### Weather Gone Wild - New Weather Math for Shockumentaries

Weather Gone Wild does a great disservice to statisticians by promoting some 'new' math related to extreme weather.  Its like Romper Room math.

The Ocean Breeze resident Frank who has lived there 43 years was interviewed.  He was told by a "geologist" that Hurricane Sandy was a 1 in 700 year storm.  And Hurricane Irene before was a 1 in 300 year storm.  Spring flood the year before 'roughly' 1 in 100 year storm.  His conclusion is that in three and a half years they had 1100 "years of storms"!

There you have it folks!  The new normal is here.  More like the new weather math is here! Does anyone remember what geologists do for a living? I didn't think it was meteorology or extreme weather statistics.  I bet kids in Romper Room even know the difference.

Here's how the math really works on extreme weather return periods (new weather math that Weather Gone Wild promotes):

1) never add the years of storm return periods ... those numbers are denominators of fractions that represent the yearly probability of an event occurring.  Just to show how bad that "weather math" is, imagine in Romper Room they teach some "pie math" to the kids.  Teacher says little Billy has 1/2 a pie, Sally has 1/4 of a pie and Joey has 1/8 of a pie ... together the teacher says they have 2+4+8 = 14 pies ... IT IS SOOOO WRONG! Wrong. wrong. wrong, wrong, wrong. Never add the denominators of probability fractions!  Its like Ghostbusters crossing the streams from proton packs!

2) Hurricane Sandy has a return period of 700 years which is not rare for a long time resident meaning:

- chance of Sandy occurring in any given year is 1/700, or 0.14%   ... its about the same probability of OLG's Pick 4 12-Way Box Play (all four digits must match the digits drawn in any order, where two of the digits are the same), which is 1 in 833 odds (OLG odds)
- chance of Sandy not occurring any 1 year is 1 - 0.14 = 99.86%
- chance of Sandy not occurring in the 43 years Frank lived in Ocean Breeze = (99.86)^43 = 94.04%
- chance of Sandy occurring in 43 years = 1 - 94.04 = about 6% .. or 1 in 17 chance  = not rare
- "the old extremes are the new normal"... quote me ;)

3) Never conflate the probabilities of different processes like hurricanes and spring floods like Frank does. Even though these are both weather/water events, they are statistically independent processes.  You might as well say your well pumped failed, say a 1 in 40 chance event and then add 40 to the 1100 years of storms.  You might as well add you just won Pick 4 12-Way Box Play so add another 833 to the total based on those odds.  And you bought an orange car which is a 0.4% or 1 in 250 chance (popular car colours) so add another 250 to the odds.  Yes it gets silly adding the probability denominators for different events like car colour, but that is what adding spring flood and hurricane denominators is like.  I bet spring floods were influenced by snow pack, melt cycle, perhaps ice build-up at bridges, joint probabilities of melt plus rain, etc... many things not related to hurricane processes.  And I'd have a better chance of winning that bet than the Pick 4 12-Way Box Play.

Ocean Breeze residents, it was said in the doc, are faced with the choices of "build back better or retreat".  Uninformed decisions are made based on bad math that underestimates risks.  Did Frank's geologist friend tell him how rare a Hurricane Sandy event would be if he settles there (did the geologist say 1 in 17 chance in 43 years (correct math), or it won't happen for 700 years (bad weather math)? Let's build better math skills and weather documentaries to inform the public about risk... let's retreat from sensationalized weather reporting, and "shockumentaries".

### Weather Gone Wild Documentary Review : Media Gone Wild Folks!

Note this is Part 1 - Media Gone Wild Folks! ... why rainfall is not more extreme.
Part 2 will share insight to why (if not more extreme rainfall) we are experiencing record extreme storm "obervations" and undeniable flood damages/insurance claims in urban areas. You'll read things I have not heard in the media before. Update - I've added a bullet summary list of these reasons at the end of Part 1.

Doc Zone Intro:
Doc Zone has a forecast - Weather Gone Wild - the narrator says it is not your imagination - weather has taken a turn for the extreme and the unpredictable.  It cites: Spring blizzards across the Maritimes;  flash flooding in Toronto; raging rivers in Manitoba; horrendous heat in Australia; record showfall in the Middle East; Hurricane Sandy.

