PARENT SESSION
Oral Session #84: Statistical Ecology.
Presiding: L. Subhash
Thursday, August 8. 9:00 AM to 11:30 AM. Mesquite Room, Radisson.
Line intersect sampling: unbiased estimators for ell-shaped intersect lines.
GREGOIRE, TIMOTHY1, VALENTINE, HARRY*,2, 1 Yale University, School of Forestry and Environmental Studies, New Haven, CT2 USDA Forest Service, Northeastern Research Station, Durham, NH
ABSTRACT- Line intersect sampling is commonly used to by ecologists to estimate plant abundance and cover. It is also used to estimate coarse woody debris and forest fuel loads. Some of the earliest papers on line intersect sampling contained recommendations for running two intersect lines, the second perpendicular to the first. It was thought that this strategy would eliminate a bias where intersected elements tended to be oriented in a common direction. Although no bias accrues under a design-based approach to sampling with straight intersect lines, the use of ell-shaped intersect lines evolved from the early recommendation. Samplers have assumed that estimators, which are unbiased for straight intersect lines, also are unbiased for ell-shaped lines. However, this assumption is false for the so-called conditional estimators. Thus, a strategy to eliminate a non-existent bias may actually induce bias because the estimator is inappropriate. We have derived new design-unbiased estimators for line intersect sampling with ell-shaped lines. The new conditional estimator requires a measurement of an element's length perpendicular to each leg of the intersect line, regardless of which leg intersects the element. The unconditional estimator requires a measurement of the circumference of the convex hull enclosing the intersected element. Our approach extends to the formulation of estimators for triangular and Y-shaped intersect lines.
KEY WORDS: conditional estimator, unconditional estimator