Bannerman J A, Costamagna A C, McCornack B P, Ragsdale D W
Department of Entomology, University of Manitoba, 12 Dafoe Rd, Winnipeg, MB R3L 1R2.
Department of Entomology, Kansas State University, 123 W. Waters Hall, Manhattan, KS 66506.
J Econ Entomol. 2015 Jun;108(3):1381-97. doi: 10.1093/jee/tov009. Epub 2015 Mar 21.
Generalist natural enemies play an important role in controlling soybean aphid, Aphis glycines (Hemiptera: Aphididae), in North America. Several sampling methods are used to monitor natural enemy populations in soybean, but there has been little work investigating their relative bias, precision, and efficiency. We compare five sampling methods: quadrats, whole-plant counts, sweep-netting, walking transects, and yellow sticky cards to determine the most practical methods for sampling the three most prominent species, which included Harmonia axyridis (Pallas), Coccinella septempunctata L. (Coleoptera: Coccinellidae), and Orius insidiosus (Say) (Hemiptera: Anthocoridae). We show an important time by sampling method interaction indicated by diverging community similarities within and between sampling methods as the growing season progressed. Similarly, correlations between sampling methods for the three most abundant species over multiple time periods indicated differences in relative bias between sampling methods and suggests that bias is not consistent throughout the growing season, particularly for sticky cards and whole-plant samples. Furthermore, we show that sticky cards produce strongly biased capture rates relative to the other four sampling methods. Precision and efficiency differed between sampling methods and sticky cards produced the most precise (but highly biased) results for adult natural enemies, while walking transects and whole-plant counts were the most efficient methods for detecting coccinellids and O. insidiosus, respectively. Based on bias, precision, and efficiency considerations, the most practical sampling methods for monitoring in soybean include walking transects for coccinellid detection and whole-plant counts for detection of small predators like O. insidiosus. Sweep-netting and quadrat samples are also useful for some applications, when efficiency is not paramount.
多食性天敌在北美控制大豆蚜（Aphis glycines，半翅目：蚜科）方面发挥着重要作用。几种抽样方法被用于监测大豆中的天敌种群，但很少有研究调查它们的相对偏差、精度和效率。我们比较了五种抽样方法：样方抽样、全株计数、扫网法、行走样带法和黄色粘虫板法，以确定对三种最主要物种进行抽样的最实用方法，这三种物种包括异色瓢虫（Harmonia axyridis，帕拉斯）、七星瓢虫（Coccinella septempunctata L.，鞘翅目：瓢虫科）和暗黑赤眼蜂（Orius insidiosus，赛）（半翅目：花蝽科）。随着生长季节的推进，我们发现不同抽样方法内部和之间群落相似性的差异表明了抽样方法与重要时间之间的相互作用。同样，多个时间段内三种最丰富物种的抽样方法之间的相关性表明抽样方法之间相对偏差存在差异，这表明偏差在整个生长季节并不一致，特别是对于粘虫板和全株样本。此外，我们表明，相对于其他四种抽样方法，粘虫板产生的捕获率偏差很大。抽样方法之间的精度和效率有所不同，粘虫板对成年天敌产生了最精确（但偏差很大）的结果，而行走样带法和全株计数分别是检测瓢虫和暗黑赤眼蜂最有效的方法。基于偏差、精度和效率的考虑，大豆监测中最实用的抽样方法包括用于检测瓢虫的行走样带法和用于检测像暗黑赤眼蜂这样的小型捕食者的全株计数法。当效率不是最重要时，扫网法和样方抽样在某些应用中也很有用。
