|
Activity Number:
|
76
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Sunday, August 6, 2006 : 8:00 PM to 9:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #307344 |
|
Title:
|
Strip Transect Sampling To Estimate Object Abundance in Homogeneous and Nonhomogeneous Poisson Fields: a Simulation Study of the Effects of Transect Width and Number
|
|
Author(s):
|
Timothy C. Coburn*+ and Sean A. McKenna and Hirotaka Saito and Orlando T. Garcia
|
|
Companies:
|
Abilene Christian University and Sandia National Laboratories and University of California, Riverside and Sandia National Laboratories
|
|
Address:
|
ACU Box 29315, Abilene, TX, 79699,
|
|
Keywords:
|
strip transect sampling ; variance and bias estimation ; spatial Poisson fields ; unexploded ordnance (UXO)
|
|
Abstract:
|
Natural phenomena often occur as spatial point processes which can be modeled by the Poisson distribution. This paper investigates the use of strip transect sampling to estimate object intensity/abundance in such situations. Both homogeneous and non-homogeneous fields are considered. Because the number and size of transects have important operational and economic implications, the objective is to determine how many and what size transects should be selected in order to satisfy specified precision and bias constraints. The results suggest that percent of field coverage is more important than specific combinations of transect size and number, and that sampling requirements to produce high quality estimates are higher than anticipated. The sampling requirements are highest for non-homogeneous fields containing clusters or targets. A specific application to unexploded ordnance is discussed.
|
