Published online by Cambridge University Press: 12 June 2017
This work was initiated to develop a method of sampling quackgrass populations. Shoot and bud distribution of 36 field populations was obtained. The results demonstrated that shoot count was inappropriate in assessing quackgrass bud populations and that bud sampling depth should be at least 20 cm if an accurate estimate is required. When shoot count is the only possibility, quadrat size should be as large as possible to minimize the error. Descriptive statistics of the field populations were used to generate a series of 24 simulated quackgrass bud populations. The field and simulated data were computer sampled with a program designed to compare the effects of varying sample size (number of quadrats) and sample shape (quadrat size and form). The intensity of sampling must be a function of resources and cost, since sampling precision increased linearly with sample size. Precision of sampling was generally higher when a large number of small quadrats were used compared to a small number of large quadrats. Sampling intensity per unit area must be increased when bud density increases to maintain a given degree of precision, but it can be decreased when plot size increases.