Overland Flood Risk - From Flood Plains to Foundation Drains

From foundation drain to floodplain there is a continuum of drainage features that affect flood risks and damages in urban areas - these are micro lot level factors and macro neighbourhood factors.
River flood risks are defined by floodplain maps (called
engineered floodplains) because of the detailed hydrologic
analysis used to define flood flows and river and bridge
surveys used to define river and valley hydraulics to define
precise flood levels. Hazards in these zones are closely
regulated, new development is not permitted, and
 redevelopment is subject to special policies if allowed.

Yes, some property reference points could identify a few flood risk factors to address, but overland flooding risk is not apparent at the property scale - that is if we agree overland flooding is not a result of poor lot grading or obvious rain entry points.  Unlike other risks like fire hazards, the structure or property does not define most of the risk when if comes to flooding. It will be interesting to see what methods Aviva has adopted for defining risk levels and overland flood endorsement premiums.

Aviva is excluding high risk properties (about 5% of properties) considering flood plain maps that identify river (and sometimes lake) flooding hazards. Micro-scale factors like the property's foundation drain and service lateral condition can affect almost any property, but issues with those do not cause overland / surface water flooding.

Overland flow areas in an urban setting.  Areas beyond blue
regulated floodplains may be several hundred hectares in size.
I'd suggest the most sensitive overland flooding occurs upstream of large floodplain-mapped valleys and up onto urbanized 'table land' where the sewer drainage system and overland drainage system are not designed to accommodate runoff from major rainfall events.  Usually this would be in catchments draining a 125 hectare or smaller area in Ontario, lying beyond the regulated flood zones above where Aviva insurance would not be available. In these areas, neither the capacity of the storm sewer system, nor the adequacy of the overland drainage system (ideally on roadways and drainage easements) is apparent at the individual property scale.

The macro-scale, overland system is only apparent from runoff accumulation and concentration from 10's of hectares to even a couple hundred hectares of table land runoff that accumulates downstream. That is, a broad neighbourhood-scale macro factor representing many hundreds upstream of properties. The map of overland flow drainage areas (red text in yellow highlight) shows the continuum of drainage areas extending beyond regulated floodplains that flow into to smaller surface features that run through the urban lot fabric.

Engineered floodplain limits are determined through
extensive engineering analysis of the valley system.  More
local overland flow limits may be determined through a
local flood remediation study that analyzes the 'major system'
of road drainage as if it is a river valley. That can be
 expensive. Alternatively the risk of being close to the
overland flow path can be estimated by defining buffer
distances from small and large runoff area flow paths.
The key overland flood risk factor is whether the property is on the overland flow path where the drainage accumulates. Maybe a property is at a sag in a road where runoff could spill toward the property (i.e., it's vulnerable to overland flooding) but if there is no large upstream area contributing overland runoff toward the property, its not exposed to meaningful risk. But with a large overland drainage area from the neighbourhood flowing into the sag there is a risk to the property.  And in that case, disconnected downspouts won't help reduce risk - downspouts drain a couple hundred square meters of rooftop, while large overland catchments can be a couple hundred hectares (meaning 10,000 times greater than the local roof area)
GIS Tools can identify major sags, or depressions,
 (called sinks) in the landscape that can indicate or amplify
overland flood risk.

Neighbourhood scale risks factors for overland flooding can be assessed with some minor effort (relative to detailed pipe-by-pipe sewer and street-by-street hydraulic simulation models that is) by just considering topography from readily available elevations models and by applying core GIS hydrology tools.

The image to the left on 'major sags' is an example showing overland flooding risks including dwellings within poorly-drained sags.  In this case it includes dwellings upstream of a railway embankment that can impede and 'back up' flows during large events.

Ranked property risk considering proximity to overland flow
path, drainage area, and major sags in topography
 (i.e., drainage-challenged during extreme runoff events).
The risk to dwellings and property within close proximity to overland flow paths can be ranked. For example a building within 3 m of a 10 hectare area runoff flow path, or 15 m from 100 hectare area flow path, would be at relatively low risk. Alternatively a dwellings within  3 m of a 100 ha flow path would be at a high risk.  A building within both a flow path and in a sag would be at the highest risk. Remember, these are estimates and are qualitative.  Nonetheless, experience shows that areas identified through these heuristic methods have be subject to overland flooding during extreme rainfall events.

