Evaluation of two sampling methods for even-age Pinus forest in northern Mexico

Abstract

The objective was to evaluate and compare the precision of estimation of dasometric variables, survey times and costs of two sampling methods versus a census, applied to a natural even-age mass of Pinus arizonica Engelm. in the municipality of Guachochi, Chihuahua, Mexico. Nine fixed-dimension and nine variable-dimension sites were performed with Bitterlich's angular sampling at coincident sample points. Times from start to finish were taken by site and technicians from the State were surveyed to determine sampling costs. ANOVA tests were developed at a significance level of 0.05, comparing both methods with the census. The results indicate that, in both samplings, the number of trees per hectare does not present statistical differences with respect to the census. The basal area by both methods was not statistically different, however, fixed dimensions present an error greater than 5%. Finally, the volume per hectare estimated was similar to that of the census, according to the rates carried out, greater precision was found by the variable area method (error=0.03%). The execution time showed significant differences (p=0.0001), the fixed dimension site required a mean time of 44 minutes and the variable dimension one 10 minutes 26 seconds. The total costs are 679.90 and 654.33 Euro to sample nine fixed-dimension and nine variable-dimension sites respectively, showing significant differences (p=0.0008). It is concluded that both methods are statistically acceptable for the variables evaluated in the type of forest mass studied.