National Guidelines on Undertaking a Comprehensive Analysis of Benefits, Costs and Uncertainties of Storm Drainage and Flood Control Infrastructure in a Changing Climate

The following paper was presented at the 2021 WEAO conference. The presentation is included at the bottom of the post. A pdf copy of the paper is available here:  paper

The full guidelines are available here to download: https://nrc-publications.canada.ca/eng/view/object/?id=27058e87-e928-4151-8946-b9e08b44d8f7

***

National Guidelines on Undertaking a Comprehensive Analysis of Benefits, Costs and Uncertainties of Storm Drainage and Flood Control Infrastructure in a Changing Climate

Fabian Papa*, M.A.Sc., M.B.A., P.Eng., FP&P HydraTek Inc., Robert J. Muir, M.A.Sc., P.Eng., City of Markham, Yehuda Kleiner, Ph.D., P.Eng., National Research Council of Canada

*FP&P HydraTek Inc., 216 Chrislea Road, Suite 204, Vaughan, Ontario L4L 8S5

INTRODUCTION

      Following more than two years of extensive research and stakeholder engagement, the National Research Council of Canada (NRC) has recently completed the development of a comprehensive resource for practitioners to assist in the development of economic assessments of initiatives aimed at reducing flooding damage to core public infrastructure assets.  The overall guidelines which bear the same title as this paper consists of a “main” guideline document supported by several appendices of foundational research, comprising over 700 pages in total.  

       The guidelines (NRC, 2021) were developed in recognition of the need to harmonize and standardize the assessment of the value of storm drainage and flood control infrastructure initiatives. The benefits provided by these infrastructure initiatives include the value of future avoided damages as well as direct and indirect co-benefits, such as enhanced health, recreational and environmental value.  These benefits are compared to the costs associated with the infrastructure initiatives whilst considering the various uncertainties associated therewith, explicitly including those uncertainties associated with a changing climate.  While such analyses are not uncommon in Canada, they have generally been limited to large-scale projects or other special situations requiring significant levels of investment and sophistication. In this light, these guidelines are intended to help promote such practices for a broader range of projects and a broader range of jurisdictions with varying degrees of sophistication as well as the quantity and quality of available information relating to their assets.

      These guidelines are intended to promote the rational assessment of projects or initiatives, using rigorous analysis through an economic lens such that competing projects or alternatives can be objectively assessed and compared.  Concepts such as the time value of money, benefit estimation, life cycle costs, net present value, net benefits, benefit-cost ratios, cost-effectiveness, sensitivity analyses and probabilistic risk assessments are explicitly considered in the guidelines.  The application of these concepts is demonstrated through five case studies that span a broad spectrum of project or initiative types as well as scales.

      This paper is intended to provide a brief synopsis of the guidelines, and the reader is encouraged to obtain a copy of the complete publicly-available document for additional details.

 Organization of the Guidelines

     The guidelines are organized into two components: (i) a main body referred to as the Guidelines Document; and (ii) nine appendices consisting of a comprehensive bibliography with over 300 entries followed by a glossary and list of acronyms, the foundational research (discussed further below) and case studies.  The Guidelines Document itself provides the reader with brief summaries of the salient findings of the foundational research work, followed by a generalized approach to conducting the economic assessments and relevant information for performing time value of money calculations, estimating benefits and costs as well as assessing uncertainties.  As with any such practice, it is evolving in nature and, while the fundamental concepts generally do not change materially over time, the ability to apply these concepts can change as the degree of available information expands and improves.  To this end, a brief section dealing with considerations for future work concludes the Guidelines Document.

      The extensive foundational research work is summarized as follows (with the relevant appendix titles in bold typeface):

 ·       A Benefit-Cost Analysis Industry Scan was undertaken to assess practices across Canada and internationally to identify the state-of-the-art as well as limitations.  Included with this work was the review and assessment of several relatively recent applications made to Infrastructure Canada’s Disaster Mitigation Adaptation Fund (DMAF) which, in turn, revealed several issues related to the estimation of benefits and application of benefit-cost analyses.

·       A thorough review of Direct & Indirect Long Time Horizon Flood Damages reports and research studies was undertaken in conjunction with a review of available insurance industry data sets for purposes of estimating benefits (i.e., avoided damages) associated with project or initiatives aimed at reducing flooding.  While bottom-up estimation of avoided damages is fairly common, damage estimation considering actual reported losses has not been readily available – insurance industry claim data has been analyzed to support such analysis.

It is not always sensible or appropriate to apply bottom-up techniques when assessing certain project types or scales.  The ability to use insurance industry data allows for top-down assessments and guidance on how this is done is provided in this appendix.  Amongst the concepts covered in this portion of the research is that related to the use of damage-probability relationships to derive the Expected Annual Damages (EAD), also referred to as Annual Average Damages (AAD), being an important value in the estimation of the overall present value of damages. 

