Childers J W, Thompson E L, Harris D B, Kirchgessner D A, Clayton M, Natschke D F, Phillips W J
ManTech Environmental Technology, Incorporated, P.O. Box 12313, Research Triangle Park, North Carolina 27709, USA.
Environ Sci Technol. 2001 May 1;35(9):1859-66. doi: 10.1021/es001744f.
Open-path Fourier transform infrared (OP/FT-IR) spectrometry was used to measure the concentrations of ammonia, methane, and other atmospheric gases at a concentrated swine production facility. A total of 2200 OP/FT-IR spectra were acquired along nine different monitoring paths during an 8-day period between January 11 and 22, 1999. Standardized quality control (QC) procedures were applied to the archived OP/FT-IR spectra to verify that the instrument was set up and operating properly during the field study and to identify outliers in the concentration data. These QC procedures included measuring the random baseline noise, the signal strength, and the relative single-beam intensity in selected wavenumber regions; inspecting the archived spectra for wavenumber shifts, changes in resolution, and evidence of detector saturation; and examining time series plots of the target gas concentrations and the uncertainty values reported by the classical least-squares (CLS) analysis. Application of these QC procedures to the archived spectra identified 252 potential outliers. After a careful review of the original spectra, 41 of the 252 suspected outliers were designated as actual outliers. Of the QC criteria used during this study, the uncertainty values reported by the CLS analysis were the most reliable indicators of actual outliers.
开路傅里叶变换红外（OP/FT-IR）光谱法被用于测量一家集约化养猪场中氨、甲烷及其他大气气体的浓度。在1999年1月11日至22日的8天时间里，沿着九条不同的监测路径共采集了2200条OP/FT-IR光谱。对存档的OP/FT-IR光谱应用标准化质量控制（QC）程序，以验证仪器在实地研究期间的设置和运行是否正常，并识别浓度数据中的异常值。这些QC程序包括测量选定波数区域的随机基线噪声、信号强度和相对单光束强度；检查存档光谱的波数偏移、分辨率变化以及探测器饱和的证据；以及检查目标气体浓度的时间序列图和经典最小二乘法（CLS）分析报告的不确定度值。将这些QC程序应用于存档光谱，识别出252个潜在异常值。在仔细审查原始光谱后，252个疑似异常值中有41个被确定为实际异常值。在本研究中使用的QC标准中，CLS分析报告的不确定度值是实际异常值最可靠的指标。