My take:
I was ready to be disappointed by the documentary. I heard about the Doc Zone's Weather Gone Wild on my Friday commute to work - a job where I am responsible for delivering a long-term, \$234 Million,  municipal flood remediation program for a GTA municipality.  John Moore - a good man on most mornings - played some snippits from the producer (I think) who suggested that anyone who denies the weather changes is crazy.  So I was ready to be offended because I regularly encounter those who think the plural of anecdote is data and who are adverse to scientific rigor on the topic of extreme weather. Please keep in mind the comments below are geared to rainfall changes (where I am a skeptic) and not temperature changes (where I am a true believer).

Get ready for some thought provocation and debunking!

Who and what is really crazy?
Let me share some facts and other considerations that are not in the media and are not in Doc Zone's Weather Gone Wild, so you, intelligent reader can be the judge of what is crazy: i) sensational media reporting that does not dig deep into subject matter, or ii) me, a licensed Professional Engineer who has worked an entire career immersed in meteorological statistics (rainfall, snowfall, snow pack, snow melt, rain + snowmelt),  hydrological statistics (flood flow, drought flows, flood levels), municipal flood remediation studies to protect basements and properties (sanitary sewer, storm sewer, combined sewers, overland flooding analyses), and riverine flood remediation studies (flood hazard mapping, flood risk reduction strategies ranging from small systems (Cavan Creek in Peterborough .. it looks like a ditch) to large systems with complex reservoir systems (Mattagami River upstream of Timmins)).  Don't answer yet on who is crazy and never mind my resume .. just read some facts below on the topics I can hopefully enlighten you on.

More Doc Zone intro:
Says one interviewee : "There's a bullet that's a comin'.  Are ya gonna stand there and let it hit ya right in the forehead?" So the narrator adds "The future is here.  Are we ready?"

Is this not sensationalizing the topic of weather?  Comparing getting hit by weather with getting shot in the head? That's what Doc Zone does.  What this blog post does is deliver some facts on some weather and flooding topics where I can bring 24 years of engineering practice and 2 engineering degrees into the conversation.

More Doc Zone set up:
Narrator: "It was as if a water bomb smashed onto Southern Alberta" and says "a massive weather system ..exploded into three days of torrential rain." Cue the creepy piano and cello music.

Is this not sensationalizing the topic of weather?  First guns now bombs?  I enjoy an good enlightening environmental documentary, Trashopolis, etc. those producers look for entertainment value in the content too but not sensationalizing it like Doc Zone does.

More Doc Zone:
Images flying over flooded Alberta.  Narrator says this is High River, Alberta and nearly the entire Town is under water and communities are in a state of emergency.

Who is surprised that the community of High River is underwater?  Where did the town get its name?  I was expecting Captain Obvious to step into the frame here. Would you expect the seaside town of Windy Shores to be immune from extreme gusts? Droughts in Dry Gulch, Arizona? The real story here is that communities settle near water for many reasons - drinking water supply, transportation, power generation, support for industry - this is what High River did a long time ago, it flooded then and was adptly named - flooding should not be a surprise.  Sh*t "happens". The pre-flood conditions of a late spring melt, saturated soils and then a good storm "happens"... statistically that is.  In Ontario, Hurricane Hazel's 212 mm of rain was preceeded by a wet week (in the business we call this Antecedent Moisture Conditions, or AMC) and it can be readily accounted for in rainfall-runoff simulations used to derive extreme flow predictions that define flood hazard zones - Google it - students, engineering younglings-in-training, and crusty engineering geezers are applying these principles every day.  It is also no surprise that Calgary flooded - parts are built and have intensified in flood hazard zones - this is a political failing not Weather Gone Wild, rather inadequate provincial and municipal leadership in Alberta (political leadership can be short-sighted where the time horizon necessary to plan for and mitigate extreme flood risk extends beyond the election cycle). Ontario is better off with its flood hazard management policies since Hurricane Hazel, and many municipalities are investing in long term flood mitigation (riverine and sewer based) and are committing to funding for stormwater management initiatives.

