Scirtothrips dorsalis, chilli thrips, is an invasive insect species in Florida and an important foliar pest of blueberry. Sound knowledge of insect distribution within the field is needed to formulate accurate sampling methods. Fourteen blueberry fields were systematically sampled for chilli thrips during the summers of 2017 and 2018. Field counts were modeled in various spatial models and determined chilli thrips had temporally stable aggregated distribution. A fixed-precision sampling plan was developed for summer sampling, requiring seven and three sampling units (sampling unit = 10 young blueberry shoots) to estimate a nominal mean density of 20 chilli thrips with a precision of 25% and 40%, respectively. The sampling plan can be used to improve the timing of control measures and assess the effectiveness of these control measures.
Scirtothrips dorsalis Hood is an invasive and foliar pest of Florida blueberry that reduces plant growth by feeding on new leaf growth. A sampling plan is needed to make informed control decisions for S. dorsalis in blueberry. Fourteen blueberry fields in central Florida were surveyed in 2017 and 2018 after summer pruning to determine the spatial and temporal distribution of S. dorsalis and to develop a fixed-precision sampling plan. A sampling unit of ten blueberry shoots (with four to five leaves each) was collected from one blueberry bush at each point along a 40 × 40 m grid. Field counts of S. dorsalis varied largely ranging from zero to 1122 adults and larvae per sampling unit. Scirtothrips dorsalis had aggregated distribution that was consistent within fields and temporally stable between summers, according to Taylor’s power law (TPL) (aggregation parameter, b = 1.57), probability distributions (56 out of 70 sampling occasions fit the negative binomial distribution), Lloyd’s index ( b > 1 in 94% occasions), and Spatial Analysis by Distance IndicEs (31% had significant clusters). The newly developed fixed-precision sampling plan required 167, 42, seven, or three sampling units to estimate a nominal mean density of 20 S. dorsalis per sampling unit with a precision of 5%, 10%, 25%, or 40%, respectively. New knowledge on S. dorsalis distribution will aid in evaluating the timing and effectiveness of control measures.