The issue with overland flood insurance (urban, table-land type flooding) is that risks are concentrated with a small portion of downstream properties for whom risk-based premiums could be unaffordable, or for whom coverage would not be available.  The majority of upstream properties would not likely ever experience overland flooding nor add coverage for it. So overland flood damages are not like wind damage (path of wind can be anywhere whereas path of water is always the lowest elevations).

At a property scale, a property with poor lot grading and a reverse slope driveway may be at high risk regardless of proximity to overland flow paths or major sags in topography. Such a property would benefit from having an overland flood endorsement in its water protection insurance. Ideally homeowners in this situation can take action to mitigate risks by improving grading, installing barriers to flow from the reverse drive, keeping grates clear of debris and perhaps installing a backflow valve on the driveway drain.  Beyond these exceptions, anyone with foundation drains that could clog or a service lateral that could become root infested is at risk of flooding - but that would not be overland flooding, but rather back-up from the floor drain or seepage from the foundation wall and cracks (a different endorsement all together).

Check out new analysis of table land flood risks and insight into correlations with basement flooding / sewer back-up incidents : overland flow flood risk factors

Be Logical About Rainfall Extremes And Flooding - Its What Mr. Spock (Leonard Nimoy) Would Have Wanted !

Mr. Spock favours facts
Insufficient facts always invite danger.
--SPOCK, Star Trek: The Original Series, "Space Seed"

Flooding from extreme rainfall is dangerous.  But if facts related to the cause of flooding are insufficient, the solutions to flooding will be misguided ... illogical.

As noted in an earlier post, when Environment Canada recently reviewed Southern Ontario rainfall statistics to see if there have been any measurable trends in rainfall patterns and intensities they found the following:

  • Significant increases, as well as decreases, were detected at some stations in a number of the extreme precipitation indicators.  However, the majority of station trends were determined to be non-significant and no consistent geographical patterns for increases or decreases were observed across Canada.  In most cases, the magnitude of the observed changes was also very small.
  • On sub-daily or short duration rainfall intensities of less than 24 hours (this is the data / or trends municipalities would use for planning, design and management of most drainage and stormwater systems that respond to flashy storms):  The majority of the trends were determined to be non-significant with no simple patterns or uniform rates of change evident in the short duration rainfall.  Trends were determined to vary with duration and regional location. 

Those are the facts.  The report is entitled "Methodologies to Improve Rainfall Intensity-Duration-Frequency (IDF) Estimates: A Southern Ontario Pilot Study", by Environment Canada, Adaptation and Impacts Research Climate Research Division, dated  December 2011.  The study was supported through the Natural Sciences and Engineering Research Council of Canada (NSERC).  It's 176 pages long.  Click for report. Mr. Spock, Leonard Nimoy (RIP), would be proud. More recently, Environment Canada had an article on extreme rain trends published in Atmosphere-Ocean that reiterated there are no overall trends in rain intensity - click for abstract.

Here are some insufficient non-facts: 

"The August 19th downpour was just one of 8 extreme rainfall events in the past 20 years that resulted in basement flooding and other damages. The August 19th storm exceeded the rainfall expected in a 1-in-100 year storm. Most of the other storms were estimated previously as having a 1-in-25 year to 1-in-50 year chance of occurring. Taken together, these events indicate that intense rainfalls are already on the rise in the Toronto area and that design standards for stormwater management need to be revised."

Mr Data favours data
This is from a report is entitled "Climate Change Adaptation in the City of Toronto: Lessons for Great Lakes Communities"  by Clean Air Partnership, dated 2008.  See Report Pages 19-20. The phrase "Taken together" takes the place of statistics, analysis and scientific conclusions.  The plural of anecdote is not data.  Environment Canada's analysis and conclusions indicated "... trends were determined to be non-significant with no simple patterns or uniform rates of change evident ...", while Clean Air Partnership's conjecture is that intense rainfalls are "already on the rise".

Once you have eliminated the impossible, whatever remains, however improbable, must be the truth.
--SPOCK, Star Trek (2009)

Dusty old data
It is possible that the proliferation of low cost, closely spaced rainfall gauges has increased the observation of intense rainfall events and improved technology has resulted in near-real time reporting of extremes in the 24-hour media.  But this does not mean that the intensities at any static location have increased - Environment Canada's analysis of individual rain gauge locations shows only very small, non-significant changes. Factually, in the municipality where I work we used to rely on Toronto's Bloor Street rain gauge statistics many decades ago when there were no reliable local records in the then township.  A decade ago we used a single rain gauge at the local airport to characterize rainfall - that's one gauge per 200 square kilometers. Today we have a dense rain gauge network with over 10 gauges covering 15 square kilometers each - this dense network observes more intense rainfall events because previously storm cells could readily pass around the single airport gauge available.

