Comparing Sampling Designs for Monitoring Ecological Status and Trends: Impact ofTemporal Patterns (Chapter 3)

摘要

The paper presents a comparison of two classes of sampling designs suggested forlong-term environmental monitoring. One class is based on a rotating panel design that prescribes visiting a panel of sites for several consecutive years with partial replacement of sites each year; the other class is based on a serially alternating design which prescribes returning to sampling sites every four years. A random effects linear model is used to investigate and compare the proposed sampling designs because the sites to be visited and the years of visits may manifest substantial random contributions. The class of designs based on the serially alternating design lead to more precise estimates of current status and of trends in the ecological resource when the site components of variance is substantially greater than the year component and/or when the sample sizes are smaller.