Table 1 provides the results of the assessment of available insurance industry data in relation to the value of sewer back-up. Losses per property may be applied to bottom-up analyses., Aggregate EAD values, to be used in top-down analysis are provided for flood-related losses as well as sewer back-up losses.  Additional details are available in the guidelines and its appendices, and it is expected that these values may continue to be refined and updated over time as additional information becomes available. 

TABLE 1 – EVENT AVERAGE SEWER BACK-UP LOSSES (PER PROPERTY) AND INSURED LOSS EAD VALUES (CAD 2018)

Jurisdiction	Event Average Sewer Back-up Loss (for bottom-up analysis)	Flood Loss EAD (for top-down analysis)	Sewer Back-up/Water EAD (for top-down analysis
		(Millions)	(Millions)
Canada	$22,300	$819	$376
Alberta	$19,700	$414	$88.8
British Columbia	$8,440	$16.1	$0.752
Manitoba	$8,870	$14.2	$1.83
New Brunswick	$13,300	$7.32	$2.33
Newfoundland and Labrador	$17,400	$10.2	$1.83
Nova Scotia	$13,900	$18.9	$14.2
Ontario	$18,500	$289	$244
Prince Edward Island	$8,500	$0.222	$0.0085
Québec	$9,890	$90.4	$35.9
Saskatchewan	$18,200	$41.8	$15.0

The aggregate EAD values may be scaled down to a municipal- or project-level to estimate existing damages that could potentially be avoided in the future with the new infrastructure under consideration. Details are provided in the case studies.

·      The appendix Climate Change & Flood Damage Considerations addresses issues related to meteorological uncertainties. Potential changes in climate and meteorology have the potential to significantly affect the shape of future damage-probability relationship. This component of the research thoroughly reviews the available literature and data to give the reader a complete reference to consult.  One of the important findings of this work was the need to differentiate between short duration meteorological events (typically less than 1 day and often on the order of hours) and long duration climate that ranges from several days to more typically on the order of months, seasons or years.  Urban (pluvial) and sewer surcharge-related flooding, where the majority of damages occur, is typically driven by short-duration rainfall events and therefore, these deserve the appropriate level of focus when considering projects at the more common scales of sewersheds and municipalities.  Environment and Climate Change Canada (ECCC) data relating to short-duration rainfall trends was assessed in detail and found to (i) generally not show strong overall signs of change in any direction (i.e., up or down), and (ii) have a significant variation that is location-dependent.  These observations regarding short duration rainstorms are also corroborated in the literature (Shephard et al., 2014; ECCC, 2020).  Methods for considering future climate conditions are reviewed, albeit with varying degrees of uncertainty, and methods for dealing with such uncertainty are also identified.

·       It is important to recognize that intangible benefits also accrue with any interventions that reduce the likelihood and consequence of flooding.  Accounting for these additional benefits will have the effect of improving the economic assessment of any intervention.  The appendix titled Post-Flood Event Economic, Legal, Social and Indirect Costs addresses matters related to reductions in property values, human health impacts, population displacement, disruption of (and stress on) municipal infrastructure services, and legal costs. The appendix titled Post-Flood Event Environmental Impacts addresses matters related to flood impacts such as erosion, quality of water in the environment, impacts to flora and fauna and greenhouse gas emissions, among others.  It is worth noting that it is often difficult, and sometimes impossible to monetize many of these co-benefits.  Nevertheless, the process of identifying and (non-monetarily) quantifying such benefits is helpful in striving for a comprehensive economic assessment and may be incorporated into multi-criteria and/or triple bottom line analyses that can accompany the monetarily-based economic assessments.

·       The appendix titled Life Cycle Costs of Storm Drainage and Flood Control Infrastructure addresses an area where the industry benefits from a considerable amount of data and experience as such assessments, or at least components thereof, are routinely conducted in relation to project selection, construction cost estimation, construction contracting as well as asset management and financial forecasting.  In addition to basic guidance on developing cost streams for time value of money calculations, an abundance of cost estimating data and relationships is also provided to facilitate this component of any economic assessment.

·       The overall document concludes with an appendix consisting of five Case Studies which demonstrate the application of the various concepts and methods promoted in the guidelines.  The case studies are largely developed from actual projects of various types, scales and locations in Canada in order to give the practitioner a sense of how such economic assessments may be approached and conducted. 

      Although the guidelines are the result of specific, deliberate and thorough research, they are not to be construed as a prescriptive approach.  Each opportunity to apply them needs to be assessed based on its own particular circumstances, including the degree of importance (i.e., value of initiative considered and/or risk associated with infrastructure in question, including its criticality), the quantity and quality of available information as well as the level of effort to acquire any additional information and conduct supporting technical analyses (e.g., hydrologic and hydraulic modelling, depth-damage curve development, etc.), amongst other matters.  To assist the practitioner in identifying how to approach an analysis, a conceptual model is provided in Figure 1.

