Texas A&M AgriLife Research and Department of Entomology, Texas A&M AgriLife Research and Extension Center, 10345 Hwy 44, Corpus Christi, TX 78406, USA.
J Econ Entomol. 2013 Jun;106(3):1310-6. doi: 10.1603/ec12422.
The ability to monitor verde plant bug, Creontiades signatus Distant (Hemiptera: Miridae), and the progression of cotton, Gossypium hirsutum L., boll responses to feeding and associated cotton boll rot provided opportunity to assess if single in-season measurements had value in evaluating at-harvest damage to bolls and if multiple in-season measurements enhanced their combined use. One in-season verde plant bug density measurement, three in-season plant injury measurements, and two at-harvest damage measurements were taken in 15 cotton fields in South Texas, 2010. Linear regression selected two measurements as potentially useful indicators of at-harvest damage: verde plant bug density (adjusted r2 = 0.68; P = 0.0004) and internal boll injury of the carpel wall (adjusted r2 = 0.72; P = 0.004). Considering use of multiple measurements, a stepwise multiple regression of the four in-season measurements selected a univariate model (verde plant bug density) using a 0.15 selection criterion (adjusted r2 = 0.74; P = 0.0002) and a bivariate model (verde plant bug density-internal boll injury) using a 0.25 selection criterion (adjusted r2 = 0.76; P = 0.0007) as indicators of at-harvest damage. In a validation using cultivar and water regime treatments experiencing low verde plant bug pressure in 2011 and 2012, the bivariate model performed better than models using verde plant bug density or internal boll injury separately. Overall, verde plant bug damaging cotton bolls exemplified the benefits of using multiple in-season measurements in pest monitoring programs, under the challenging situation when at-harvest damage results from a sequence of plant responses initiated by in-season insect feeding.
监测绿叶蝉（Creontiades signatus Distant）和棉花（Gossypium hirsutum L.）棉铃发育的能力，以及棉铃对取食的反应和相关的棉花烂铃，为评估单季测量值在评估收获时棉铃损伤中的价值以及多季测量值是否增强其综合使用提供了机会。2010 年，在德克萨斯州南部的 15 个棉田进行了一次单季绿叶蝉密度测量、三次单季植株损伤测量和两次收获时损伤测量。线性回归选择了两个测量值作为收获时损伤的潜在有用指标：绿叶蝉密度（调整 r2=0.68；P=0.0004）和心皮壁内部棉铃损伤（调整 r2=0.72；P=0.004）。考虑到多次测量的使用，对四个单季测量值进行逐步多元回归，使用 0.15 选择标准选择了一个单变量模型（绿叶蝉密度）（调整 r2=0.74；P=0.0002），使用 0.25 选择标准选择了一个双变量模型（绿叶蝉密度-内部棉铃损伤）（调整 r2=0.76；P=0.0007）作为收获时损伤的指标。在 2011 年和 2012 年使用品种和水管理处理进行验证时，双变量模型的表现优于单独使用绿叶蝉密度或内部棉铃损伤的模型。总体而言，绿叶蝉损害棉花棉铃的例子说明了在昆虫取食引发的一系列植物反应开始后，在收获时损伤的挑战性情况下，在害虫监测计划中使用多个单季测量值的好处。
