Webinar Q&A: innovative & intelligent stormwater modeling approaches to combat flooding: 3 high-impact projects
As part of Stormwater Awareness Week, this Stormwater Solutions and Stormwater University webinar dived into flood control. As communities face more frequent and intense flooding, advanced stormwater modeling is becoming essential for designing resilient infrastructure to manage community risk. Speakers from OHM Advisors explored three high-impact projects that demonstrated how project-tailored modeling approaches and advanced simulation techniques can guide infrastructure investments and improve flood mitigation strategies.
Attendees gained a practical understanding of how recent advances in stormwater modeling tools can help identify management solutions to some of today’s most pressing water infrastructure challenges.
Below, speakers Kristen Chaffin and Brandon Ellefson answer audience questions.
To view the webinar on-demand, click here.
Audience: How long did the modeling take?
Speakers: The time it takes to set up a model is dependent on the state of the input data and the size of the system. If the system is a very large and complex system, or the GIS records of the collection system are outdated and missing key pieces of information, then significant time is needed, both in the field and sifting through city files to gather required system information to build the model. Also, if flow data will be used in the calibration, time is needed to collect that data in calibrating the model.
The model for a system as complex as Dearborn was a two-year build, and it is still undergoing updates as the city carries out maintenance and improvements. This was also a simplified version of the collection system, so it would take more time to build the entire system.
On the other end of the spectrum, the Watervliet model is a small neighborhood model that had all the network information digitized and in the city’s GIS database. This model was built in one 8-hour day. Also, it was not calibrated with flow data, only anecdotal flooding location information as reported by residents.
Audience: Regarding the dam, I am trying to understand how something so small could have such a large impact.
Speakers: The aerial extent of the dam is not important in evaluating the upstream effects of a dam in a system where the hydraulic grade line (water surface elevation) upstream of the dam is controlled by the dam elevation. Regardless of the size of the pond upstream of the dam, or the distance the hydraulicly connected pipe network is away from the dam, if the HGL (water surface elevation) set by the dam is higher than the inverts of the pipes in the collection system upstream, then it will cause flows to back-up in the system to match that HGL (water surface elevation).
Audience: When you created the modeling, did you obtain survey data for inverts or was that data already available?
Speakers: We started with available data from city records. This information was supplemented with collected survey data at critical/key locations. Remaining missing inverts are interpolated using the inverts of surrounding pipes.
Audience: For the third example, did the flooding problem stem from the ghost stream or the pipe sizing?
Speakers: Both. First, the ghost stream was the predominant topographical feature in this particular neighborhood. This was not a curb and gutter neighborhood, so the runoff flow did not always flow to the streets and enter the conveyance system at the first catch basin it encountered. Instead, runoff flows traveled towards the ghost stream/low topography, and in some cases, they flowed across streets and down into a swale on the other side rather than entering the system at the upstream catch basins. The existence of the ghost stream changed the flow patterns of the runoff from the pattern that the client perceived was happening. The water was entering the system along the ghost stream rather than higher up in the system along the roadway. Second, the system pipes that ran along the ghost stream did not have enough catch basins to accept the flow that was now entering the system at this location rather than being distributed along the length of the system. Third, the pipes along the ghost stream were undersized and lacked the capacity to transport the water out of the system. This was due to two factors, one, growth in the area so the current system was undersized (more houses equals more impervious area, equals more runoff volumes) and two, additional flow was now tributary to the pipes along the ghost stream that, if the ghost stream was not present, would have flowed into other areas of the system and out other outfalls.
Audience: How close are the FEMA flood maps to your modeled surface flooding?
Speakers: The modeled area in the third case study in which a 2D model was developed and surface flooding mapped was not within a FEMA floodplain so this comparison could not be made.
Audience: I'd love to see where the AMM model is located.
Speakers: The details on the Antecedent Moisture Modeling used in the Dearborn case study can be found at the following web location.
Related