 

FIGURE 1:  CONCEPTUAL MODEL TO IDENTIFY ANALYSIS METHODS TO CONSIDER WITH EXAMPLE APPLICATIONS

EXPECTED ANNUAL DAMAGES (EAD)

      As noted earlier, the main benefit of any flood-control measure is the value of the damages this measure is expected to avoid over the duration of its service life.  The high uncertainty that is inherent in estimating future damages necessitates a probabilistic approach, where damage from a flooding event is associated with the probability of occurrence of this event.  This is illustrated graphically in Figure 2, and involves determining the area beneath each of the damage-probability curves (i.e., both the existing curve and the future, after the improvement project, curve) to arrive at the area between these curves.

 

FIGURE 2:  CONCEPTUAL CALCULATION OF EXPECTED ANNUAL DAMAGES (EAD) AVOIDED AS A RESULT OF IMPROVEMENT PROJECT

      While this calculation seems straightforward enough, it could be complicated by the fact that many relevant factors are not necessarily stationary over time.  Matters such as economic growth, representing the real (rather than nominal) value of the properties and assets that are impacted by flooding that can be expected to grow over time as the overall wealth of the society increases.  Further, climate change as well as increased urbanization may potentially change the character of the relationship between damage and probability.  Changes in the value of assets at risk of flooding (i.e., economic growth) will have the effect of shifting the damage-probability curve vertically (i.e., positive growth will result in an upward shift), while changes in climate, assumed to be represented by changes in rainfall intensities for different return periods (or probabilities of occurrence) will result in a horizontal shift (i.e., increasing rainfall intensities will result in a shift to the right).  The combined impact of these changes, as well as how EAD may change over time as a result thereof, is illustrated in Figure 3.

 

FIGURE 3:  CONCEPTUALIZATION OF IMPACTS OF
ECONOMIC GROWTH AND POTENTIAL CLIMATE CHANGE
ON FUTURE DAMAGE ASSESSMENT

ASSESSING UNCERTAINTY

 A good forecaster is not smarter than everyone else,

he merely has his ignorance better organized.

- Anonymous

      It is extremely important to acknowledge, understand and explicitly consider the uncertainties inherent in any analysis, particularly given the scale of investment and potential damage implications associated with storm drainage and flood control infrastructure.  Ignoring these uncertainties may result in significant under- or over-investment, yielding poor returns on investment and depriving society of value by wasting resources that could otherwise be deployed for other, more valuable, infrastructure improvements.  The impact of neglecting to account for uncertainty is magnified by the extremely long service life of these infrastructure assets – typically in the order of 75-100 years.

      Uncertainties exist in relation to the estimation of baseline benefits and costs, as well as in relation to potential changes in the benefits of avoided damages associated with economic growth and potential climate change impacts.  Additionally, the assumptions applied for the time value of money calculations, including the discount rate and time horizon, may play meaningful roles in changing the economic outlook of a project. 

      The guidelines identify various methods for assessing uncertainty, including the relatively simple application of sensitivity analysis and, as a subset thereof, stress tests.  Both of these types of analyses have been adopted by various jurisdictions to help identify vulnerabilities in existing or proposed infrastructure systems and allow for the allocation of additional costs and efforts only where they are most needed.  A more sophisticated method of dealing with uncertainty covered in the guidelines include the application of the probabilistically-based Monte Carlo analysis, where probability distributions for each of the uncertainty-bearing parameters can be developed and applied to produce a probability distribution for the output (e.g., benefit-cost ratio or other outcome sought through the analysis). 

      The guidelines also contain a discussion of Real Options Analysis, which is expected to be particularly relevant to the intended audience given the potential scale of investment required for certain projects aimed at reducing flooding.  This approach focuses on adaptability and incorporates the ability to incorporate information as it becomes available to help decide on the appropriate next step.  It limits possible over-investment that may result from high uncertainty, which in turn leads to over-estimation of the severity of future conditions, whilst not constraining the ability to implement the investment when supported by the then available evidence.  This approach bears a strong resemblance to the Observational Method of the American Society of Civil Engineers (ASCE) in relation to the adaptive design and risk management of infrastructure for climate resiliency (ASCE, 2018).  It is a practical approach and is typically practiced in a less formal manner, although there is merit in formalizing (and documenting) such assessment processes to promote a comprehensive consideration of influencing matters as well as for purposes of communications, both when the analyses are occurring periodically over time, as well as across time periods.  The Canada in a Changing Climate: National Issues report (Government of Canada, 2021) identifies Real Options Analysis as one main approach for accommodating uncertainty in the economic appraisal of adaptation actions.

