Restricted Access for Students in:
Lectures/Labs SAL

Geospatial Pattern Analysis and Geostatistics

### Ocean 453/503

 WHEN:  Mon. & Wed. 11:00 - 12:20 WHERE:   Ocean Science Building, Room 111   The Spatial Analysis Lab. WHO:     Dr. Miles G. Logsdon; mlog@u.washington.edu

Course Overview:

This course is focused on the Application of Geospatial pattern analysis and the application of geostatistics in Earth Science Research.  The course is ideally suited for those students who seek to develop more in-depth skills and an advanced understanding of the concepts behind modern tools for detecting, describing, and estimating spatial patterns and trends.  The course specifically addresses ways of investigating the spatial continuity that is an essential feature of natural phenomena.  This course will explore:

• Descriptive Statistics of spatial pattern,
• Estimation & Prediction
• Interpreting the Variogram & Fitting a Model, and
• Quantitative Pattern Metrics of Composition and Configuration

This is a 3 credit course.  Grades are based upon two "take home" test, and a project design by the student.

"An Introduction to Applied Geostatistics" by Edward H. Isaaks and R. Mohan Srivastava, Oxford University Press, 1989.

"Spatial Data Analysis: Theroy and Practice" by Robert Haining, Cambridge University Press, 1993.

"Statistics for Spatial data" by Noel a. c. Cressie, Wiley & Sons, Inc. 1991.

Descriptive Statistics and Semivariance
• Explore data formats, define continuous vs. discrete, integer vs. floating point
• Review the histogram: mean, mode, median, st dev.

Spatial Description
• Introduction to autocorrelation, covariance, and semi-variance
• Moran-I demostartion
• Define Regionalized variables

Spatial Pattern
• Discuss spatial dependence, spatial correlation
• Introduce aniso vs. isotrophy, heteroscedasticity,

Introduction to ArcGIS Geostatistical Toolbox: Terms & Tools

• Exploring the ArcGIS geostatistical analyst and Fragstats.
• Empirical Semivariogram
• Input data, view histogram, and run demo geo-stat Wizard using defaults?
• Prepare data for analysis – degenerated dataset examples & considerations
• Kriging Models
• Construct and interpret a variogram
• Explore fitting a model & predicting surfaces
• Discuss Trend, QQ Plot, Voronoi Map, Cross Covariance Cloud
• Student demostrations

Pattern Analysis via pattern Metrics

• Introduce Pattern Metrics
• concepts of landscape ecology -
• Describing spatial heterogeneity
• Discuss Composition vs Configuration
• Choosing appropriate metrics
• Short demo of Fragstats
• Review results
• Calculating metrics on individual datasets using fragstats
• Describing spatial patterns and relating to processes

Estimation and Prediction: Building and validating surfaces

• Comparing predicted surfaces
• Interpreting and validating the predictions
• Assessing accuracy

Other information

All participants should be familiar with the tools found in commercial GIS software.