The state of flood mapping in Canada2 October 17, 2014 at 2:42 pm by Glenn McGillivray
On October 10, ICLR held a Friday Forum workshop entitled ‘National Riverine Floodplain Mapping Framework and Advancements in Urban Overland Flood Risk Assessment’, which largely looked at the state of flood mapping in Canada.
The workshop was lead by Tim Mereu, Vice President of MMM Group, with responsibility for water resources and environmental services. He has thirty years of consulting experience with a focus on water resources, including flood risk management, policy and standards development, master planning, channel restoration, and storm water management. Tim was the project manager for the recently completed National Floodplain Mapping Assessment undertaken on behalf of Public Safety Canada (PSC).
PSC is the lead federal agency responsible for disaster prevention and mitigation. In response to the increase in flood intensity and flood damage over the past number of years, PSC issued a public tender in August 2013 seeking to assess the current state of flood plain maps in Canada in order to help inform potential mitigation activities. In support of this objective, PSC requested three deliverables in its tender: 1) A report on international flood mapping practices in seven countries other than Canada and a recommended standard; 2) An assessment report on the current state of flood plain mapping in Canada; and, 3) A costing report on requirement to bring current Canadian flood plain maps up to a recommended standard.
MMM Group was awarded the contract.
The October 10 presentation focused on the National Floodplain Management Framework including proposed mapping standards, anticipated implementation steps, and the scope and intent of a proposed national flood risk database.
The National Floodplain Management Framework produced by MMM identified:
• The type and extent of information that must be collected and managed;
• The key standards and guidelines that will apply to the development of flood hazard maps and a flood risk database.;
• The anticipated cost of updating and preparing new hazard maps, as required, and the compilation of data for the flood risk data base; and
• A list of initiatives to be completed as part of implementation.
The second half of the presentation focused on recent advancements in effectively defining urban overland flood risk. Urban overland flooding is caused by intense rainfall events in areas where there is insufficient storm sewer system capacity and poorly defined overland flow routes.
Currently, areas in Canada prone to urban flooding are not generally defined. Moreover, in areas that are known to be flood prone the risk is rarely quantified. By better defining areas at risk, municipalities can focus efforts in effective urban overland flood mitigation.
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.
It is quite sad to see that citizens can obtain earthquake coverage in Ontario, yet do not have the ability to obtain flood coverage, even though flood events have been prevalent for decades. There has definitely a demand for this kind of coverage, however, insurers were busy providing frivolous coverages (ie collision damage to shrubbery) rather than taking on the opportunity to find ways to quantify flood risk, to help with what is truly a devastating risk that relies on the pooling of risk in order to protect the unfortunate few who experience flood damage. Insurers have been more intent on inserting exclusions into policies to avoid damage caused by flood, when they should have been working on determining the cost of risk in order to offer their customers a choice. It didn’t take any time for the insurers to start offering cyber risk coverage – why is it taking so long to find a product and solution to provide coverage for flood risk? No-one likes purchasing insurance – it is a grudge purchase. So let the insurance community start tackling the real issues, such as flood insurance!
Seems there are different areas to address – Alberta’s riverine flood risk (incompletely mapped, not well regulated) and Ontario’s urban overland flood risks (rarely mapped, never regulated).
The MMM presentation suggests there are two extremes for assessing the overland flood risks i) topographic mapping that can only define areas of interest but not extent of flooding, and ii) complex data and labour intensive modelling required to “understand risk and assess mitigation measures”. There is a middle ground beyond the topographic screening suggested (slides 85-88) which is GIS-based hydrology and hydraulics using provincial data sets – its neither data nor labour intensive. Here is an example of such analysis:
Why is it so important to not shy away from urban overland flood risk mapping?
1) Because that is where so much of the damages are focused – the Toronto example shows for the May 2000, August 2005 and July 2013 flood events only 2% of reported flooding was in the river flood vulnerable areas. You find a similar breakdown for Burlington’s 2014 where valley flooding documented by the conservation authority is in the double digits but urban flooding is in the 1000’s.
2) Because that is where the regulatory and policy gaps lie. You can’t manage what you don’t map.
We don’t need expensive coupled minor system – major system – 2D spill simulation models to define urban flood risks – those are certainly needed to design remediation. For extreme events, the minor system can be ignored for the purpose of risk management – if that system is woefully undersized everyone knows because of the number of sewer back-up reports or claims. For extreme events the urban overland topography explains risks, and drives the minor system risks too. As the MMM presentation suggests 100 year level of service is insufficient – so the minor system does not figure into those large events of interest.
For urban flood risk mapping, any urban drainage path with 200 ha of area flowing into it is at risk (there are some examples at the end of the slideshare deck). Any urban drainage path with 500 ha of area flowing into it is more at risk. One can apply buffers to the flow path or do the basic hydrology/hydraulics in GIS to define the risk areas as in the slideshare example.
Do we want to end up 10 years from now with updated floodplain maps in valleys where only a couple percentage of Ontarian’s live (per Conservation Ontario)? And no overland flood risk mapping for urban areas where 90% of damages are occurring?
When it comes to urban overland flood mitigation, municipalities need more support at the provincial / agency level to map, manage and regulate the risks. This will take some great conviction – not because of the cost (extending risk mapping onto urban tableland can be done cost effectively with 1D GeoRAS in my experience), but because of the land value impact of mapping hazards on developed properties.
The floodplain mapping framework should look closely at the urban overland flood risk question in the context of high historical damage areas, and ensure the updated mapping includes extending mapping up into tableland, beyond valleys. The real gap is not the age and accuracy of the floodplain maps but their extent – professionals know that there is no correct storm to simulate, no ideal return period, no realistic hyetograph pattern, no perfect rain-runoff transformation, no correct combination of antecedent conditions and event conditions, no correct storm positioning and movement relative to the catchment, no correct hydraulic roughness of the conveyance system, no correct loss coefficients of the hydraulic structures, etc. etc. etc. – we can update the maps and we still will not address the fundamental uncertainties.
We could be better off to leave the floodplain maps alone and look harder at extending risks mapping into urban flooding areas.