Improved Estimates Of Storm Water In Sewers Could Help Reduce Flooding

Recent PhD graduate Gabrielle Migliato Marega (CivMin) developed an improved tool for estimating how much water from heavy rains ends up in sanitary sewers

A new strategy for estimating how much rainwater finds its way into sanitary sewer systems could help prevent problems such as backups and floods, while reducing treatment costs.

The tool is particularly useful in areas where accurate data about sewer flow rates is difficult to find, such as in low and middle-income countries, where many new sewer systems are being built.

“In theory, storm sewers and sanitary sewers should be separated, but in many older systems, they are combined,” says Gabrielle Migliato Marega (CivMin PhD 2T6), lead author on a new paper published in Water Science & Technology.

“Even if they are theoretically separate, there’s always crossover: it’s unrealistic to think that no stormwater is getting to the system. That extra stormwater can overwhelm the plants designed to treat the wastewater, so you get raw sewage flowing into lakes and rivers. You can also get problems like sewer backups and basement flooding.”

In the past, several different methods have been used to estimate how much storm water is getting into sanitary sewers.

“One common approach is based on flow rates: you basically take an average flow at times when it’s not raining heavily, another average at times when it is raining heavily, and then subtract the two,” says Marega.

When this level of detail is not available, the amount of storm water can be estimated by comparing the long-term average sewer flow to the amount of sewage a city of a given size is expected to generate.

“Another approach is to use some sort of tracer for the presence of sewage. For example, a common water metric is called biological oxygen demand (BOD), which is proportional to the amount of organic material in the wastewater,” says Marega.

“When that wastewater gets diluted with storm water, the BOD will go down, so you can measure that difference to estimate the storm water inflow and infiltration.”

Marega says that previous studies in the field have mostly focused on comparing which of the various methods provided the most accurate estimate of storm water inflow and infiltration into sanitary sewers.

“What’s different about our paper is that we decided to combine two different methods: the long-term average flow rate and BOD methods,” she says.

“If each one is appropriately weighted, we get a much better estimate than would be possible using either method alone.”

To test this out, Marega, who was co-supervised by Professor David Meyer (CivMin) and Professor Jennifer Drake at Carleton University, built a mathematical model that she used to simulate the function of a hypothetical city’s sewer system.

Using what’s known as a Monte Carlo approach, she simulated thousands of scenarios, each with a different storm water inflow and infiltration rate. She then compared how accurately each of the three methods — BOD, average flow rate, or the combined approach — estimated the simulated inflow and infiltration rate.

“We found that our combined method improved accuracy by more than 10%,” says Marega.

Armed with the new strategy, Marega then analyzed data from 46 different cities in her home country of Brazil. In each of these cases, the true rate of storm water inflow and infiltration was not known, and in some, the available data was patchy: flow rates might only be measured on average every three months, rather than every hour as they would be under ideal conditions.

But despite the data-poor environment, there was enough data to estimate storm water inflow using two different methods. Marega’s new approach combined both these methods, creating more plausible estimates of storm water infiltration that could be used to benchmark potential improvements to the system.

“For example, one of the most common causes of storm water inflow and infiltration is from households that connect the drain spouts from their roofs directly into sanitation sewers,” says Marega.

“If you find out that’s happening, you can update the building codes to prevent it or increase inspections to improve compliance. And if you’re building a new system, you can size it larger to ensure it’s going to be better prepared to deal with the larger flows that come with rainstorms.”

Meyer says that the team’s hope is for urban designers around the world to adopt the new method in their planning and maintenance processes.

“The key insight here is that by combining methods, we’re shifting the target,” he says.

“The goal is no longer to answer: what’s the best method? Instead, we’re asking: how can we use existing methods, plural, to learn the most about the system? That should enable us to design better sewers, improve wastewater treatment, and prevent the backups and floods that cause so much damage.”