Estimating Yield Losses in Soybean Due to Sourgrass Interference

ABSTRACT: Efficient management of sourgrass is one of the main challenges faced by Brazilian soybean farmers. Biological and ecological characteristics of this weed, as well as wide distribution of glyphosate-resistant biotypes, make it difficult to control this species. The objective of this work was to estimate interference of sourgrass in soybean yield. Seven experiments were conducted during three consecutive years (2012/2013, 2013/2014 and 2015/2016 cropping seasons); six of them were on-farm trials where sourgrass infestation was mainly originated from clump regrowth, while infestation originated from seeds in one trial. Data from all sites and experiments were submitted to a combined analysis, according to the study factors and variables, to determine the greatest number of information points for the variable under each circumstance. To each data set, a 1st degree polynomial regression was fit, with a 95% confidence interval. The degree of sourgrass interference varies according to plant origin (seed or clump regrowth) but, in both cases, yield losses are directly correlated with sourgrass densities, aboveground dry mass accumulation and soil coverage. Plants coming from clumps tend to cause higher yield losses than those originated from seeds. These data highlight the importance of performing proper management of sourgrass, especially in its initial stages of development, in view of the high level of losses that perennial plants can cause to crops.

摘要：对铺地黍（sourgrass）的高效防控是巴西大豆种植从业者面临的核心挑战之一。该杂草的生物学与生态学特性，加之抗草甘膦生物型的广泛分布，使其防治难度显著提升。本研究旨在评估铺地黍对大豆产量的干扰效应。研究于连续三个种植季（2012/2013、2013/2014及2015/2016）开展了7组试验：其中6组为农场田间试验，其铺地黍侵染主要来源于丛株再生，剩余1组试验的侵染源为种子。所有试验位点与试验数据均依据研究因子与变量进行合并分析，以确定各情境下目标变量的最优信息点数量。针对各数据集，我们拟合了一阶多项式回归模型，并设置95%置信区间。铺地黍的干扰程度因繁殖起源（种子或丛株再生）存在差异，但在两种情境下，产量损失均与铺地黍密度、地上干物质积累量及土壤覆盖度呈显著正相关。丛株再生的铺地黍对大豆产量造成的损失显著高于种子起源的个体。本研究结果凸显了对铺地黍实施精准防治的重要性，尤其在其生长初期，鉴于多年生杂草可对作物造成极高水平的产量损失。