 CONCLUSIONS

    The development of a national set of guidelines to undertake comprehensive economic assessments of storm drainage and flood control infrastructure is intended to promote, advance and to some extent harmonize and standardize such practices across Canada and at various levels of implementation.  The guidelines, supported by extensive foundational research, represent a comprehensive reference for practitioners in assessing the value of initiatives that may be considered, including the consideration of uncertainties (including those related to potential climate change impacts), such that competing projects and project alternatives can be objectively and rationally assessed so as to inform decisions and promote the judicious allocation of investment capital.  In many ways, these guidelines represent somewhat of a renaissance of the application of classical engineering economics in relation to matters that are of current relevance and in light of the vast (and growing) amount of information and tools available to inform these assessments.

BILBIOGRAPHY

American Society of Civil Engineers (ASCE) (2018) Climate-Resilient Infrastructure: Adaptive Design and Risk Management. Committee on Adaptation to a Changing Climate.

Environment and Climate Change Canada (ECCC) (2020) Climate-Resilient Buildings and Core Public Infrastructure.

Government of Canada (2021) Canada in a Changing Climate: National Issues.

National Research Council of Canada (NRC) (2021) Guidelines on Undertaking a Comprehensive Analysis of Benefits, Costs and Uncertainties of Storm Drainage Infrastructure and Flood Control Infrastructure in a Changing Climate

Shephard, M.W., Mekis, E., Morris, R.J., Feng, Y., Zhang, X. (2014) Trends in Canadian Short-Duration Extreme Rainfall: Including an Intensity-Duration-Frequency Perspective, Atmosphere-Ocean, 52:5, pp. 398-417

***

The follow presentation was made at the conference in London, Ontario.

National Research Council - Guidelines on Undertaking a Comprehensive Analysis of Benefits, Costs and Uncertainties of Storm Drainage and Flood Control Infrastructure in a Changing Climate from Robert Muir

What COVID-19 Taught Us About Observed Data vs. Model Projections: They Are Different - Let's Remember That When Interpreting Climate Models

COVID-19 - observed data on ICU cases and projected capacity

"All models are wrong, some are useful".  Predicting COVID-19 conditions has taught us that models come with a great deal of uncertainty, and are based on a lot of assumptions.  Furthermore, models have to be constantly updated over time with real observed baseline data to represent the starting point for future predictions. At least we recognize the difference between theoretical model projections and the past observations on COVID-19 conditions.  More attention should be given to the difference between theoretical models of climate effects and observed changes in extreme weather.

In early April, COVID-19 ICU cases were projected to increase to 1200 in a best case to about 1500 in a worst case in Ontario, increasing considerably from actual data counts in late March.  The chart at right shows that ICU beds peaked at under 300 cases by mid April, a fraction of the best case model prediction, and has declined since.  So model projections should be viewed with some caution, and the reliability of the projections should be questioned and validated where possible with real data.

Predicting future weather extremes due to climate change effects has a great deal of uncertainty as well.  The recurrence time of extreme rainfall is predicted to decrease due to climate change effects, meaning that the "return period" of storms would become smaller.  For example, a rainfall event that had a return period of 35 years today (meaning a probability of occurring in any year of 1/35, or 1 in 35) has been predicted to occur every 12 years in the future (i.e., a higher probability of happening each year of 1/12 or 1 in 12 ... that a greater chance than today's 1/35).  That is what is projected to occur in Canada from now to 2100.

The above example on decreasing recurrence times is from a simulation presented in Canada's Changing Climate Report by Environment Canada (link: https://changingclimate.ca/CCCR2019/).  It is for a future scenario with several assumptions about growth and emissions called the RCP8.5 scenario, representing a Representative Concentration Pathway of just one of several future scenarios.  The shift in 24-hour precipitation recurrence times are presented on Figure 4.20 b shown below:

Canada's Changing Climate Report Figures 4.20 b), Projected Extreme Precipitation Recurrence Time / Return Periods for Past, Present and Future Time Periods, RCP8.5 Model Simulation Scenario

As annotated above, today's recurrence time is noted as 35 years, the future recurrence time is 12 years and the past time was 50 years. So the model predicts these shifts in recurrence time (return period) and annual probability:

   Period         Recurrence Time       Probability Each Year
1986-2005             50 years                       2.0 %   (1/50)
2016-2035             35 years                       2.9 %   (1/35)
2081-2100             12 years                       8.3 %   (1/12)

Some have misinterpreted the theoretical, simulation model changes from past to present as 'actual' observed changes in extreme precipitation when in fact the Environment Canada report clearly notes these are 'projected changes' and are 'simulated by Earth system models' for the scenario RCP8.5.  A different scenario's simulated results, with different assumed emissions and growth, and different recurrence time shifts are presented in Figure 4.20 a) as well.

CBC News In Our Backyard - Flooding

CBC's In Our Backyard interactive notes "Climate change is no longer theoretical. It’s in our backyard" - unfortunately it presents theoretical past model trends as real changes that are "In Our Backyard" now.  Here is the online report link: https://www.cbc.ca/news2/interactives/inourbackyard/

CBC News report: "Climate change is making extreme rainfall a more frequent occurrence. Storms that historically happened only once every 50 years are now coming every 35 years or less. By the end of the century, they could happen once every 12 years on average, according to a recent climate report from Environment Canada. All this increases the potential for urban flooding."

CBC News In Out Backyard Extreme Rainfall Frequency - Past, Present, and Future Recurrence Times Confuses Simulation Model Projections With Observed, Historical Trends

So while predicted changes are only theoretical, CBC News mistakenly reports that changes have already occurred and are 'now coming' at smaller recurrence intervals (i.e., higher frequency and higher probability each year).

The CBC Ombudsman has indicated that the CBC should be careful to distinguish between past, present and future extreme rainfall trends, as noted in a recent post: https://www.cityfloodmap.com/2020/05/past-present-or-future-cbc-ombudsman.html

We agree.

A review of historical extreme rainfall trends in one region of Canada affected by may flooding events has shown no decrease in the recurrence time, or return period, of extreme precipitation.  A previous post showed that today's 35 year storms are actually occurring less frequently than in the past. In southern Ontario, long term climate station observations show that the average 25 to 50 year rainfall intensities today are actually slightly smaller than they were considering observations up to 1990. See previous post: https://www.cityfloodmap.com/2020/05/southern-ontario-extreme-rainfall.html

Analysis of the Version 3.10 Engineering Climate Datasets IDF Files updated in March 2020 show that southern Ontario long term rainfall intensities have decreased slightly since 1990, on average by 0.1%.  The 50 year return period rainfall intensities are on average unchanged.

If 50 year rainfall intensities actually occurred more frequently and now occur at a 35 year return period, as CBC mistakenly reported, then the magnitude of the 50 year intensities would have had to increase by about 6%.  This considers the example long term climate station at Toronto's Pearson International Airport - the 35 year 24-hour rainfall intensity of 99.7 mm at the airport would have to increase to the 50 year intensity of 105.7 mm.  Back in 1990, the 50 year 24-hour rainfall intensity at the airport was 109.3 mm, meaning the 50 year rainfall has decreased by several percentage points.  Here are the 1990 data (copied from my top desk drawer):

Toronto Pearson International Airport IDF Table With Data Up to 1990 - 50 year design rainfall intensity of 109.3 mm  (shown here) was higher than today's version 3.10 Engineering Climate Datasets intensity of 105.7 mm (see table below to 2017).  

Here are the recently updated IDF values from Environment Canada considering data up to 2017:

Toronto Pearson International Airport IDF Table With Data Up to 2017 - 50 year design rainfall intensity of 109.3 mm (see previous table to 1990) shown is lower than today's version 3.10 Engineering Climate Datasets intensity of 105.7 mm (shown here).

Climate models that predict more frequent future rainfall intensities, characterized by shorter recurrence times (i.e., lower return periods = higher probabilities of occurrence) are not necessarily in step with observations (see Toronto airport example above and previous post on southern Ontario long term stations).  Here is a comparison of past trends in 100-year rainfall intensity based on observed data and projections from various studies - the actual data curve is already 'flat', so the need to flatten the curve can only be made based on projections and not past data.

Extreme Rainfall IDF Trends - Toronto 24-Hour 100-Year Rainfall Volumes per Environment Canada Engineering Climate Datasets - Past Data and Linearly Projected Trends Shown in Black.  Various Studies and Models Project Significant Increases That Have Not Shown Up In The Data Observed Data Statistics

Just like COVID-19 models have considerable uncertainty and must rely on observational data to calibrate and validate them - so they they are more reliable and useful in making projections of the future - climate models require checks on accuracy and usefulness.  Media like CBC News may not discern between model predictions and actual trend data which can mischaracterize trends in extreme weather.  Since models predicting extreme rainfall do not appear to match past observations over the recent past few decades, the accuracy and reliability to project conditions over the next 80 years should be closely scutinized.

While in the case of COVID-19, the need for "flattening the curve" is clear given the close scrutiny of observed data that has shown rising counts of infections, hospitalizations or deaths - that gives clear direction on actions to be taken to mitigate observed phenomena.  In the case of COVID-19, these values may even increase at an exponential rate.  In contrast, the IDF curve trends are largely flat if not already declining based on observed data in some regions.  Any change in extreme rainfall trends has been explained by natural variations (i.e., trends can go up).

