-validation statistics to determine which kriging model is most suitable for your data and assess the accuracy of your prediction surface.
• Export a prediction surface generated by a kriging model to a raster dataset so it can be used for further visualization and analysis.

To complete course exercises, you need the following software:

• ArcGIS 10.0 for Desktop (ArcView, ArcEditor, or ArcInfo)
• ArcGIS 10.0 Geostatistical Analyst

What is geostatistical interpolation?

• Deterministic and geostatistical interpolation
• Geostatistical interpolation models
• Create your initial kriging surface
• Kriging assumptions

Kriging modeling techniques

• Binning and the semivariogram model
• Workflow for modeling the semivariogram
• Optimizing your semivariogram model
• Create your semivariogram model
• Modify your data parameters
• Defining searching neighborhoods
• Modify your remaining model parameters

Interpreting your interpolation results

• Cross validating your models
• Comparing model results
• Cross validation and comparison
• Interpreting your final surface
• Interpret your interpolation results

• Completion of ArcGIS Desktop II: Tools and Functionality or Learning ArcGIS Desktop or equivalent knowledge is required.
• Completion of Exploring Spatial Patterns in Your Data Using ArcGIS 10 or equivalent knowledge is also required.