||Properties of Designs for Sampling Continuous Spatial Resources.
Stehman, S. V. ;
Overton, W. S. ;
||SUNY-ESF, Syracuse, NY. ;Oregon State Univ., Corvallis.;Corvallis Environmental Research Lab., OR.
||EPA-R-815422 ;EPA-R-816721; EPA/600/A-94/075;
Mathematical models ;
Environmental Monitoring and Assessment Program ;
||Most EPA libraries have a fiche copy filed under the call number shown. Check with individual libraries about paper copy.
Three designs were investigated for sampling continuous response variables defined over a continuous spatial region. Two of the designs were based on the triangular grid proposed for sampling environmental resources in the Environmental Monitoring and Assessment Program. Analytical and empirical results were derived under various surface models used to represent the continuous response variable. The tessellation-stratified design was generally more precise and permitted better estimation of variance than either a strict systematic design or an unrestricted random sample.