Sunnyside Needs Improved Flood Barrier Protection
Please be sure to drop in to the important City of Calgary Open House on the Sunnyside Flood Barrier Project to be held in September.
When: Tuesday, September 24, 2019
Where: Sunnyside School Gymnasium
211 7th St. NW
The City is exploring various barrier/berm height options that will provide different levels of protection from high river levels. At the open house you can learn more about the options being explored in Sunnyside. Your feedback will be important to ensuring the right option is selected.
For the last 2+ years the HSCA EPARC Flood Infrastructure Committee has been advocating for a flood barrier (aka berm) that will provide adequate protection for Sunnyside. Pressure from this group was key to pushing the City to consider alternatives to their original plan. Now we are calling on community members to make their voices heard so that an appropriate (higher) alternative is selected – the more voices, the better.
The City’s original plan for our berm is for no material height increase and this is not acceptable. Sunnyside needs a higher berm, and we need it now. How much higher has not been determined, but a one-half metre height increase is the minimum plausible.
Below are explanations of some terminology used in discussing river flooding and the berm. These are intended to provide clarity and understanding. One does not have to get very deep into the subject of river flooding for it to get very complicated. Although this note is very long we have tried to simplify for brevity.
Explanation of Terms Used When Discussing the Berm
(1) River Flow. Rainfall and snowmelt west of Calgary can cause high flows in the Bow River. In the case of large floods the most important factor is the rainfall, with snowmelt contributing perhaps only 10-20% of the flow. At any point in time the flow in the Bow River is different at different locations. The river starts with a small flow at Bow Glacier and the flow increases with tributaries joining as the river runs eastwards.
The Ghost Reservoir west of Cochrane is also important. Depending on the available room in the reservoir and the management decisions taken the reservoir can temporarily hold back water and reduce the outflow to Calgary to less than it would otherwise be. More on this below. On the other hand, the Bearspaw Reservoir has little storage capacity and therefore little potential impact on river flow.
Complex hydraulic models are used to project river flows under different scenarios of rainfall and snowmelt.
To simplify discussion, when we speak of river flow, we always refer to the flow of the Bow River upstream of the confluence with the Elbow River, as measured by an Alberta Environment and Parks flow meter on the Bow at Reconciliation Bridge in downtown Calgary. The flow is measured in cubic metres per second which is abbreviated either as m³/s or cms. The flow at Reconciliation Bridge is a good approximation of the flow at Sunnyside because the contribution from runoff between here and there is relatively small.
(2) Return Period. The risk associated with major floods is expressed as a return period. For example, 1:100 means that a specific high flow is expected to return on average once every 100 years. This is more correctly stated as “a 1% chance of that flow occurring in any year”.
Below is a table translating return periods into probabilities of floods in any one year and over time. Note that at a constant risk the specific flow will be different at different locations up and down the Bow River. The river flow at Reconciliation Bridge corresponding to each return period is shown at the right of the table. These flows are calculated by taking historical river flow data, using statistical techniques to fit the data to a curve and then picking flow values off the curve for the chosen return periods.
The probability of occurrence or the return period is a measure of the risk that we are willing to tolerate. Risk tolerance varies from person to person and represents our perception of the consequences of the flood versus our perception of the cost of mitigating against that flood.
An example of acceptable risk tolerance is the Alberta provincial standard for flood protection which is 1:100, or a 1% probability in any year. In the Netherlands their standard for flood protection can be as high as 1:1000 or more. Calgary has no established flood protection standard, instead individual flood protection infrastructure is built to different risk tolerances, seemingly at the whim of City Administration.
It might seem counterintuitive that there is only a 65% chance of a 1:10 event occurring in any 10-year period. However this is correct, and is a result of the way probabilities are mathematically combined.
(3) River Level. The level of the water in the river depends on the flow in the river, but it also depends on the width of the river at any point. The same flow increase in a wide section of the river will result in less level increase than in a narrower section of river.
The effective width of the river at a specific location can vary with the depth of the water. For example, at 10th Street NW there is a gravel bar. When that gravel bar is dry the river is flowing in a narrower channel and a smaller change in flow will result in a given change in water level. When the river level rises above the gravel bar its width is effectively greater and a larger change in flow is required for that same change in water level. This change in width with depth is referred to as the river cross-section, and it is different at every point along the river. Sophisticated river surveys and flow models are used to estimate projected river levels for specific flows.
A further complication is that the Bow River is in some ways more like a mountain stream when compared to most major rivers. In other words, the Bow River is falling rapidly downhill as it passes through Calgary. Just from one end of Sunnyside to the other the river elevation drops about two metres.
