SAMPLING GUIDELINES FOR OLD-GROWTH FORESTS IN THE MIDWEST, USA

Old-growth forests in the Midwest occur in tracts ranging from a few t o a few hundred hectares in size. As more resources are directed towar d understanding the unique features of these old forests, there will b e new opportunities to inventory the vegetation on these areas and to compare the observed composition and structure with that of other old- growth and second-growth forests. Such comparisons are meaningful only if the sampling procedures are sufficiently sensitive to detect diffe rences that are of practical importance. Based on detailed inventories of five upland old-growth tracts in Missouri, Indiana, and Illinois, we examine the relation between sampling intensity in old forests and precision of estimates for (1) measures of stand density, (2) coarse w oody debris, and (3) species richness/diversity. Sample sizes necessar y to estimates mean basal area and number of trees per hectare within +/- 10% of the mean can often be accomplished with 30 or fewer 0.1-ha plots. Estimating the density of dead trees, the volume of coarse wood y debris, or the density of trees by individual species groups within +/- 10% of the mean can require several hundred 0.1-ha plots. Effects of plot size are discussed, and some practical aspects of inventorying midwestern old-growth forests are addressed.