费城水利部（PWD）最近制定了一项全面的水源保护计划，作为其改善费城水环境工作的一部分。该项目包括一个水源评估（SWA）合作项目，该项目是根据《安全饮用水法案》为宾夕法尼亚州舒尔基尔河和特拉华河流域的供水商开展的。这些评估包括在整个流域内确定5000多个潜在污染源，并在50个取水口处确定取水口特定污染源的优先顺序。为了高效、经济地开发这样一个大规模的水源水评估项目（SWAP），建立了一个统一、全面的优先顺序流程。本文介绍了创新数据管理系统的数据采集和开发、用于描述非点源特征的降雨径流建模过程、其他此类大型水源水保护举措的工具和技巧的优缺点。 污染物清单数据库管理系统包括点源和径流污染物负荷。使用选定的标准对水源进行分析，并对每个饮用水取水口进行优先排序。这一潜在来源汇编是为两个大型流域实施的互换过程的一个组成部分。最初的数据汇编程序严重依赖于从几个联邦和州数据库下载，以快速生成一份包含5000多个废水、RCRA和其他来源的清单。利用当地供水商的数据对数据进行了整理和验证。结合使用地理信息系统（GIS）处理和数据库管理技术，开发了一套综合工具，对每个取水点的水源进行初步筛选，并为本项目中使用的多标准优先排序模型EVAMIX生成输入数据文件。利用数据库和GIS工具对模型的输出进行了编辑和处理，以编制各种进水口的评估报告。 本文解释了数据库和GIS工具开发过程，为其他SWAP项目提供了指导，并讨论了使用此类大型流域软件工具时应避免的陷阱。交换计划的关键要素之一是非点源的特征描述。需要对11000平方英里以上的综合流域的潜在非点源和径流污染物负荷进行估算，这是一个更为困难的挑战。研究区域从高密度城市化区域到农业区域。流域水文和水质负荷的全面、全流域、长期连续模拟模型满足了污染物负荷建模和量化的挑战。本文描述了模型方法及其对此类项目的有用性。包括2个参考文献、图表。

The Philadelphia Water Department (PWD) has recently developed a comprehensive Source Water Protection Program as part of its effort to improve Philadelphia's water environment. The program includes a source water assessment (SWA) partnership project, conducted in compliance with the Safe Drinking Water Act, for water suppliers in the Schuylkill and Delaware River Watersheds in Pennsylvania. These assessments include identification of over 5,000 potential sources of contamination over the entire watershed and intake-specific source prioritization at 50 intakes. To efficiently and cost-effectively develop such a large-scale source water assessment program (SWAP), a uniform and comprehensive prioritization process was established. This paper describes the data acquisition and development of an innovative data management system, the rainfall-runoff modeling process used to characterize the nonpoint sources, the advantages and disadvantages of the tools and tips for other such large source water protection initiatives. The contaminant inventory database management system incorporated both the point sources and the runoff contaminant loads. The sources were analyzed using selected criteria and prioritized for each of the drinking water intakes. This compilation of potential sources is one component of the SWAP process implemented for the two large-scale watersheds. The initial data compilation program relied heavily on downloads from several federal and state databases to quickly generate a list of over 5,000 wastewater, RCRA and other sources. The data was organized and verified using local water suppliers' data. Using a combination of geographic information system (GIS) processing and database management techniques, a comprehensive set of tools were developed to conduct preliminary screening of the sources for each intake and to generate input data files for EVAMIX, a multi-criteria prioritization model that was used in this project. The output from the model was compiled and processed using the database and GIS tools to develop the assessment reports for the various intakes. This paper explains the processes involved in the development of the database and GIS tools, provides guidance for other SWAP programs, and discusses the pitfalls to be avoided when using such large-scale watershed software tools. One of the critical elements of a SWAP program is the characterization of the nonpoint sources. The need to provide estimates of potential nonpoint source and runoff contaminant loads from a combined drainage area of over 11,000 square miles presented one of the more difficult challenges. The study area ranges from high density urbanized to agricultural areas. A comprehensive, basin-wide, long-term continuous simulation model of watershed hydrology and water quality loading met the challenge of modeling and quantifying contaminant loads. This paper describes the model methodology and its usefulness for such projects. Includes 2 references, figures.