(4) Berm Height. From the paragraphs above we can say that the height of an adequate berm at a specific point depends on the desired risk tolerance as indicated by the probability or return period of a damaging flood, the river flow associated with that return period at that point in the river and the river cross section, or effective width, at that point.
In addition, good engineering practice is to allow for “freeboard”. This allowance is to compensate for any modelling errors, upstream or downstream debris dams, and wave action including any standing waves. Freeboard is to ensure that the berm is not overtopped in a flood of the return period for which it was designed. In the case of the Bow River the freeboard allowance recommended by City engineers is one-half metre (0.5 m). We are using that number.
The table below takes all the discussion above and identifies the increase required to the Sunnyside berm at various points along the berm. The elevation of the existing berm is given in metres above sea level.
There are a lot of numbers in that table. Here is a simple summary:
For 1:20 protection: No change to current berm height (the City’s original proposal)
For 1:50 protection: The berm must be raised about 0.5 m
For 1:100 protection: The berm must be raised about 1.0 m (the Alberta provincial standard)
For 1:200 protection: The berm must be raised about 1.5 m (to match the new flood protection on the south bank)
Note that in the case of 1:100 and 1:200 protection the berm should also be raised in Hillhurst.
It is plausible that for modest height increases over modest lengths temporary “water tube” type berms could be deployed when high river levels are anticipated. This might be an acceptable plan for Hillhurst and perhaps everywhere west of the Peace Bridge, but it is implausible that temporary berms could be deployed for the full height and along the entire length required in Sunnyside. The City has limited resources and many other priorities for water tube berms.
(5) Ghost Reservoir TransAlta Agreement. The most complex and contentious issue has been left for last. To properly understand the impact of the TransAlta agreement for management of Ghost Reservoir levels and outflows requires the use of differential equations that I last studied in second year university, as well as an understanding of human nature and people’s reactions during crises that I only know through observation during our 2013 crisis. This is to say that I find it difficult to explain the possible operation of the Ghost reservoir in a clear and succinct way.
There are other upstream reservoirs besides Ghost included in the agreement between the Government of Alberta and TransAlta for reservoir management before and during a flood event. For the purpose of simplicity, they will be lumped together as “Ghost”.
The Ghost Reservoir has a working capacity of up to 61,000 dam³ (dam³ is an abbreviation of cubic decametre. 1 dam³ equals 1000 m³, and m³ means cubic metres). Working capacity means the amount of water the dam can store if it is as empty as possible at the beginning of an event and is allowed to fill to just before overflowing.
The flow rates into the Ghost Reservoir from the Bow River and the Ghost River depend on the amount of rainfall upstream. The flow rate out of the Ghost Reservoir is controlled by TransAlta in consultation with the Province and the City at whatever rate they choose, until the reservoir is full. Outflow rates that might be chosen correspond to the following flows at the Reconciliation Bridge in Calgary:
Up to 400 cms – no special action required in the City. Allows time for thorough evacuations.
Between 400 and 800 cms – Pathways closed, some basement groundwater flooding expected.
Between 800 and 1230 cms – Significant overland flooding in Bowness (until their berm is built)
Over 1230 cms – Damaging flooding in Sunnyside unless a higher berm is built.
For comparison the Bow River flow was about 250 cms at its highest this summer, the peak flow for a 1:100 flood is 2020 cms, and the actual flow at the peak of the 2013 flood was 1850 cms.
The river flows corresponding to return periods listed in the berm height table above are “peak flows”. If the Ghost Reservoir is not used (outflow = inflows) it will take some time for the river flow to rise to that peak flow, where it will remain for a relatively short time before gradually declining, eventually returning to a normal low flow (the flow if the Ghost Reservoir is not used can be called the naturalized flow). Representing this rising-peak-declining inflow pattern and calculating the time to fill the reservoir at a chosen outflow is where the differential equations are used. When the reservoir becomes full there is no option other than to increase the outflow to whatever the inflow happens to be. If the reservoir is full before the naturalized peak flow is reached then that full peak flow becomes the outflow.
The key point for us is the amount by which the actual peak flow is less than the naturalized peak flow. Estimates of the amount by which the naturalized peak flow can be reduced vary from 0 cms to 500 cms.
The 0 cms peak flow reduction benefit occurs if the reservoir capacity is used to delay the onset of high flow to ensure complete in-city preparations. The use of the capacity for this purpose has been suggested by both the City and the Province in previous communications. Reservoir capacity used for high flow onset delay will not be available for peak flow reduction.