***

There is a long-standing gap in the media mixing up predictions of extreme weather and actual Environment Canada observed data trends - sometimes a single report can start a narrative that can go unchecked for some time.  The "Telling the Weather Story" report is one such example where a theoretical shift in extreme weather has been reported, and repeated endlessly in the media as actual data when it is clearly not:

Storm intensity not increasing - factual review of engineering data - Canada and Ontario from Robert Muir

Climate Change and Infrastructure Resiliency Assessment - What Representative Concentration Pathways Should be Used to Estimate Future IDF Curves? Caution Using RCP8.5.

There is considerable uncertainty in modern infrastructure design methods when rainfall design intensities are well established using past observations and derived return period values. The uncertainties include:

1) runoff coefficients or other hydrology parameters, especially for pervious surfaces
2) catchment response time (time of concentration), typically estimated using empirical methods
3) rainfall temporal pattern, aka design storm hyetograph, derived from input IDF data (for hydrologic and hydraulic simulations) ... and the spatial pattern which is always ignored because it is chaos
4) hydraulic performance of inlets and grates
5) hydraulic roughness and energy losses in junctions, etc.

Considering future climate scenarios, the input IDF is also uncertain. Estimated future values often depend on the assumed Representative Concentration Pathway (RCP) which are called 2.6, 4.5, 6.0 and 8.5 and which represent progressively more extreme emissions, CO2 concentrations, and energy entering the troposphere (the number 2.6 represents the energy per area warming the planet you could say).

Future design IDF estimation tools like the University of Western IDF_CC Tool give a choice of 3 pathways - 2.6, 4.5 and 8.5. The description of RCP 8.5 indicates that this scenario gives the most severe climate change impacts as noted below. But the extreme intensities do not always support that statement.

The following future IDF intensity tables for RCP2.6, RCP4.5 and RCP8.5 show that the RCP4.5 scenario can give the highest short duration intensities that affect infrastructure capacity and resilience. The 5-year 5-minute intensity is highest for RCP8.5 but only 3.5% above RCP4.5, which is very small in the context of infrastructure design. But the 100-year 5-minute intensity for RCP8.5 is below the RCP4.5 values by over 8% - and the 24-hour intensity is also about 8% lower. So RCP8.5 is not the most conservative, 'most severe' scenario for extreme 100-year events. Note these tables are based on a period of 2050-2100.

The reasonableness of the RCP8.5 has been questioned. In Energy, University of British Columbia note “RCP8.5 no longer offers a trajectory of 21st-century climate change with physically relevant information for continued emphasis in scientific studies or policy assessments.” Researchers add:

"This paper finds climate change scenarios anticipate a transition toward coal because of systematic errors in fossil production outlooks based on total geologic assessments like the LBE model. Such blind spots have distorted uncertainty ranges for long-run primary energy since the 1970s and continue to influence the levels of future climate change selected for the SSP-RCP scenario framework. Accounting for this bias indicates RCP8.5 and other ‘business-as-usual scenarios’ consistent with high CO2 forcing from vast future coal combustion are exceptionally unlikely. Therefore, SSP5-RCP8.5 should not be a priority for future scientific research or a benchmark for policy studies."

Because there is such as wide range of future IDF possibilities already, it is good to know that RCP8.5 could be discounted as implausible in sensitivity analysis. The chart below shows various projections for 5-minute 100-year rainfall intensity for a range of RCP scenarios. Dropping RCP8.5 from further consideration will help focus assessments of infrastructure resiliency. But considering RCP4.5 may yield even higher IDF values than the assumed 'most severe' RCP8.5 scenario.

Future IDF Uncertainty - Moving Target Under Various Representative Concentration Pathways

RCP2.6 scenarios may not result in future IDF intensities that are above current design standard values as shown in the chart above.

Climate Adaptation in the Age of Weather Zoltar (A Short Story on Water Infrastructure Design Limitations and Uncertainty)

What if a "Weather Zoltar" machine answered
all our questions on future climate change
conditions - would we know what to do with
that precise detailed data?
 Are our current decision making processes even
sophisticated enough to make use of it?

The pursuit of extreme weather resiliency and climate change adaptation technical tools and planning / management approaches for core public infrastructure requires a broad review of industry practice and commonly held assumptions about adaptation needs that have been held as tenets. Many of the assumptions do not hold up to scrutiny upon review of fundamental economic and technical data. Some assumptions are overly simplistic and do not recognize the diversity of systems and system components in real-world planning and design environment - the diversity is wide, ranging from newer systems with negligible existing or future climate risks due to inherent design safety factors, to older systems with significant existing risks and fundamental physical and economic constraints to adaptation. The review of newer and older systems and their components and their existing and future risks will help focus the identification of knowledge, technical tool and planning / management framework gaps on areas with the highest risk and highest benefits associated with adaptation efforts.

