Establishing Chemometric and Statistical Foundations of Receptor Models

受体建模是一个寻找污染源化学指纹的跨学科研究领域。受体模型使用化学计量学、化学、统计学和图像分析以及许多其他领域的方法。这些模型通常采用主成分分析法和因子分析法作为主要成分。受体建模需要良好的科学知识来合理地模拟观测数据。例如，知道乙炔是机动车排放的示踪剂对成功的受体建模至关重要。本研究开发的方法将用于提供污染源的指纹识别。该研究也将为该方法提供坚实的数学基础。这项研究为受体建模提供了坚实的化学计量学和统计学基础，是对联邦环境战略领域的重要贡献。受体模型将为确定污染源提供公正的科学依据。受体模型应该能够在立法和法庭诉讼中不受玷污。

Receptor modeling is a cross-disciplinary research area that finds chemical fingerprints of pollution sources. Receptor models use methods from chemometrics, chemistry, statistics, and image analysis, as well as many other fields. These models commonly use principal component analysis, and factor methods as their main ingredients. Receptor modeling requires good scientific knowledge to reasonably model observed data. For example, knowing that acetylene is a tracer for motor vehicle emissions is essential to successful receptor modeling. The methods that this research develops will be used to provide fingerprint identification of pollution sources. The research will also provide a firm mathematical basis for the methods. This research provides a firm chemometric and statistical basis for receptor modeling and is an important contribution to the Federal Strategic Area of the Environment. Receptor models will serve as unbiased scientific evidence for identifying pollution sources. Receptor models should be able to stand untarnished in legislative and court proceedings.