Estimating the volume of coarse woody debris of forest floor using sampling with probability proportional methods

Abstract

A variety of new methods have been proposed by ecologists and biologists for data collection on different attributes of coarse woody debris (CWD). Amongst others methods based on probability proportional to size are the newest techniques. These approaches are based on selecting individual logs into the sample unite using some forms of unequal probability sampling. In the present study, we aimed to introduce “Distance Limited Sampling” (DLS) and “Sausage protocol” of “Fixed Area Method” as well as assessing their precision and accuracy for estimating volume of CWD. These methods select individual logs into the samples using probability proportional to length of the logs. To collect the data, a set of attributes including location, top and bottom diameter and length of all logs in a 250×250 m plot in an un-harvested temperate forest were fully inventoried, and the location of logs were mapped. Then different methods with different sample unit sizes were simulated in R programming language environment using SampSurf package. Results indicated that both methods are unbiased. Considering precision of the estimates, Sausage protocol of fixed area sampling led to more precise estimates of volume of the loges than those estimated by DLS. Consequently, Sausage protocol is concluded for further usage as a promising method in the field due to its unbiasedness and production of highly-precise estimates.

Keywords

• Bias
• Sausage protocol
• fixed area sampling
• distance limited sampling