“Do I really look like a guy with a plan? You know what I am? I'm a dog chasing cars. I wouldn't know what to do with one if I caught it! I just do things.” The Joker in The Dark Knight

It is important that engineers don’t ‘just do things’. If they plan to adapt infrastructure systems to any changes in societal requirements, whether environmental or economic, or to modify approaches to planning and design, there should be a sound basis. Unlike Jokers, they don't 'just do things'.

Imagine the car is future climate and weather details, and the engineering and scientific community is the dog. Imagine if future 5 minutes rainfall intensities and temperatures across Canada were now predicted with exact certainty on a fine 10 m grid across Canada out to the year 2100 and all hydrologists, musicians, municipal engineers, and hydrogeologists were all given the complete future data.  What would they do with it? What would be the plan with this new data? Would they have a sound basis to adapt their practice to this data? Evaluating what they would do with this data can help us review assumptions about systems and system components and explore other fundamental uncertainties in planning, analysis and design.

The Hydrologist

The hydrologist could do things with the minute-by-minute future weather data. Like quickly screen for major events and use some coarse existing hydrologic and hydraulic tools to prevent damages with exact military precision. He would aggregate future rainfall data on a broad catchment basis for his existing hydrology models, and silently curse the fact that he has to now load different storms for each catchment (because he feels that it is less conservative that hitting all catchments with the same intensity peak all at once).

The hydrologist would do his first modelling for areas upstream of known river flood hazards, e.g., Special Policy Areas in Ontario that have big historical and exiting weather flood risks. He would use his steady state river hydraulic model to predict floods levels to the centimetre and get the GIS department to plot the flood limits (but not until after they debate the difference between the hydraulic model elevation benchmarks and the GIS digital elevation model benchmarks) and he would then coordinate with emergency services, engineers and utilities to reinforce existing infrastructure components in the predicted floodplains, and safely evacuate residents and businesses well in advance of each event. He may communicate future flood levels to all property owners in the flood zones and using this information, some owners may or may not flood-proof their properties, depending on whether their insurance policy will cover damages and depending on the benefit/cost of flood-proofing efforts.

After the first flood event, it will be revealed from crowd-sourced drone video footage that the hydrologist’s models were not perfect - in fact the hydrologist ‘rounded down’ to AMC II conditions in the hydrology model to predict peak flow, underestimating the actual antecedent moisture before the first flood, instead of the wet-antecedent condition AMC III model which he thought was too conservative. The hydrologist also argued that the hydrologic model was actually calibrated with smaller storms so he never really had a hope of matching actual storm flows for a large event exactly anyway. The hydraulic model was also not perfect. It overpredicted levels in some areas an underpredicted levels in others, even once the actual vs. predicted flow discrepancies were factored out. Underpredicted model levels were later attributed to culvert grates clogged with debris during the flood that raised actual levels in some locations. The hydrologist argued that the model hydraulic model should not be blamed for these operational issues in the real world system. Also, several cars, picnic tables, pedestrian park bridges, and large dumpsters washed into some large urban channels and clogged culvert openings, raising actual flood levels above those predicted. In some ‘flat’ river reaches the hydraulic model underpredicted the flood level because the model incorporated ‘ultimate roughness’ values to characterize the overbank flow areas that were much higher than the existing values. Lastly, operation of a couple small dams may not have occurred exactly as predicted in the model. You win some you lose some.

Some roadways will wash out during the flood and a provincial inquiry would be held to determine who was responsible. Justice Riviere will conclude that in many cases the flood exceeded the design capacity of the roadway and so wash outs could be expected. He’d struggle with the definition of a ‘dam’ because expert witnesses representing municipalities argue that road embankments are not dams. The railway companies would hire the most expert of experts and argue their embankments are not dams either. In the end, Justice Riviere would recommend mandatory screening of roadway embankments for dam safety based on semi-quantitative risk assessment (like Ontario did for Drinking Water Source Protection) to be developed by a group of provincial ministries and newly conceived Flood Protection Boards. We would wait a long time for the regulations, director rules and guidance documents under the new Ontario “Roads are dams. Yes, it’s a ‘Thing’ “ Act.  The hydrologist would retire well before this inquiry was over.

The Musician

The musician would write songs about the drought of 2041-45, rhyming ‘dust’ with ‘rust’, and  the great Flood of ‘67, quietly wishing for an earlier flood year, like one with more long vowel sounds to draw out like ‘49 - or a ‘54 flood, “that would be much better alliteration - flood o’ fifty-four” he mused. Hearing about the upcoming droughts, the musician’s brother would open a successful lawn painting business like they have in Las Vegas.

