An analysis of the adaptive cluster sampling design with rare plant point distributions

Abstract

A sampling design that can provide estimates of abundance with low variance is very valuable to biologists working with limited budgets and time. Estimates that are precise even with low sampling efforts allow researchers to cheaply and confidently monitor rare populations. Adaptive cluster sampling has the potential to be much more efficient at sampling rare populations than conventional sampling designs, but it has also been shown to be highly inappropriate for some populations. Applications of adaptive cluster sampling (ACS) have had inconsistent results in real-world settings, leading to increasing scrutiny of the factors that influence the efficiency of this design. Much more work still needs to be done in order to provide samplers with the knowledge of when ACS is appropriate and how to maximize its effectiveness through constructing an optimal design. This study develops a procedure in a GIS environment for rigorously examining the effects of design parameters on the variance of ACS estimates, and applies this procedure to some real-world point populations. The relative efficiency of adaptive cluster sampling to simple random sampling is shown to be dramatically influenced by design parameters. This highlights the need for further investigation and a better understanding of how these parameters interact with point distributions through the use of procedures and tools such as those introduced here.

Cite this paper

@phdthesis{tout2009analysis,
  title = {{An analysis of the adaptive cluster sampling design with rare plant point distributions}},
  author = {Tout, Jeremy},
  school = {CSUN ScholarWorks (California State University, Northridge)},
  year = {2009}
}

Made from the index record. Check names, capitalisation and page numbers against the paper. Need another style? AI for Research has over 10,000.

Related papers