Do you want to see if your house or business is in a floodplain in Ontario?  It's not Weather Gone Wild .. it is expected so the nice engineers have mapped it out for us.  Try this site:

Conservation Ontario Interactive Flood Maps

When these flood hazard zones experience flooding, get ready for the media to be surprised.

More Doc Zone:
"Welcome to the new normal". Melancholy piano.

What is normal?  Is extreme weather normal?  Yes.  Is extreme weather unusual? No. When engineers, climatologist, and hydrologists analyze rainfall and flood flows we use 'extreme value statistics'.  We don't use the normal bell-curve that helped journalists and doc film makers pass high school math.  My personal go-to extreme value distribution is Gumbel - not a "new normal", but a distribution Gumbel derived in 1954 and which is commonly used for Environment Canada's rainfall statistics.  Interestingly, Environment Canada's Adaptation and Impacts Research Climate Research Division recently reviewed Southern Ontario rainfall statistics to see if there are any trends in rainfall patterns and intensities.  A brief summary for those without time to read all 176 pages:

• Page 76-78 discusses trends in Intensity-Duration-Frequency, i.e., rainfall intensities.  Here are some highlights under extreme daily and multi-day precipitation indices since the early and mid-20th century.
• The indicators included:
• highest 1-day, 3-day, 5-day, 7-day and 10-day precipitation, rainfall and snowfall
• days with precipitation, rainfall, snowfall exceeding 90th, 95th, and 99th percentiles
• days with precipitation ≥ 10 mm, ≥ 20 mm and ≥ 50 mm
• Conclusions: 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):
• A number of studies have conducted a trend analysis of Environment Canada’s short-duration rainfall intensity data.
• Conclusions: 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.

Translation: the new normal is basically the same as the old normal, but with the variations that we expect in a natural systems, characterized by averages and also by extremes.  Environment Canada's best and brightest have nothing to show that rainfall is becoming more extreme.  THIS IS THE AUTHORITY ON CLIMATE IN CANADA TELLING US THIS.  Do the Weather Gone Wild producers, writers think Environment Canada is crazy?  If you lean toward conspiracy theories and think it is a cover up I think that is wrong.  What is wrong is that those "in the know" do not speak as loudly and often as those who want to share another story.

More Doc Zone:
The documentary jumps from historical trends (that Environment Canada has shown are up or down and non-significant for short term < 24 hour duration rain that affects urban drainage) and lands on predictions.  Gordon McBean, President International Council for Science cites every degree of temperature rise means 7% more moisture in the air.  Narrator says in 2050 we can expect double the number of extreme weather events.

Please do not confuse "hot and muggy" with extreme rainfall.  I measure humidity in my house ... its harder to humidify my acoustic guitars in the winter when it is cold and there is less moisture in the air - that's a scientific fact.  And temperature trends are undeniable across Canada - say at the Pearson Airport weather station the 10 hottest years were observed in the last 15 years.  So how come Environment Canada cannot show rainfall trends to support the Doc Zone corollary that 1-degree = 7% more moisture extremes to bomb us with or shoot at us?

I have looked closely at some of the Toronto Environment office study results widely reported in the Toronto Star that predicted double the rainfall intensities for some events (their frequency, or probability of occurrence is called 'return periods').  I observed that the 10-year return period, 1-hour rainfall intensity was predicted to double.  I then compared the existing 10-year return period, 1-hour rainfall intensity to the 'old normals', i.e., the Pearson Airport rainfall statistics that I keep in my top drawer of my desk.  I found the existing intensity was under-reported by half, meaning the future doubling of intensity was meaningless for that event.  Did anybody else check?  I was at a seminar shortly after and asked senior professionals at Toronto and Region Conservation Authority and Toronto Water if they checked / were involved and the answer was no.  Maybe that has changed but I remain skeptical when studies under-report today's rainfall by half and then predict that it doubles.

More Doc Zone:
Interviewee, University of Waterloo, says we are 1 or 2 catastrophic events away from really bringing cities on-board across Canada to realize the dire situation we are in.

At this point it is useful to distinguish between riverine flooding (valley flood plains and bridges) and provincial roles and urban flooding (road and basements) and municipal roles.

On riverine flooding, are the townsfolk who named High River surprised at the dire situation?  Are you in one of the Ontario Conservation Authority flood zones shown in the interactive maps at the link above?  Is your home/business in the flood plain in Calgary?  Risk are predicable, mapped and actionable already - and requires strong provincial policy and municipal implementation strategies to mitigate risks.  The Ontario model is working: development is prohibited in the flood plain, and existing development can continue under special policies in some cases, however risks are mitigated through flood proofing of structures and ensuring safe access.

On urban flooding municipalities have been implementing flood reduction programs since the 1970's.  This is 'bread and butter' work for the engineering profession.  But there is a long, long, backlog and infrastructure funding deficit that prevents municipalities from doubling the size of all the sanitary and storm sewer networks.  The post-WWII building boom in Ontario happened under an old set of design standards whereby sanitary sewers were prone flooding (i.e., groundwater infiltration from foundation drain connections, and inflows from connected roof downspout), and storm sewers were size to handle 2 to 5-year return period storms with no allowance for overland flow relief on roadways and drainage easements/channel.  Moderately sized channels that did not regularly flood where encroached upon with buildings and parking lots or piped entirely into storm sewers.  No surprises here - the old normal is the new normal in developments built pre-1980.

More Doc Zone:
Narrator notes that Canadian insurance claims related to weather have increased, exceeding 1-billion for several years and in 2013 were 8 billion.  And these damages will rise with temperature.

Are there other explanations for increased flood damages than more extreme rainfall?  See Part II of this blog post for my thoughts.

More Doc Zone:
Narrator takes us around the world to show us hurricane surge walls in New York, floating housing complexes in the Netherlands and in Toronto, they are "rushing" to fix infrastructure in the most vulnerable of neighbourhoods.  The adaptation strategies shown are backflow valves for sanitary sewers, river diversions and 'retreat'.

Why does Doc Zone  show a Lake Ontario breakwall in Toronto's west end? So the viewer can be fooled into thinking that it is a surge wall too, like New York?  Do we have a surge risk like New York during Hurrucane Sandy? No.  Lake shorelines are managed for risks considering water levels (which are regulated for Lake Ontario), and also for short term wind related waves.  These risk zones are managed like river flood hazard zones in Ontario.  Look at the coasts if the topic is ocean surges like in New York or the Netherlands.  Mixing Toronto's sewer upgrade programs with other topics is misleading and irrelevant.

As for 'rushing to fix' infrastructure, work is ongoing since the 1970's.  My first engineering summer job in the late 1980's was related to basement flooding in Toronto's North York due to storms in 1986.

More Doc Zone:
Narrator says after the 2013 floods, Albertan's were thrust into climate change adaptation.

Wait. Rewind please.  High River flooded.  Lands in the Calgary flood plain flooded.  Everything worked according to the inadequate plans in place.  Where did climate change get weaved into this story?  See Environment Canada report on rainfall trends.  Development and natural hazard management policies are inadequate in Alberta.  Rainfall extremes do happen.  Rainfall on saturated soils do happen.  Extreme flood flows do happen.

Tangent alert:. Did you ever see the documentary Heart of Darkness about the making of Apocalypse Now?  There is a great quote ""We were in the jungle, there were too many of us, we had access to too much money, too much equipment, and little by little we went insane".  The same holds true for climate change researchers.  All the research money is flowing into climate change research and they could be going crazy, or making documentaries.

Conspiracy alert: Doc Zone describes Waterloo's Mr. Feltman as a concerned parent and shows him walking over a river with his family.  It should be noted that his centre at Waterloo has received funding from the insurance industry to promote climate change adaptation:

Intact Insurance Funds Waterloo Climate Adaptation

The insurance companies want municipalities to upgrade flood-prone sewer systems build before 1980.  It costs them a ton of money in damages.  They will fund research into adaptation that pushed municipalities to upgrade their systems.  So climate change rainfall is a very convenient scare tactic for insurance companies to use to promote this.  Environment Canada did not show increasing trends in rainfall intensity.

Doc Zone Weather Gone Wild has conflated several topics and issues worth commenting on:

• ocean hurricane surges are not relevant to urban infrastructure flooding (sewer back-up or flash floods on roads)
• rivine flooding related to poor development policies (identified hazards in Alberta) is not related to urban infrastructure flooding
Increasing reported damages due to flooding can be explained by a list of explanations none of which is increased rainfall intensities due to climate change :

• intensification in Alberta valleys/floodplains - more development = more losses
• intensification over urban sewer infrastructure - more development = more runoff into the same size system (yes there are now stormwater controls for large sites, but not for all intensification since the systems were built)
• aging infrastructure = more water infiltration through cracks and sewers and manholes settle over decades, using up sewer capacity
• plugging the relief overflows in sanitary systems = sewage back-up into basements instead of into watercourses (see Ontario's Policy F 5-5 that limits overflow volume to watercourses . Google it)
• underpinned / lowered basements = a lot of carpets and hardwood floors in expensive finished basements are now closer to sewer levels and back-up
• inadequately maintained foundation drains = post WWII boom dwellings are old and so are the foundation drains around the perimeter of the homes; many flood calls to municipalities relate to these and not the City's mainline sewer in the street (similar issue for private drain connections, often blocked by roots that enter old clay tile pipe)
• infill in poor groundwater conditions = the best sites have been built on, and the low-lying soggy ones were built on next; many flood calls to municipalities relate to properties with chronic groundwater conditions, aggravated by today's normal rainfall
• expensive basement finishing and content = more expensive repairs (my house basement was used to store coal in the early 1900's, later an oil tank and furnace and tools, now there is an extension and hardwood floor and lots of electronics - nicest rooms in the house actually!)
• construction sewer by-pass failure = more back-ups during extreme events (contractors size by-bass pumps so they can take sewers out of service during construction (looked around Toronto lately? there is a lot) and the pumps have finite capacity and cannot handle large storms.  This is why Union Station flooded recently and I have some other examples in the City where I work. Toronto recently updated the contract specifications for by-pass pumping to require more capacity to reduce back-up risks. Google Engineering and Construction Services Division Standard Specifications for Sewers and Watermains TS 4.01 April 2013 and see for yourself).
• Keating Channel / Don River outlet dredging = frequency of Don Valley Parkway Flooding ( “If you don’t dredge, it floods the Don River and the Don Roadway and the Don Valley Parkway,” explains Angus Armstrong, the harbour master. “That’s the practical reason.”; read about it here:National Post Article
Hint to why more extreme storms are being observed and reported:
• more dense rain gauge network that doesn't miss any storms = more storms reported (rain gauges used to record on analog paper rotating drums which was expensive to process - today's technology is better and gauges are cheap, electronic and data is easy to download over the net to process and report; for example York Region municipalities used to have only a gauge at Buttonville Airport .. now on city has 19 of its own to support municipal investment in urban flood risk reduction; just like fishing with a finer net, you will catch more fish)
• 24 hour talk radio, the weather channel, weather as a business (storm chasers), promotion by the insurance industry who want municipalities to address pre-1980 standards, documentary writers ......

### Identify Flood Risks with the Toronto Flood Map

The Toronto Flood Map can help answer questions for Toronto residents, businesses, tenants, home-buyers, builders, insurance policy holders and insurance brokers / adjusters /actuaries interested in flood risk:
• Is a neighbourhood at risk of flooding during severe rain storms? Along what roads or portions of roads have property owners reported flooding during large rainfall events (July 8, 2013, August 19, 2005, May 12, 2000)?
• Is the area you are considering buying or renting in flood prone (so does the property of interest have flood risk mitigation measures in place, like a backflow valve to control sewer back-ups, or good lot drainage to keep runoff away from windows, doors and the foundation wall)?
Residents and businesses have reported flood incidents to the City of Toronto after extreme storm events. Calls are made during a flood event or afterward, sometimes as part of an insurance claim related to property damage. The City's map is is shown below.

 City's map: Toronto Water Basement Flooding Location Map
As shown on the close up of the City's map below, only major roadways and points of reference are available, making it difficult to pinpoint a local area of interest. Unfortunately the City's map is not at a sufficient scale to show local streets with high historical flooding risk.

 Close up of City's map (no local streets shown)
 May 2000 locations estimated with symbol (!) on City's map
As a further limitation, May 2000 flood report locations are only shown with a symbol (!) as the original City map did not embed Geographic Information System fonts in the exported PDF file. These symbols are off-set from the reported basement flooding location (red dots in the legend).

Our Toronto Flood Map overcomes City map limitations and reveals high risk areas and neighbourhoods that have flooded during recent severe weather events.

Of course there are limitations in interpreting the flood location map:
• The City of Toronto is investing heavily in improvements to the sewer system to limit risks so some areas may no longer have the same risk after construction as they had in the past.
• Affected property owners may have invested in individual risk reduction measures (back-flow valves, sump pumps, improved lot grading, etc.) to protect their property from further damage.
• Not all flooding is reported so virtually any property along any street could experience flooding under certain conditions. For example, any property could experience flooding if the private sewer or drainage systems on the individual properties are not maintained (clogged foundation drains / weeping tiles, lateral service connection blocked with tree roots or other debris), or if there are other risk factors (reverse slope driveway, etc.).  This explains some isolated flood locations.
• Because the high intensity area of a storm may cover only part of the city, the density of flood locations reflects the relatively higher local rain stress on the drainage system, and not necessarily the local drainage system capacity compared to other parts of the city.  Other storms may affect parts of the City in a greater way (e.g., August 19, 2005 affected the north part of the City, while July 8, 2013 affected the west part).
• Individual addresses cannot be identified due to the scale of the map and spatial accuracy.
Nonetheless, Toronto Flood Map can still show the relatively higher flood risk streets within a neighbourhood.  Currently the map shows flood reports for the August 19, 2005 and May 12, 2000 storms. The May 12, 2000 basement flooding locations on the City's map have been corrected / adjusted so that affected streets can be identified.  The July 8, 2013 storm flood report locations, available through other City reports, will be available shortly as will other historical reports reported through local studies.

### Can Twitter Accurately Document Flooding Incidents?

Bright Planet has explored whether social media could "help in a flood", especially in the context of reviewing tweet data for insurance claims.  Here's what they did in relation to the July 8, 2013 Toronto storm and flood:

"To assist with the vetting and triage of insurance claims, BrightPlanet harvested all tweets within the Greater Toronto Area. BrightPlanet then filtered and curated the tweets down to only tweets discussing specific effects of the flooding. Any tweets containing a latitude and longitude were then mapped in a heat map format in Toronto to show where the most chatter was happening about the flooding."

See the full description on their page here: BrightPlanet.  The harvesed image is here:

There are a lot of tweets downtown, mostly south of Bloor Street. Matthew Dance analyzed tweets about Toronto - not about flooding, and produced this cluster map:

Of course the extents are different, but I'd suggest that we see the same pattern in BrightPlanet "flood tweets" as we do for Matthew Dance's "generic Toronto tweets".  Dance explains tweet density in a few ways:

• "The most densely Tweeted area is bounded by Bloor Street to the North and Lake Ontario to the South, connected by Young Street.  There is a greater density along the Lake, away from Young to the West."
• Hot spots include the Eaton Center, Rogers Center and tourist and suburban destinations - the areas around Yonge and Bloor and Front Street, including the sports stadiums - "destinations for those interested in shopping or taking in the sights in Toronto."
• Areas that are strictly neighbourhoods (Hillsdale Avenue running east from Yonge) have low density.
• Poorer neighbourhoods like Regent Park have few tweets (between Dundas and Gerrard west of the Don Valley)

Where does the City of Toronto report flooding for July 8 2013? It does not match the BrightPlanet flood tweet locations very well as shown below:

Tweet-rich downtown had only 50 mm of rain and limited reported flooding, compared to 130 mm of rain in tweet-poor Etobicoke where most of the flooding was reported.  Perhaps the BrightPlanet flood tweets should first be "normalized" by the density of generic non-flood tweets, to give a better representation of relative spatial flood activity?

It would also be worth checking if what people tweet more about downtown is spectacular surface flooding (e.g., a sewer geyser blowing a manhole 10 feet in the air, a Ferrari floating in an underpass lake, an amphibious GO Train) as opposed to basement flooding, which is the focus of City of Toronto flood complaints.  Tourist and passersby have time to tweet about flooding, while a homeowner with sewage gurgling up their basement floor drain may have more pressing thing to do (save the family photos honey!).