Protocol Development for Converting Large/Complex Legacy SWMM Models to SWMM5

George Remias, Mike Gregory, Denis Zaharija and Shirly Schluchter


The U.S. Environmental Protection Agency’s StormWater Management Model (SWMM) has been in the public domain for nearly 40 years. It remains among the most popular simulation tools in the world for hydrologic, hydraulic, and water quality analyses in stormwater and sanitary sewer systems. SWMM Version 5 was released in October 2004, representing a significant update from previous versions including rewritten and restructured code, new input file structure and data format, new graphical user interface, modified algorithms, and new features not available in previous versions of SWMM. There are strong incentives to use SWMM5 including: improved computational efficiency, accuracy, and numerical stability; compatibility with GIS and asset databases; and a wide selection of third party add-ons and tools to enhance the ease of use, interpretation and presentation of results. Further, many of SWMM5’s new features are essential for continuous hydrologic simulation and the analysis and design of Low Impact Development facilities.

Given its distinctive lineage, there have been many major investments in model development and ongoing maintenance using previous SWMM versions. While SWMM5 includes a tool to convert previous SWMM models, it has only basic functionality, poor error checking capabilities, and a lack of documentation and support. In addition, modeling methodologies and naming conventions that were acceptable with earlier versions of SWMM are not necessarily compatible with SWMM5 or its converter program. Therefore, the conversion of legacy models to SWMM5 can present a significant challenge, particularly for large or complex stormwater networks. Consequently, simply utilizing converted SWMM models prior to following a proper QA/QC procedure can lead to erroneous/inaccurate results and potential mistakes when assessing problems and evaluating solutions.

One such major model investment was the Regional Intercommunity Drainage Evaluation (RIDE) Study, an initiative undertaken by the Northeast Ohio Regional Sewer District (District) from 2001-2005. The objective of the RIDE Study was to develop planning-level solutions to alleviate intercommunity drainage problems that presented an immediate threat of flooding or streambank erosion to structures, roadways, and infrastructure within the Cleveland metropolitan area. The study area represented a 414 mi2 (1,072 km2) multi-jurisdictional drainage area, including over 760 mi (1,230 km) of open channels, detention basins, culverted streams, and pipes. A key deliverable from the RIDE study was the development, calibration and application of two dozen watershed models throughout the study area using Version 4.4h of SWMM.

The District is currently updating the RIDE watershed models to use as technical planning tools for evaluating and implementing stormwater related projects under their construction program to address flooding and erosion problems on a watershed-wide basis throughout their service area. As part of this effort, a SWMM5 model conversion protocol was developed that includes step-by-step procedures for ensuring that the converted models properly reflect the hydrologic and hydraulic conditions represented in the original models. This paper highlights the key project features and critical lessons learned from the SWMM5 model conversion effort that will benefit other agencies, communities, and SWMM modelers faced with the same challenges.

Permanent link: