Southern Ontario Extreme Rainfall Trends and Engineering Design Considerations for Effective Management

This blog has many posts that illustrate how the media, some researchers, and/or the insurance industry have missed the mark when it comes to defining past changes in extreme rainfall and the effects on communities. What is also apparent is that government agencies and ministries have also missed the mark when it comes to defining these changes and what it means to stormwater management policies and design standards. 

This post provides:

1) Examples of media, researcher and insurance industry data gaps on extreme rainfall compiled over the years;

2) A presentation made to the the Ontario Ministry of the Environment, Conservation and Parks with information for Ontario (MECP) showing inaccurate government statements on past trends, and actual data characterizing southern Ontario trends to help guide development of standards for stormwater systems; and

3) References to updated material that expand on and confirm the above.

***

1) Examples of the media, researchers, and the insurance industry misstating extreme weather and rainfall trends

CBC and Radio-Canada - this blog post compiles corrections by CBC and Radio-Canada related to extreme rainfall frequency since 2015, including corrections to or the deletion of articles that violated their journalistic standards and practices, including the commitment for accuracy in reporting: https://www.cityfloodmap.com/2019/06/cbc-correcting-claims-on-extreme.html

Key take-away: even large media organizations are not well-equipped to critically report on the basics of extreme rainfall. 

TVO - this blog post includes a critique of storm intensity and extreme temperature reporting in 2017 and 2019: https://www.cityfloodmap.com/2017/11/tvo-articles-on-climate-change-extreme.html

Key take-away: smaller media organizations are not well-equipped to critically report on the basics of extreme rainfall or temperature trends either. 

Insurance Industry / Researchers (plus more media) - this presentation shows that many cannot tell the difference between a theoretical, arbitrary 'bell-curve' shift in extreme weather frequency and actual observations, and that there is no fact-checking in most media. This relates to the widely-reported incorrect claim in the Institute for Catastrophic Loss Reduction (ICLR) report "Telling the Weather Story" for the Insurance Bureau of Canada (IBC) that "weather events that used to happen once every 40 years are now happening once every size years": https://www.slideshare.net/RobertMuir3/storm-intensity-not-increasing-factual-review-of-engineering-datasets

Key take-away: media organizations, and even the chief economist at a large Canadian bank, do not verify suggested extreme rainfall or weather trends in Canada, and will widely echo unsupported/incorrect/false statements. In this case the insurance industry had not checked the researcher statements in a report they commissioned.

2) MECP presentation on extreme rainfall trends to guide stormwater design standards

The MECP embanked on the development of design standards for wastewater and stormwater systems in Ontario to support Consolidated Linear Infrastructure (CLI) Environmental Compliance Approvals (ECAs). CLI ECA's for stormwater systems (like wastewater systems) are intended to consolidate individual ECA's for stormwater infrastructure in a municipality.

Adapting to a changing climate is one consideration in developing stormwater management standards and criteria. New, expanded criteria have been proposed for stormwater CLI ECAs, largely echoing the MECP's draft Low Impact Development Stormwater Management Guidance Manual criteria such as managing 90th percentile storm events (see January 2022 ERO posting: https://ero.ontario.ca/notice/019-4971). The ERO posting notes several goals including:

"The draft Low Impact Development Stormwater Management Guidance Manual aims to ... Increase resiliency of communities and associated stormwater infrastructure to climate change and help mitigate climate change"

A presentation was made to the MECP Stormwater Design and Permissions Working Group on May 1, 2019 to support the stormwater CLI ECA development, providing actual data on past trends. Some highlights are below illustrating that:

i) the former Environmental Commissioner of Ontario also echoed the above incorrect/unsupported insurance industry statements on increasing extreme weather frequency:


ii) Ontario government documents claims more frequent extreme weather is 'already underway':


iii) The LID Manual, including the November 2017 draft cited below (link: https://drive.google.com/open?id=1NHtrjCglDgox4tYISLU5LLZYS32Ea4MN)
and the January 2022 noted above, cites only the increases in maximum rainfall intensity in Ontario, omitting significant negative trends in southern Ontario that demonstrate decreasing risks: 


The 2022 draft LID Manual includes the same highlighted 'red box' text above in Table 6.1 – Examples of Observed Changes in Ontario Climate, namely:

"The maximum intensity for 1-day, 60-minute and 30-minute duration rainfall events increased on average by 3%-5% per decade from 1970 to 1998 (Adamowski et al., 2003)."

The citation is as follows: Adamowski, K., J. Bougadis, and G. Pessy., Influence of trend on short duration design storms, pp. 15. Department of Civil Engineering, University of Ottawa, 2003.

To provide a more complete perspective on rainfall intensity trends, Kaz Adamowski and John Bougadis also analyzed trends in Ontario rainfall intensities in 2003 in HYDROLOGICAL PROCESSES (Hydrol. Process. 17, 3547–3560 (2003)), published online in Wiley InterScience (www.interscience.wiley.com), DOI: 10.1002/hyp.1353. As shown below, that analysis found both increasing and decreasing trends - not just the increases cited by MECP.

Adamowski and Bougadis found significant increasing trends only in northern Ontario. In southern Ontario 'significant negative trends' were observed for all durations, except the 2 hour. Central Ontario had positive trends, but none were statistically significant:

The full slide presentation to MECP including the selected slides above is as follows:



Clearly the Ontario government and MECP has not adequately characterized extreme rainfall trends in the province, and presents only partial, dated analysis that is inadequate to define and support new stormwater management standards or policies considering actual risks.

3) References to updated material that expands on and confirm the above

Historical extreme rainfall trends in Canada and across Ontario have been summarized in the National Research Council of Canada (NRC) National Guidelines on Undertaking a Comprehensive Analysis of Benefits, Costs and Uncertainties of Storm Drainage and Flood Control Infrastructure in a Changing Climate (see post: https://www.cityfloodmap.com/2022/02/nrc-national-guidelines-on-flood.html).

The Ontario trends noted in the 2019 slides above, that included IDF data Environment and Climate Change Canada's Version 3.00 Engineering Climate Datasets, have been updated with Version 3.10 data. For example, southern Ontario annual maximum trends and IDF trends based on recent updates are shown below:


In this heavily populated region, where there is a vast amount of infrastructure, clearly there are many decreasing trends in observed annual maximum rainfall at the long term climate stations. This contradicts the statements by the media, some researchers, the insurance industry and the Ontario government.

The following figure shows IDF trends since 1990 for long term climate stations. On average small 2 year rain intensities are lower and extreme intensities are unchanged.


Therefore data continue to not show any 40 to 6 year frequency shift in extreme rainfall across Canada, as claimed by some researchers. Overall, extreme rainfall intensities have decreased at over 200 climate stations with available data in Canada as illustrated in the NRC guideline (see the blog post noted above for details). In southern Ontario, some intensities have decreased on average, like the 2-year rainfall intensities, consistent with Adamowski and Bourgadis' 2003 analysis noted earlier.

Stormwater management standards in Ontario, whether in a LID Manual or CLI ECA criteria, should consider the actual rainfall data trends and the resulting risks to systems. This is currently not occurring. 

Consistent with past observations of extreme rainfall trends are the future projections for southern Ontario (see post: https://www.cityfloodmap.com/2022/01/future-extreme-rainfall-idf-values-in.html). In Assessment of non-stationary IDF curves under a changing climate: Case study of different climatic zones in Canada in the August 2021 Journal of Hydrology: Regional Studies, Silva et al. projected future IDF curves in several regions of Canada under various emissions. When more plausible emissions scenarios are considered (see RCP2.6 projections below), the 100-year intensities decrease in London and Hamilton:



For RCP4.5, Hamilton has projected increases and decreases in 100-year intensities depending on the duration, and London has minor increases (no more than 5%) - increases of such small magnitudes in one input parameter are not relevant in engineering design given the dominance of other factors related to the transformation and application of the data in hydrologic analyses (e.g., hyetograph selection).

Other research by the University of McMaster entitled Assessment of Future Changes in Intensity-Duration-Frequency Curves for Southern Ontario using North American (NA)-CORDEX Models with Nonstationary Methods predicts that extreme 50-year rainfall amounts will decrease in Southern Ontario by 2050, that moderate 25-year rainfall amounts will remain flat, and frequent 10-year rainfall amounts will increase overall for long durations, but are mixed for short durations. See previous post: https://www.cityfloodmap.com/2018/04/climate-models-predict-decreasing.html. The following tables illustrate projected 50 year return period trends from that study, with decreasing trends shown in blue:


The projected change in extreme rainfall for the majority of durations and locations in southern Ontario is negative, i.e., decreasing rainfall design intensities.

Stormwater design policies and standards in Ontario should consider future climate effects and future, long term risks. In southern Ontario, where most stormwater infrastructure exists, it is not clear that projected rainfall intensities used in design will be increasing significantly, if at all, based on several sources above. Accordingly, a robust risk-based approach to setting standards or policies for adaptation is needed that weighs the cost of any new standards against their benefits, especially in regions with past decreasing trends and limited projected increases - those regions include southern Ontario. Where benefits are low due to limited expected rainfall changes, and adaptation costs are extremely high (see this presentation to MECP on costs of LID measures for example: https://www.slideshare.net/RobertMuir3/cost-effectiveness-of-lid-implementation) such a robust risk-based approach is even more essential to ensure that investments in any higher infrastructure standards are worthwhile. Such an approach has not yet occurred in the development of Ontario's draft LID Manual or CLI ECA criteria, when it comes to managing stormwater and extreme rainfall.

Based on the above, Ontario's draft LID Manual should be critically reviewed, expanded and updated with respect to actual extreme rainfall trends and projections in Ontario. The need for increased resiliency (and the associated cost of generic one-size-fits-all measures), when actual observed and projected risks are low, should be reconsidered

Posted by