The 500 cms peak flow reduction benefit is a very optimistic case where perfect knowledge of the incoming flow patterns is available and immediate and aggressive action is taken to increase the reservoir outflows. Of course, perfect information is not available in a flood emergency. I suggest that it is unlikely that officials with less than perfect information will quickly take the decision to increase the flows and knowingly flood Bowness. And failure to take quick action means the full 500 cms reduction cannot be achieved.
In our analysis in the next section we have chosen an intermediate Ghost/TransAlta peak flow reduction benefit of 300 cms. Applying this to the berm height table above we can see that our “1:20” berm will protect us up to a flood of about 1:40, while a “1:50” berm would protect us in a flood up to about 1:90 (the peak flows quoted in the table are naturalized flows). Similarly, a “1:100” berm would protect to about 1:180 and a “1:200” berm would protect to almost 1:350. Is our 300 cms peak flow reduction benefit choice correct? Maybe, but maybe it is too optimistic. Probably the peak flow reduction is not constant for floods of different sizes as these calculations assume.
In summary, we do not accept the City’s contention that a Ghost/TransAlta benefit of 500 cms is appropriate and we feel that a reasonably optimistic assumption is 300 cms peak flow reduction benefit.
(6) Upstream Mitigation. The province of Alberta is now studying options for a new or expanded dam on the Bow River upstream of Calgary. It is proving challenging to find a suitable location for such a dam. Even if a location is identified the cost of a new dam will be close to $1 billion. With the current state of the province’s finances it is not clear that money would be available to begin significant work anytime soon. In the very best case scenario, a new or expanded dam would not be operational for 10 years minimum with 15, 20, 25 years or never all more likely cases. When/if this dam is completed it will provide good protection for all of Calgary and a 1:20 or 1:50 berm would be sufficient for Sunnyside. During the time we are waiting for this dam we will be at significant risk of a damaging flood.
Comments on Alternative Berm Heights
[This section was previously distributed as part of the HSCA August e-newsletter]
The City’s original plan was for no material change to the height of the Sunnyside berm, which on its own provides protection against a 1:20 flood (1:20 means a 5% probability of occurrence in any year). When the 1:20 berm is combined with reasonable estimates of upstream dam operation and future upstream infrastructure construction there is a 30% probability of a damaging flood in the next 15 years. This is unacceptable, especially when a higher berm is technically, economically, environmentally and socially feasible.
The stress that is palpable in the community each year when flood season rolls around is an important social factor that a higher berm would help alleviate.
Another important social factor the lack of equity and fairness between communities. Recently constructed or proposed barriers protecting Inglewood, Downtown, Eau Claire, and even the Zoo and the Field of Crosses are substantially higher than that originally planned for Sunnyside. This discrimination against our densifying inner city, mixed income community is unacceptable.
If Sunnyside is protected with an adequate berm the City would consider relaxation of redundant mitigation in buildings. This refers to the prohibitively expensive requirements in the case of significant renovations and to the sub-optimal designs required in new construction. Both impact the character of our community – an important social factor.
In addition to the social factors above the following technical points are important.
The City’s own Triple Bottom Line economic analysis shows that a higher berm for Sunnyside is solidly justified.
The provincial standard for flood mitigation is 1:100 (1% chance in any year). Even in combination with improved upstream facility management the berm originally planned falls short. This jeopardizes any provincial contribution to the cost of the project, and is the reason no provincial funds have been awarded to this project to date. It is likely that the provincial money attracted by a higher berm would more than cover any additional cost for the added height.
All of the probabilities mentioned in this email assume no impact of climate change. If more frequent extreme weather events occur as a result of climate change then larger floods will be more common. For example, what is now projected as a 1:100 flood (1% chance in any year) could be as frequent as 1:50 in future (2% chance in any year), and so on. Consideration of climate change would further justify a higher berm.
For more information on all the Flood Committee concerns with the flood barrier project please refer to
The original flood barrier plan for a 1:20 berm is unacceptable. We hope you will join as we advocate for one of the other options to be presented. These include berms that would protect against river levels that would be expected to occur 1:50 (2% chance in any year), 1:100 (1% chance in any year) or 1:200 (0.5% chance in any year). Each of these options will have their own pros and cons.
We are nearing a critical point regarding the protection of our community in the event of future river flooding. Your support and some of your time is needed at the Community Engagement Open House on the Sunnyside Flood Barrier Project to be held on Tuesday, September 24. I look forward to seeing you there.
I would be happy to discuss this whole issue or any part of it further individually or in a group. If there is enough interest we can convene a meeting of those who want a better understanding. Either way, please contact me at firstname.lastname@example.org
Chair, Infrastructure Group, Emergency Preparedness and Response Committee
Hillhurst Sunnyside Community Association.