USDA-ARS Environmental Microbial and Food Safety Laboratory, Beltsville, MD, USA.
Dep. of Environmental Science and Technology, Univ. of Maryland, College Park, MD, USA.
J Environ Qual. 2020 Nov;49(6):1633-1643. doi: 10.1002/jeq2.20164. Epub 2020 Nov 16.
Several manure-borne microorganism removal models have been developed to provide accurate estimations of the number of microorganisms removed from manure or manured soils undergoing rainfall. It has been commonly assumed that these models perform equally well when used to simulate microbe removal in runoff from manures of different consistency and levels of weathering. The objectives of this work were (a) to observe kinetics of the removal of Escherichia coli and enterococci with runoff for two different manure consistencies and three manure weathering durations, and (b) to compare performance of the log-linear, Vadas-Kleinman-Sharpley, and Bradford-Shijven models in simulation of the observed kinetics. Liquid and solid dairy manure were applied to grassed soil boxes that received simulated rainfall immediately after application and subsequently at 1 and 2 wk. Runoff samples were collected for 1 h at increasing time intervals during each event. Only the effective rainfall depth at the start of runoff was significantly affected by manure consistency (p = .033), whereas other parameters were not (p > .05). Substantial differences in microorganism removal kinetics during the initial, 1-, and 2-wk rainfall events were manifested by the significant (p < .05) effect of the degree of manure weathering in about 70% of cases. The log-linear model produced the largest fitting error especially during the initial rainfall event. The Vadas-Kleinman-Sharpley model and the Bradford-Schijven model were comparable in accuracy for all events. The latter model was slightly more accurate, and the former model had better expressed dependencies of parameter values on manure weathering. Ignoring manure weathering may lead to incorrect parameterization of manure removal models.
已经开发了几种粪肥传播微生物去除模型，以提供对降雨过程中从粪肥或施肥土壤中去除微生物数量的准确估计。通常假定，当用于模拟不同稠度和风化程度的粪肥径流中的微生物去除时，这些模型的性能相同。这项工作的目的是：（a）观察两种不同粪肥稠度和三种粪肥风化持续时间的粪肥径流中大肠杆菌和肠球菌的去除动力学；（b）比较对数线性、Vadas-Kleinman-Sharpley 和 Bradford-Shijven 模型在模拟观察到的动力学方面的性能。将液态和固态奶牛粪肥施用于草地土壤箱，施粪后立即进行模拟降雨，然后在 1 周和 2 周后进行。在每次事件中，以增加的时间间隔收集 1 小时的径流样本。只有在径流开始时的有效降雨深度受到粪肥稠度的显著影响（p=0.033），而其他参数则不受影响（p>0.05）。在最初、1 周和 2 周降雨事件期间，微生物去除动力学的显著差异表现为粪肥风化程度的显著影响（p<0.05），约有 70%的情况下会出现这种情况。在初始降雨事件中，对数线性模型产生的拟合误差最大。在所有事件中，Vadas-Kleinman-Sharpley 模型和 Bradford-Schijven 模型的准确性相当。后一种模型略为准确，前一种模型更能表达参数值对粪肥风化的依赖性。忽略粪肥风化可能会导致粪肥去除模型的参数化不正确。