Decades old analog chart for tipping bucket rain gauge
Factually, extreme rainfall data were not reported in real time decades ago, and data used to be recorded with analog systems (rotating drum charts) that needed to be reviewed by hand, transcribed and digitized before being reported or compared to statistical data to report on frequencies.  Today, rainfall intensity and volume data are recorded and analyzed and compared to statistical data in real time.  Results over thresholds associated with local or widespread flooding are texted in real time and also emailed. This and 24-hour media contribute to a focus on extreme weather, observed more often due to the denser monitoring networks available today.

It is possible that infill and intensification in Toronto has increased runoff rates and volumes. Anecdotally, I need only to look around my house to see the residential intensification in the last 45 years: on my block the rear laneway has been urbanized (used to be gravel, now it is concrete with storm sewer), the two smallest homes across the street are now monster sized, the third smallest is now the vestibule for the massive new home in the old back yard, the semi I'm attached to has an extension on the back as do several others, smaller homes on large lots that had some greenspace have been replaced with multiple homes (includes the neighbours to the north of me, and the neighbours north of them, and the neighbours north of them).  The latter two homes now have 90% paved front yards for parking and many other homes have added or expanded front yard parking (more impermeable surfaces = more runoff). Factually, as part of flood remediation hydrologic model refinements, I have quantified increases in rooftop area of several percentages over 10 years in a GTA municipality using GIS data.  Additional driveway increases, not measured, would add more pavement. Over multiple decades infill and intensification increases runoff.

Example intensification, just the 8 addresses north of me since the late 1970's ... add it up.

It is possible that infill an intensification in Toronto has obstructed major overland flow paths, aggravating flood damages during extreme rainfall events. Factually, I have observed drainage easements encumbered by sunroom additions and landscaping such that drainage of low points in residential roads is compromised, and have observed overland flow paths encumbered by monster home rebuilds.  This occurs because overland flow paths (unless there is an old easement) are not regularly mapped or managed by municipalities, yet are prevalent in areas built pre-1980's when overland drainage was not engineered/designed to stay on right-of-ways and away from homes.

It is possible that removal of sanitary overflows for environmental protection has increase basement flooding risks by removing the "relief valves" from the system during extreme events.  See Ministry of the Environment's Procedure F-5-5Factually, the basement flood remediation Class Environmental Assessment (EA) Study for my neighbourhood (Toronto Remediation Area 32) identified the adverse impact of sewage overflow measures on basement flooding.  These measures built in the early 1990's were effective at achieving water quality improvement for small storms (see Council report - Eastern Beaches Water Quality - Effectiveness of the Kenilworth Avenue and Maclean Avenue Detention Tanks (Ward 26)). But the Kenilworth tank and related flap gates were cited in the Class EA as the cause of flooding (excerpt from the Class EA study):

The solution to flooding includes reestablishing sewerage overflows as highlighted in the causes and remedial measures summary for Area 32 flood cluster area 1:

The Toronto Area 32, flood cluster 1 shows environmental improvements implemented in the 1990's to reduce sewage overflows aggravated basement flooding during the extreme August 19, 2005 storm.  This is often referred to as the "law of conservation of poop" - poop either it spills to the river / or lake or backs up in the system and potentially into basements - it is an incompressible liquid and it has to go somewhere.  Area 32 is one specific example in my neighbourhood, but you can find many others.

Another example is in Hamilton, Ontario, in relation to their Ward One Combined Sewer Overflow Master Plan Class Environmental Assessment.  The 2003 report to the Committee of the Whole recommends implementation of overflow reduction measures to prevent pollution in Cootes Paradise Hamilton Harbour - it is also to address MOE Procedure F-5-5 control requirements.  The report notes the risk of interfering with the law of conservation of poop as follows in the analysis of Class EA alternatives on page 4:

"• Combined Sewer Overflow Regulator Adjustments
Adjustments to combined sewer overflow regulators to send more flow to the treatment plant can reduce combined sewer overflows, but must be carefully investigated to prevent basement flooding and overloading of the wastewater treatment plant."

