Efficiency Of Adaptive Cluster Sampling for Estimating Density of Wintering Waterfowl

An evaluation of adaptive cluster sampling was based on a simulation
experiment where samples were drawn from an enumeration of three
species of waterfowl wintering in central Florida. The initial
samples were taken either by simple random sampling or with
probability proportional to available habitat. Efficiency of adaptive
cluster sampling relative to simple random sampling was highest when
1) the within-network variance was close to the population variance,
and 2) the final sampling fraction was close to the initial sampling
fraction. The within-network variance is determined by the spatial
distribution of the population, quadrat size, and the condition that
determined when to adapt sampling. The final sampling fraction
depends on the previous factors as well as the size and selection of
the initial sample. Some combinations of these factors led to
increased precision compared to simple random sampling and some did
not.

Geographic Description:
Central Florida (5,000 km2). The study region extended 100 km east
and 50 km north from the southwest corner at 0438000, 3056000
(Universal Transverse Mercator coordinates; zone 17). 1.5.2 Bounding
Rectangle Coordinates

Methodology:
An effort was made to count every individual duck of the three
waterfowl species in a 5,000 km2 area of central Florida by making
systematic flights over the entire study region. Two biologists
counted waterfowl from separate helicopters (Bell Jet Rangers) during
13-15 December, 1992 and used the LORAN-C and GPS systems to determine
flock locations Field.