Optimizing sample unit size for sampling stink bugs in Midwest soybean

Stink bugs are an increasing concern to soybean, Glycine max (L.) Merrill. Sampling plans have been recently developed for herbivorous stink bugs of soybean. Sample unit size is an important component of sampling plans because it can affect the precision and costs. Here, two sample unit sizes (10 vs. 25 sweeps) were compared for the sampling of Euschistus spp. On a weekly basis, a total of 21 soybean fields were sampled with sweep nets in Illinois, Indiana, Michigan, Minnesota, Nebraska, North Dakota, Ohio and Wisconsin during 2018. On each sample date for each field, 15 sets of 25 sweeps and 15 sets of 10 sweeps were collected. The probability of detecting at least one insect, the mean number of insects per sweep, standard error of the mean, relative variance, relative net precision, and the spatial pattern (using Taylor’s power law) were used to compare these two sample unit sizes for Euschistus spp. nymphs, adults, and nymphs + adults. Overall, the probability of detection was higher for 25 sweeps. The mean number of insects per sweep was similar between the two sample unit sizes. The standard error and relative variance were lower for 25 sweeps. However, the relative net precision was higher for 10 sweeps. In conclusion, sampling plans using the 25-sweep sample unit size remain effective, but a 10-sweep sample unit size would be more efficient. Furthermore, the optimization of sample unit sizes is an important factor that should be accounted for in the development of sampling plans.