PROJECT MENU:

RATIONALE FOR THE STUDY

(This Page)

INTRODUCTION:

Data Properties

Summary of the Previous Study

STATISTICAL ANALYSIS:

Test for a Normal Distribution

Summary of Interpolation Methods:

    Statistical Methods

    Geostatistical Methods

RESULTS OF THIS STUDY

DISCUSSION

CONCLUSIONS

REFERENCES

APPENDICES

    A Kriging Example

   1-Hour Prediction Maps

MORE ON FOREST FIRE HAZARD MODELING

(Coming Soon)

FIGURE 1. Location map for the Study area MESA del HURRACAN

Predicting Forest Fire Fuel Loads using the Geostatistical Analyst Extension of ArcGIS 9.3

Goal of this Project:  To compare the accuracy and precision of the ArcGIS Geostatistical Analyst 9.3 to a previous spatial statistics study by Flores-Garnica (2001), that utilized the interpolation methods available in ArcView 3.X, GS+ and S-Plus spatial statistics software.  This study will apply Inverse Distance Weighted, Spline, Ordinary Kriging, and Universal Kriging interpolation methods.

Rationale:  Most spatial statistics studies apply either the software options mentioned above (GS+ and S-Plus), the Geostatistical Library (GSLIB) open source software from Stanford University, or increasingly the R open source statistical software from the University of Auckland, New Zealand.  Open source software, although free, often has the problem of poor documentation. The wide availability of ESRI products with documentation and support should make Geostatistical Analyst a better option for organizations that already use ArcGIS 9.X.

ArcGIS Extensions: Geostatistical Analyst Wizard, Spatial Analyst, and other Toolbox extensions were applied as warranted.

ACKNOWLEDGMENTS:

This project could not have been completed without the data and analytical guidance provided by:

Dr Rafael Moreno-Sanchez

Geography and Environmental Sciences

University of Colorado-Denver

Dr Jose German Flores-Garnica

Instituto Nacional de Investigaciones Forestales,

Agricolas y Pecuarias  (INIFAP), Mexico