Stiber Neil A, Small Mitchell J, Fischbeck Paul S
a Carnegie Mellon University , Pittsburgh , Pennsylvania , USA.
J Air Waste Manag Assoc. 1998 Sep;48(9):809-818. doi: 10.1080/10473289.1998.10463736.
A methodology is presented for estimating the probability that particular classes of environmental contaminants will be of concern at brownfield redevelopment sites. These probabilities are predicted by a logistics model that is based on qualitative information about site history and status. This qualitative information comprises data that would be collected through a Phase I Environmental Site Assessment (ESA), including historic site use, current use and ownership status, and the nature of adjacent properties. The model is fit and demonstrated using a set of 59 former industrial sites in southwestern Pennsylvania that were collected from the files of the Pennsylvania Department of Environmental Protection (PADEP). Predictive models are developed for exceedances of contaminants as grouped into the following classes: metals, chlorinated hydrocarbons, fuel hydrocarbons, and PCBs. A procedure for estimating the parametric uncertainty of the model predictions is also illustrated. This method can serve as a starting point for more effective usage of existing Phase I ESA information and for evaluation of the benefit of obtaining additional site information. By increasing the decision-making value of existing (or inexpensive) data, this method can help to reduce the information asymmetry that may be an obstacle to redevelopment.
本文介绍了一种方法，用于估算特定类别的环境污染物在棕地再开发场地引起关注的概率。这些概率由一个基于场地历史和现状定性信息的逻辑模型预测。该定性信息包括通过第一阶段环境场地评估（ESA）收集的数据，包括历史场地用途、当前用途和所有权状况以及相邻地块的性质。使用从宾夕法尼亚州环境保护部（PADEP）档案中收集的宾夕法尼亚州西南部59个 former industrial sites对该模型进行拟合和验证。针对以下分类的污染物超标情况开发了预测模型：金属、氯代烃、燃料烃和多氯联苯。还说明了一种估算模型预测参数不确定性的程序。该方法可作为更有效利用现有第一阶段ESA信息以及评估获取额外场地信息效益的起点。通过提高现有（或低成本）数据的决策价值，该方法有助于减少可能成为再开发障碍的信息不对称。
