Comparing Sampling Strategies for Estimating Age Characteristics of a Forest

• Activity Number: 619
• Type: Contributed
• Date/Time: Thursday, August 7, 2014 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistics and the Environment
• Author(s): Brent Burch*+ and Andrew J. Sanchez Meador
• Companies: Northern Arizona University and Northern Arizona University
• Keywords: Distance sampling ; Fixed plot ; Ratio estimator ; Variable radius plot

Abstract

We investigate the use of different sampling strategies to estimate the age characteristics of a forest. This includes estimating the mean tree age and estimating the age-class distribution of the forest. The sampling units are based on a grid structure for a simulated forest where tree location and tree size are known. The age-size relationship is described by a model relating age to diameter at breast height (dbh) where the variability in age increases as dbh increases. The sample itself is based on plot, point, or distance sampling associated with the underlying grid structure. Both simple random samples and systematic random samples are considered. The properties of the estimators are determined by repeated samples from the forest. The six possible estimation strategies (based on combinations of three types of sampling units and two sampling methods) are compared to one another.