This post summarizes risk factors affecting urban flooding and explores the example of flood risk in the City of Waterloo, Ontario.
Two key factors explain basement flooding risks in many urban areas:
1) sanitary sewer design practices, and
2) overland flow design practices.
Why?
Virtually all urban properties have gravity-drained sanitary sewer connections to the municipal sanitary sewer systems, and this collects wastewater from homes as well as infiltrated groundwater from foundation drains in most pre-1980 areas and occasionally direct rain and melt water inflows thorough illicit collections to the home plumbing and drainage systems and ultimately the municipal sanitary sewer system. Because of this connection, any surcharging of municipal sanitary sewer systems during extreme weather can back-up into low-lying floor drains, flooding basements.
So the capacity of the municipal sanitary sewer system will partially define basement flooding risk. Design standards in Canada have evolved over time as described in a
previous post. While each municipality is a little different, we can consider 1975 as a year in which systems became fully-separated, with no more foundation drain connections that serve to overwhelm the system with infiltration and, more importantly, provide a pathway for illicit inflow connections, like from rooftops or other property drains (in York Region we once even found an outside kitchen sink connected to foundation drains - it was near the garden and used to rinse vegetables!).
Overland drainage began being considered in urban drainage design in the late 1970's - the former Town of Markham's design standards recognized overland 'major' system design requirements in 1978, under the guidance of University of Ottawa's Dr. Paul Wisner. Many other municipalities in Canada adopted dual minor-sewer/major-overland drainage design standards throughout the 1980's. Historical development grading and old subdivisions that did not integrate overland flow are prone to flood stresses due to i) water entry into building openings via windows, doorways, recessed walkouts/stairs, and reverse-sloped driveways, ii) storm sewer surcharge that backs up into foundation drains and through basement walls and under flood slabs, and iii) sanitary inflows into maintenance hole lids (e.g., at roadway locations with deep ponding over the lids pick-holes and edge). The insurance industry refers to overland flooding pluvial flooding, an unheard of term in Canadian engineering design (this is to distinguish between urban overland flooding and 'fluvial' riverine flooding that occurs in valleys).
Show me !
The City of Waterloo has an extensive
Open Data portal that includes information on sanitary sewer installation date. This GIS data has been used to characterize neighbourhood flood risk according to era of construction and engineering design practices.
Overland flow risks can be mapped in many ways with increasing complexity on aspects of:
i) Input Data - e.g., elevation model detail and conditioning as input to the hydrologic and hydraulic analyses can be based on coarse provincial datasets (raster cell sizes suitable for macro-scale neighbourhood assessments), local datasets such as detailed 3D breaklines used for other image rectification (raster cell size of a metre or two for master drainage planning), to LiDAR datasets (to generate sub-metre cell size for fine-scale lot-by-lot, or gutter-by-gutter analyses),
ii) Defining Risk Zones / Hazard Area - e.g., this can involve the simple delineation of flow accumulation paths and definition of sinks (ponding areas), to setting of buffers around flow paths based on drainage area size (a surrogates for hydrology and hydraulics but good for screening), or more advanced flow spread calculations (i.e., applying hydrologic and hydraulic principles) to identify risk zones.
Data to the above can include province of Ontario processed topographic data (through
Land Information Ontario (LIO)), including a conditioned elevation model and flow direction raster grid that has been used to map overland flow paths and spread across much of the province.
Examples:
Simple Flow Path and Ponding (Sink) Delineation: My City-wide Storm System Master Plan for the City of Stratford in 2004 was one of the first applications of major drainage system / overland / pluvial flood risks using ESRI's Spatial Analyst and the emerging hydrology tools (that would later become the familiar ArcHydro tools), and first introduced by the University of Texas as an extension to ArcView 3. The following map illustrates the assessment of overland flow path drainage issues and ponding issues. No base data was available for the analysis and the elevation model was derived from half-metre AutoCAD contours to generate a 2-metre DEM raster for analysis. The integrated GIS-modelling approach was subsequently presented at the 2004 AWRA conference in Nashville, Tennessee.
 |
| Stratford City-wide Storm System Master Plan - Major Overland Flow / Pluvial Flood Risks Based on GIS-based Flow Paths Delineation and Ponding Areas using ArcView GIS Spatial Analyst Extension. |
Buffered Flow Paths and Ponding (Sink) Delineation: A similar approach was taken in Markham, Ontario in 2013 to conduct a screening-level identification of properties in close proximity to flow paths or within potential ponding areas. This was shown in a
previous post. The images below illustrate some of the outcomes that were subsequently aggregated over catchments to identify areas for detailed study. In this example 3D breaklines from a recent orthophoto rectification were used to generate the DEM raster within the city - this was integrated with a more-coarse elevation model outside of the city boundaries to ensure a complete watershed delineation. The final DEM was refined after extensive manual editing of the 3D breaklines and reprocessing of overland flow paths and ponding areas/sinks.