The recommended solution for the Sterling outfall :

"For the Sterling outfall the recommended solution is a regulator adjustment, which would involve the raising of the existing weir within the combined sewer system to increase the amount of wet weather flow going to the wastewater treatment plant and reduce CSOs. This is a Schedule A project and may proceed to implementation."

Translation: Schedule A = no further study required as impacts are assumed to be small per the Municipal Class EA Process. The law of conservation of poop would tell us that reducing the overflow of sewage at the Sterling outfall by raising the height of the overflow weir will increase the volume of poop in the sewer system, will raise the level of poop in the system and can put basements at higher risk of flooding. It is not clear if those flooding impacts have been quantified - they may be low or they may be high, just like the Toronto Area 32 impacts. Google F-5-5 and see what other municipalities have tightened up their sewer systems to achieve pollution reduction but may also have unintended impacts.

The challenge in large municipalities is that there can be competing interests and programs.  And it is possible that right hands are not fully informed of what left hands are doing or of impacts of actions or inaction.  A good example of that is in Ottawa where Petrie Island beach closure in August 2006 were likely caused by a major unreported sewage spill at the Keefer regulator site adjacent to the John Street-Sussex Drive intersection.  The 2008 audit report is available online. Like in Toronto Area 32 or Hamilton Ward 1, the management of sewage spills will affect the system - too much regulation in Toronto and perhaps Hamilton can adversely affect basement flooding, while too little regulation as in Ottawa will cause sewage spills/pollution. Municipalities are caught between a rock and poop.

To sum it up:
  • There is factual, scientific analysis of rainfall extremes that shows no increasing trends.  There are also anecdotes (MR Spock's dangerous insufficient facts) about a lot of big storms that are turned into misguided conclusions - this is really like "Jenny McCarthy vaccine science".  
  • Without more intense rainfall, there are also many factual explanations for increased flooding in urban areas based on i) tangible, even measurable, runoff factors (infill and intensification), and ii) the regulation/modification of the underlying sewer systems to meet other often competing objectives like pollution prevention (F-5-5).
  • Insufficient facts could cause us to connect flooding solely with climate change, and cause us to more fully invest in green energy and carbon pricing to reduce greenhouse gases as a mitigation measure - or as Mr. Spock says 'invite danger' by trying to solve the wrong problem, and misdirecting focus and resources.
  • Scientific facts should cause us to promote greater runoff source controls for infill development, and to preserve overland flow paths in old neighbourhoods with easements, grading or infrastructure improvements.
  • Scientific facts should cause us to carefully trade off pollution control objectives and natural environment impacts with social and human impacts due to flooding, and to restore overflows where pollution controls have caused flooding (see Area 32 structural recommendations above).
The world is full of compromises. Infill and intensification builds compact, livable communities that can be effectively serviced by public transit, promoting a healthy environment ... but there are runoff impacts. Regulating sewers to keep beaches swimable improves communities and benefits aquatic environments by treating small storms ... but can adversely affect sewer operation and cause flooding during extreme storms. 

The world is also complex and requires sufficient facts to make informed decisions on important topics. This is especially so when the facts about problems influence the billion dollar decisions on solutions.

Why We Cannot Predict Rainfall Extremes - Chaos, Lies and Butterlies

Edward Lorenz
How accurate are weather forecasts?  Not very.

Call it chaos or call it the 'butterfly effect' - http://en.wikipedia.org/wiki/Butterfly_effect - it is "the sensitive dependence on initial conditions in which a small change in one state of a deterministic nonlinear system can result in large differences in a later state."

This is the name given by Edward Lorenz (23 May 1917 – 16 April 2008), an American mathematician, meteorologist, and pioneer of chaos theory.  He was not consulted the the Weather Gone Wild authors, or others who hang their hat on faulty science .. maybe because he is dead and because science is secondary to crafting sensational journalism.

