Six methods of estimating pest-caused tree mortality were evaluated by computer-based sampling of the aerial photo and numerical database of a 9,320 ha area of ponderosa pine (Pinus ponderosa) in the central Sierra Nevada, California. Two sampling universes were defined, one comprising 9 X 9 in photo units, the other 70 mm photo units. Of the sampling designs simulated, 2-stage designs involving subsampling of photo strips with either equal or variable probability provided estimates with the lowest standard deviations, and hence would require smaller sample sizes (and less photo-interpretation time) for a specified precision. The findings of this study have application to operational surveys of pest damage for which complete aerial photo coverage is available, or to the design of surveys for which only sample aerial photo coverage is feasible. West J. Appl. For. 1:84-89, July 1986.
Document Type: Journal Article
Biostatistician-programmer, Environmental Health Associates, Oakland, CA
Publication date: July 1, 1986
More about this publication?
Each regional journal of applied forestry focuses on research, practice, and techniques targeted to foresters and allied professionals in specific regions of the United States and Canada. The Western Journal of Applied Forestry covers the western United States, including Alaska, and western Canada; WJAF will also consider manuscripts reporting research in northern Mexico that has potential application in the southwestern United States.