 |
| Overland flow / pluvial flooding risk defined by buffers on overland flow path as a function of drainage area. |
 |
| Overland flow / pluvial flooding risk defined by buffers on overland flow path and ponding with building pluvial flooding risk risk estimated by proximity to flow buffer or to ponding area.. |
Hydrologic-Hydraulic-Based Overland Flow Paths: Analysis of
City of Toronto
overland flood risks was completed in 2015 using a pre-conditioned provincial DEM - as it is conditioned it cannot be used to generate ponding limits. Simplified rational method hydrology was applied considering individual cell-by-cell time-to-peak and individual 100-year design rainfall intensities, along with a standard runoff coefficient. Overland hydraulics to define flow spread were applied on a derived vector-based overland flow network that considered 100-year flow along each overland reach and flow spread defined by longitudinal slope and uniform flow conditions for a typical roadway cross section. The presentations below illustrates the overland flood hazard / flow spread that was then used to explain the location and density of reported basement flooding during recent extreme rainfall events.
Refined Hydrologic-Hydraulic-Based Overland Flow Paths: The Toronto-based overland risk mapping approach was refined using SOLRIS land use classification to derive cell-by-cell weighted rational method runoff coefficients, for a more precise hydrology. This was required as both rural and urban areas across south-west and central Ontario were assessed. The analysis was completed in 2016 as summarized in
a previous post. The result is an overland drainage network with over 800,000 flow segments (reaches) with an individual 100-year design flow rate and flow spread. A snapshot of the analysis is shown below.
 |
| Ontario Overland Flow / Major Drainage / Pluvial Flood Risk Assessment |
This last overland flood risk analysis approach is used to help assess City of Waterloo flood risks. The map below shows flow paths in the western part of the city and and highlights buildings (in red) that intersect the overland flow path - in this analysis flow paths with 3 hectares of contributing drainage area (i.e., 30,000 square metres or more) are shown. The presence of modern stormwater management and drainage design, as suggested by the municipal stormwater ponds in the western-most areas, would mitigate the possible impact of these overland flow paths by capturing and controlling the release of major flow during extreme events. In addition, modern minor systems in these modern, post-1980 subdivisions may be designed to capture and convey runoff generated by extreme rainfall.
 |
| City of Waterloo - Example Overland Flow Risk (Urban Major Drainage / Pluvial Flood Risk) - Buildings along Flow Path Highlighted (Surface Flooding and Sanitary Inflow Risk) |
Multiples of the 100-year flow spread are shown for catchments of 3 to 1000 hectares. For larger areas, only the flow centreline is shown and those assessing valley-feature overland flood risk should refer to regulated floodplain limits that are determined through more advanced hydrologic and hydraulic analyses.
The next map shows installation date of sanitary sewers with pre-1975 sewers shown red (highest risk for infiltration and inflow stresses during extreme weather), 1975-1989 sewers shown in orange, and post-1990 sewers shown in green.
 |
| City of Waterloo - Sanitary Sewer Installation Date (Inflow and Infiltration Risk) - Pre-1975 sewers (red), 1975-1989 sewers (orange), 1990 and newer sewers (green). |
The map suggests that sanitary sewer replacement has occurred in the older core ares to the east (new green sewers surrounded by older red sewers).
This next map illustrates the intersection of overland flow path attributes onto sanitary sewer features that they intersect. Specifically the drainage area is assigned to each sewer segment it crosses and the sum of the intersected overland flow is aggregated to each segment and then weighted by the age of the sewer - post 1990 sewers have the area reduced by a factor of 5 considering modern drainage design and low infiltration and inflow stresses in modern fully-separated systems, while 1975 to 1990 sewers have the area sum divided by a factor of 2 considering lower fully-separated systems stresses. This is an approximate screening method, of course, but consistent with industry understanding of risk factors based on more detailed studies. The width of the red highlighting surrounding sanitary sewer segments illustrated thee age-factored sum of intersected flow area.
 |
| City of Waterloo - Overland Flow Impact on Sanitary Sewer Systems - Intersection of Major Drainage Flow Path Areas To Sanitary Sewer Segments, Factored by Age of Construction. |
Red highlighted areas are or interest for further study. It is clear that in some core areas with predicted flood risks, sanitary sewer replacement has already occurred (i.e., newer green sewers in eastern areas), meaning that some flood risks may have already been mitigated.
The last map adds average age of dwelling construction in census areas. Clearly, the is a strong correlation to the sanitary sewer age risk factor and overland drainage design risk factor and the average age of construction. It is interesting to note that the broad, census-area neighbourhood risk does not account for local sanitary sewer upgrades, nor does it help identify individual properties that are at risk of significant overland flooding, as those buildings are isolated to the major overland flow path hazard area.
 |
| City of Waterloo - Urban Flood Risk Factors and Average Age of Dwelling Construction |