In practice the butterfly effect means we can't predict future extremes (like rainfall) that cause flooding.  Nope. All we can practically predict is weather will regress to its mean, but highs and lows we won't know.  Per the Wikipedia butterly effect article (or anyone who plans picnics based on weather):

"Recurrence, the approximate return of a system towards its initial conditions, together with sensitive dependence on initial conditions, are the two main ingredients for chaotic motion. They have the practical consequence of making complex systems, such as the weather, difficult to predict past a certain time range (approximately a week in the case of weather) since it is impossible to measure the starting atmospheric conditions completely accurately. "
Example 1-3 month precipitation forecast

In practice do we just predict weather to be average?  Yes. Environment Canada has predictions months in advance.  Here is the 1-3 month precipitation forecast:

 And the 10-12 month forecast:

White on the map means precipitation will be 'average'.  Or as Environment Canada puts it on the main page (http://weather.gc.ca/saisons/prob_e.html):

Example 10-12 month precipitation forecast 
"The occurrence of the white colour areas over Canada is more predominant for precipitation forecasts, which are generally less skillful than temperature forecasts, and increases with forecast range at longer lead times. The increasing occurrence of areas coloured in white over Canada implies diminishing ability of the forecasting system to make reliable and accurate predictions in these regions."

Look at the example to the right - ITS ALL WHITE, meaning in 10-12 months we can expect average precipitation  .. that is recurrence, trending back to the mean.  Any extremes get washed out in the predictions.

What about short time periods within the predicted month? Well, climate models do not have the temporal precision to drill down to minutes and hours because they have daily times steps.  Some day they likely will simulate smaller time steps, but don't confuse precision with accuracy - extra decimal places that infer precision in the output do not mean the overall number is close to reality.

So you have to ask yourself how you can predict changes in 5 minute and 1 hour extreme rainfall intensities 50 years from now with any accuracy. Climate modellers will suggest that you can but reputable scientists, like professor emeritus Edward Lorenz, would tell them its just lies and butterflies.

"As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality."  — Albert Einstein, Sidelights on Relativity (1920)

Vaccines and Rainfall Extremes

What do vaccines and rainfall extremes have in common? Both have been victims of bad science when it comes to public education and media reports. And this can lead to bad policies and misguided actions that fail to address real problems.

Recently the outbreak of measles has led the media and medical professionals 'call out' those falsely linking increased autism to vaccines through their anecdotes, beliefs and unscientific arguments. Similarly, increased flood damages have been linked to more extreme weather - more intense storms - due to climate change.  But what do the statistics of extreme rainfall show? How does the scientific data compare to the media hype? Data in Canada (see links at the bottom of this post to maps, charts, tables, etc.) show no increasing trend in extreme rainfall.

But its hard to argue with pictures like a stranded GO Train in the Don River Valley, and suggest that flooding isn't increasing.  That is unless you realize that the river outlet - the Keating Channel - was designed with a low capacity and was the subject of a flood inquiry in the early 1980's.  Keating Channel flood inquiry report.  If it is not dredged it gets clogged with sediment and backs up.  Its not news, its fluviogeomorphology and hydraulics.  TRCA has a nice summary of the century-old problem:

"Since its construction between 1914 and 1922, the Keating Channel has been subject to heavy sediment loads, requiring regular dredging to maintain sufficient depths to allow for and maintain shipping activities at the mouth of the Don River. Between 1950 and 1970, widespread development throughout the Don Watershed and the construction of the Don Valley Parkway increased sedimentation rates by up to four times that of the pre-was era. After 1970, decreases in the number of new watershed disturbances and improved sediment control structures likely contributed to the decline in sedimentation in the Keating Channel to levels similar to the pre-war era. A reduction in shipping activities within the Keating Channel, combined with restrictions on the open water disposal of dredgate imposed by the International Joint Commission (IJC) in 1974, resulted in a cessation of dredging in the Keating Channel. In the following five to six years, the Keating channel filled with sediment and debris to the point where it became visible under all but high lake levels, resulting in increased flood risk along the lower Don."

Because of the low capacity, engineering solutions have been built or proposed like the Lower Don Flood Berm, or the Don Mouth Naturalization Project (DMNP) that just received provincial EA approval (see file EA 03 03 02). Of the DMNP TRCA: "This project will transform the existing mouth of the Don River (the “Don Mouth”) including the Keating Channel, into a healthier, more naturalized river outlet to the Toronto Inner Harbour and Lake Ontario, while at the same time removing the risk of flooding to over 290 hectares of urban land to the east and south of the river." To see how the Keating Channel system was designed to spill all over with limited capacity, check out the hydraulic simulation video above that shows - this is the widespread flooding based on today's design and climate conditions.

