Sampling to determine density of arthropods in intensively grazed grasslands

• Published in: Journal of applied entomology (1986) 2020-07, Vol.144 (6), p.519-533
• Main Authors: McNeill, Mark Richard, Koten, Chikako
• Format: Article
• Language: English
• Subjects: Adults; Aquatic insects; Arthropods; Beetles; Dry matter; Electric appliances; Grasslands; G‐vac; Heat; Heat transfer; Heat treatment; invertebrates; Measurement methods; Methods; Pasture; Population density; Population dynamics; Population studies; Sampling; sampling efficiency; Sampling methods; Species diversity; Spiders; Turf; turves; Vacuum; Vacuum cleaners; vacuum sampling; Vortis
• ISSN: 0931-2048; 1439-0418
• DOI: 10.1111/jen.12754

Arthropod population field studies undertaken to understand the impacts and population dynamics of the target organism rely on sampling methods that provide accurate measurements of population density. Unsuitable methods may underestimate or provide widely variable measures of population density. The suitability of three vacuum sampling methods: a domestic vacuum cleaner, Vortis™, and blower‐vac (G‐vac), plus heat extraction of turves were compared for sampling common species in an intensively grazed irrigated dairy pasture. Each method used a different approach to sampling with the number of samples per method based on protocols that had been used in grasslands. Overall, for adults of the weevil Listronotus bonariensis, no method provided a consistently high mean density, while for adults of the weevil Sitona obsoletus, the vacuum cleaner and turf heat extraction methods gave generally higher mean densities. For predatory beetles and lacewings, heat extraction was the most effective for density measurements, while for ladybirds and spiders, the Vortis™ provided the best estimate of mean density. Increasing pasture dry matter (kg Dm/ha) generally had a significant negative effect on S. obsoletus density but not for L. bonariensis. While no method was consistently superior than another, the vacuum cleaner and heat extraction methods generally provided higher mean densities. The G‐vac generally produced the lowest density estimates, but the sampling protocol meant a larger collection area per unit effort and therefore the probability of detection was higher compared with the three other methods. In conclusion, the study showed that the optimal sampling method depends on the taxa targeted, and more than one method may be required to measure density and diversity of species in both natural and modified grasslands.