Our storm sewers can’t handle today’s extremes, they never really could
3 April 1, 2016 at 2:43 pm by Glenn McGillivray
You’ve probably heard this little chestnut on at least a few occasions: “Our underground storm sewers can’t handle the rainfall we get these days.” The implication is that under climate change and with increased runoff in urban centres from the proliferation of impermeable payment, our storm sewers can no longer handle the load from extreme rainfall events.
This is a misnomer.
While, true, underground storm sewer systems weren’t designed to handle some of the extremes we experience today, the truth is, they really never where designed to handle large extremes. It is not as though the systems were able to handle really severe rainfall events when they were first designed and built in the 1940s, 1950s or 1960s, but can’t now due to climate change and increased urbanization. This is a wild mischaracterization of how storm sewers are designed and built, what they are intended to do, and perhaps even the current impacts of climate change on precipitation patterns in Canada.
Canada’s storm water infrastructure can essentially be divided into two major categories, the minor system and the major system (with both essentially representing different storm water management eras in Canada).
For many decades, the minor system – i.e. the system of eavestroughs, downspouts, service connections, catchbasins and underground pipes – was the only formal infrastructure in existence that could be relied upon to convey storm water away from homes and other buildings, and prevent ponding and street flooding during rainfall events. Most areas built before the early 1970s could only really rely on this system to manage storm water in a systemic way.
Then, from about the early 1970s onward, the major system came into play. This system can broadly be characterized as those storm water management assets that use the surface of the earth to convey, collect and store storm water. These can include both natural waterways (such as streams and rivers) and man-made assets, including roadways, drainage ditches, swales, channels and wet/dry ponds.
The underground portion of the minor system is largely designed to handle more frequent (less extreme) 2 year, 5 year and, sometimes, 10 year events (i.e. events whose probability of occurring are 1 in 2, 1 in 5 and 1 in 10 in any given year, respectively). In some cases, a few municipalities are replacing old underground systems with new infrastructure that can handle 1 in 100 year events (Toronto is a prime example), but such efforts are relatively few and far between. (It is important to note that even these more robust systems can – and have been – overwhelmed by extreme rainfall events.)
The major system, conversely, is largely designed to handle runoff from less frequent (i.e. more extreme) rainfall events, often in the range of 1 in 100 year.
Examples of recurrence intervals based on probability of occurrence (Source: climatica.org.uk)
Hence, and as noted by the faculty of engineering at Ryerson University, “By providing the major and minor system for urban drainage, a higher level of flood protection can be provided and the chance of basement flooding can be reduced.”
So, the underground portion of storm water management systems in Canada were most commonly designed and intended to handle more frequent, less severe, rainfall events. In the event of severe rainfalls in older parts of cities (where a large amount of rain falls in a relatively short time and there is no engineered major system in place), flooding often would occur.
So, while it is accurate and truthful to say that our underground storm water networks can’t handle the kind of extreme rainfall events we have been experiencing the last few years; it is equally truthful to say that they never really could handle extreme rainfall events.
This is an important distinction.
Note: By submitting your comments you acknowledge that insBlogs has the right to reproduce, broadcast and publicize those comments or any part thereof in any manner whatsoever. Please note that due to the volume of e-mails we receive, not all comments will be published and those that are published will not be edited. However, all will be carefully read, considered and appreciated.
Thank you Glenn for promoting this – we all need to emphasize design standard adaption to prevent flooding, not climate change mitigation.
As an example, storm sewer capacities have been increased up to 400% to adapt to a higher level of service (up to 100 year) in the GTA municipality where I have developed the long term flood control program. Local rainfall intensities are decreasing for most durations at our local airport, just like many southern Ontario rain gauges. So rain is not the issue.
One factor not noted in the post is the interaction of the major overland system with the sanitary sewer system in older, pre 1970’s areas, particularly because of the risk of extraneous inflows that overwhelm the sanitary system during extreme events. Our experience is that late-1970’s areas with somewhat limited major overland drainage networks have relatively low flood risks if the sanitary sewer is fully separated (that is foundations drains go to the storm and not the sanitary).
In the municipality where I work we have mapped the overland flow paths, and determined estimated ponding depths using an elevation model to identify areas to seal sanitary manholes from overland inflows and have disconnected downspouts in the high inflow areas to manage that risk factor. This has proven to be an effective approach in an area that had extensive flooding in August 2005 but very little in July 2014 – this was achieve with relatively low cost measures (i.e., downspout disconnection and pick hole plugs). And this risk reduction was achieved prior to storm sewer upgrades.
I’m guest lecturing in the Civil Engineering class at Ryerson this Monday and I will highlight the nod to their site!
Rob M
Good insight buddy. Appreciated.
I’ve just mapped out and correlated design standard date (estimated by age of watermain) with reported basement flooding for the 3 recent Toronto floods in 2000, 2005 and 2013, and have shown the relative flood risk in each construction era (normalized by the total number of addresses/buildings per era). Its pretty clear that design standard adaptation is needed before climate adaptation because the post 80’s construction is done pretty good (i.e., lower risk). Here are the results:
http://www.cityfloodmap.com/2016/04/design-standard-adaptation-vs-climate.html
Basement flood clusters are also shown to be correlated spatially with the alignment of ‘lost rivers’, meaning inadequate overland flow routes in pre-1980’s Toronto developments with limited major system design capacity (analysis is based on reported flooding sites in 2000, 2005 and 2013, and estimated 100-year overland flow paths and spread):
http://www.cityfloodmap.com/2016/03/lost-rivers-newtonbrook-tributary.html
Environment Canada scientists have reported that there is no change in rainfall intensity in most parts of Canada:
http://www.cityfloodmap.com/2015/12/trends-in-canadian-shortduration.html
So any meaningful dialogue on flood risk mitigation really has to turn to design standard adaptation.
I’ve also done a similar analysis correlating flood reports to age of sewer construction in the municipality where I work. This is even more ‘granular’ than the Toronto analysis noted above as I have pipe by pipe age of construction that can be related to design standards. Again it shows the post 1980’s flood densities drop of significantly and the new millennium standards do even better – you don’t even need quantity controls to make a difference in the flood history, just dual drainage design (which we adopted in 1978) and full sewer separation (foundation drains to storm no sanitary). Rain trends are downward at our airport. Just more support for design standard adaptation.
Rob M