In a separate post we explain that the GO Train flood was an operational mishap and not an extreme event - record rainfall at Pearson Airport in the Etobicoke Creek Watershed on July 8, 2013 did not hit the Don River Watershed where the GO Train was stranded, and where the river peaked at a measly 80.7 m, lower than the May 29, 2013 weeks earlier, and lower than the river's one-in-five-year design flow.

Back to rainfall extremes.  Environment Canada's Adaptation and Impacts Research Climate Research Division recently reviewed Southern Ontario rainfall statistics to see if there have been any measurable trends in rainfall patterns and intensities.  These guys are no dummies. A brief summary for those without time to read all 176 pages:

  • Significant increases, as well as decreases, were detected at some stations in a number of the extreme precipitation indicators.  However, the majority of station trends were determined to be non-significant and no consistent geographical patterns for increases or decreases were observed across Canada.  In most cases, the magnitude of the observed changes was also very small.
  • On sub-daily or short duration rainfall intensities of less than 24 hours (this is the data / or trends municipalities would use for planning, design and management of most drainage and stormwater systems that respond to flashy storms):  The majority of the trends were determined to be non-significant with no simple patterns or uniform rates of change evident in the short duration rainfall.  Trends were determined to vary with duration and regional location. 
Temperatures have gone up in recent years - here's a graph at right showing global temperatures increasing.  The internet is full of these, and it is undeniable that temperatures are increasing.

And insurance losses have been increasing in Canada.  You'll have no problem finding graphs and statistics showing the number and size of weather related claims increasing (see below).

But are there any statistics of increasing rainfall intensities?  Apparently Environment Canada does not have any.

But this has not stopped the media from reporting increases as facts.  This may be because the media confounds future predictions with historical trends.

The CBC's Weather Gone Wild documentary
was written by Helen Slinger and Melanie Wood.  To carry on the vaccine science analogy, they are the Jenny McCarthy's of meteorology.  They cite that every degree of temperature rise means 7% more moisture in the air, implying this means more extreme weather (more rain?).  If this linkage is true, how come the rainfall intensities are not increasing like the temperature graph does?  In other words, why does Environment Canada not have a parallel graph showing rainfall intensities increasing too? Jenny McCarthy should consider autism rates were increasing while MMR vaccination rates were decreasing - no correlation.  Helen Slinger and Melanie Wood should consider temperatures were increasing but but rainfall intensities were not - no correlation.

Linking insurance losses to climate change rainfall impacts can cause a province like Ontario to focus on distractions instead of solutions; for example cause it to embark on an expensive green energy program that hurts the economy and consumers.  But what if there have been no increases in extreme rainfall statistics - just extreme weather reporting?  Would the push for green energy be so strong? Would there be as strong a support for GHG reductions?  Yes, there are legitimate reasons to mitigate temperature increases and the effects on our environment, but so far, increased rainfall intensities is not one of those effects (not scientifically measurable). Insurance industry claims have been dismissed based on CBC's consultations with Environment Canada.

Do images of flooded GO Trains support the need for carbon taxes / carbon pricing so that we can tackle supposed rainfall increases?   Maybe we should have a carbon tax, but please use the money to dredge the Keating Channel instead of subsidizing solar panels, and maybe improve some other 100 year old infrastructure that still can't handle today's weather.

Models vs Data - WHat has COVID-19 taught us?: https://www.cityfloodmap.com/2020/05/what-covid-19-taught-us-about-observed.html

More on the 1981 flood inquiry:

More on why the GO Train flood on July 8, 2013 was completely avoidable:

More on why cognitive biases that hamper rational though on flood causes:

Extreme rainfall trends in Canada (Environment Canada Engineering Climate Datasets):

Static Maps: http://www.cityfloodmap.com/2015/12/severe-storm-trends-canada-rainfall.html

Interactive Map: http://www.cityfloodmap.com/2015/12/canadian-extreme-rainfall-map-climate.html

Table Summaries: http://www.cityfloodmap.com/2015/12/canadian-extreme-rainfall-summary-by.html

Chart and Table: http://www.cityfloodmap.com/2015/12/top-weather-story-in-canada-2015-less.html

Long-term Station Table: http://www.cityfloodmap.com/2015/12/long-term-climate-change-short-term.html

Environment Canada Denies Changes: http://www.cityfloodmap.com/2015/10/bogus-statements-on-storms-in-cbcnewsca.html

Contradicting Insurance Industry Claims: http://www.cityfloodmap.com/2015/12/trends-in-canadian-shortduration.html