Suter, Ann . 2002. Estimating Methods, Variability, and Sampling for Drop-Test Data. 0257 2826. Missoula, MT: U.S. Department of Agriculture, Forest Service, National Technology and Development Program. 30 p.
Discusses the testing process the Forest Service has used for the past six decades to analyze the ground patterns made by aerial drops of fire retardants or suppressants. The process involves dropping firefighting chemicals from an airtanker flying over open cups arranged in a regularly spaced grid. This report uses data collected from six airtanker drops to investigate estimation methods, variability, and sampling. Five estimation methods were compared: triangulation, ordinary kriging, polygonal declustering, inverse distance squared, and local sample mean. Cross validation showed that triangulation and ordinary kriging were the two best estimation methods for drop-test data. Replicate drops should be made when-ever investigators need to know whether differences in line length are due to changes in factor levels or whether they are just a reflection of the inherent variability in the test. Investigation of the sampling scheme shows that increasing the spacing of the cups reduces the accuracy of the estimates. In the crossrange direction (perpendicular to the flight path), a 10-foot spacing is recommended. In the downrange direction (in the direction of the flight path), spacing could be increased slightly from the present 20 feet without seriously affecting the accuracy of the estimates.
Keywords: airtankers, coverage levels, cross validation, ground pattern, history, kriging, line length, sampling, spatial statistics, triangulation, variance
View HTML
These files are best for printing:
View high resolution PDF (1492 kb)
If you experience problems using the higher resolution PDF file, use the lower resolution PDF files if they are listed.
View 72 dpi PDF Part 1 (272 kb)
View 72 dpi PDF Part 2 (679 kb)
View 72 dpi PDF Part 3 (296 kb)
Click the icon to download a free version of Adobe Acrobat Reader.