Municipal Engineers

Municipal engineers would talk to their managers about upcoming training in Banff to attend a 1-month symposium to learn about how to download the new future data sets, how to do statistical analysis on it, and how to consider it in updating design standards. Why Banff? Its beautiful and everyone can go on nice hikes after the sessions of course. And that funny musician who sings about future weather events will be there too for entertainment on the ‘ice-breaker’ opening night. They heard his repertoire includes old stuff like When the Levee Breaks, but a nice acoustic version. Why statistical analysis? Because most undergraduates have limited competence in this and many practitioners do not use statistics, but rather apply simple conservative fixed, deterministic design values in their everyday practice. It would be worthless trying to apply the future data without considering the variability.

Municipal Engineers’ Managers

The municipal engineers’ managers would go to Banff, not the engineers.  The private sector managers would leave staff behind to work and ‘pay the bills’, and public sector municipal engineers would approve their own training requests. When they returned they would know this new future data represented a dilemma and they would hold a meeting of the secret society of municipal engineering managers, cleverly named the “SWMinatti” during an earlier BEvERage-infused meeting. They said it stands for Storm-Water-Management-Is-Never-A-Truly-Tested-Initiative. They laughed but their wives just shook their heads once this was shared at home.  At this special meeting they would discuss how they would ‘come clean’ in the eyes of the public and regulators and dew-eyed engineers-in-training about the ‘real’ state of practice in municipal engineering design and why they had absolutely no appetite for the new, perfect, high resolution, future climate datasets. The future datasets would mean their municipal and stormwater management practice could now be ‘truly tested’.

The municipal engineering managers would meet. They’d initially joke about the “Weather Zoltar” machine that knows all and tells all - and question if it was really predicting accurate temperature and precipitation on a 10 m grid scale across Canada every 5 minutes for the next 84 years? It turns out it was indeed. How? Early on, several GTA conservation authorities got together and made a $20M grant request to confirm the Weather Zoltar predictions and installed high-precision climate stations every 10 metres in a test catchment north of Toronto - the grant request was immediately fulfilled. The total precipitation measurements every five minutes for 2 years of continuous monitoring were within 3 percent of the Weather Zoltar predictions but the extreme rainfall intensities were less accurate given the bias in equipment measurements. So just a minor measurement error. Temperatures were “bang on” to the first decimal except at the stations affected by shading. Weather Zoltar did know all.

Using a napkin, the municipal engineering managers would start a list of the things that they do in planning, evaluation and design that could be shown to be questionable by the Weather Zoltar data (i.e., overdesigned or underdesigned) and the things that would really benefit from the precise future climate and weather datasets. They would start with things they design for new developments and decide on a code for evaluating issues with each. The code “Not a Thing” meaning that the future data would be of little value, and the need for adaptation for the component design was unnecessary. The code “Might be a Thing” meaning some analysis could be pursued to confirm if any adaptation was required:

New Development System and Component	Will Future Weather Datasets Help? / Why? / How?
New Storm Systems
Draft plan lot layouts	No, lots are far beyond river floodplains (no events above Hurricane Hazel freeboard before 2100). Seems reasonable as Pielke has shown decreasing tropical storm frequency and intensity in the US. Adapting to larger floodplains affecting new development limits - Not a Thing.
Stormwater pond sizing	No, so many uncertainties in the subdivision design that pond blocks are oversized in the draft plan. Plus ponds have spillways with freeboards to handle flows well in excess of current 100-year design. Adapting new pond sizes for quantity control - Not a Thing. What might be a Thing? Review design hyetographs for existing weather - if conservative for existing, adaptation to future weather not expected to be a Thing.
Storm sewer sizing	No, pipes flow partially full with today’s 100-year and predicted higher intensities and peak flows will be throttled by inlet control devices. If some enter the storm sewer system (exceed today’s 100 year design flow) it will be accommodated in the freeboard to the basement elevation (basement slab’s 1 m above HGL). Adapting new storm sewer sizing for basement flood reduction - Not a Thing.
Local culverts	No, they are sized for 10 year events so they will overtop 6 times in their 40 year design life vs 4 times in the old climate. But it is sort of an arbitrary “even number” design return period anyway (as humans evolved with 10 fingers and toes probably and gave us this base 10 number system). If culvert was something critical, with consequences of failure, it would be sized to a higher design standard. Adapting new small culvert sizing to manage overtopping - Not a Thing.
Storm outfalls	No, they are susceptible to erosion wash-out by the receiving watercourse. Ideally they are set back from the meander belt-width with local connecting channels to limit risk of wash-out. They are often ‘fail-safe’, meaning some headwalls can fall into the creek with little consequence. Knowing future weather could be used to assess different future shear stresses in the creek but there is so much uncertainty in selecting critical shear stresses for reaches, etc. that no different design action beyond common sense set-backs would be followed with the perfect future climate data, imperfect derived flow data, and highly uncertain derived stresses and resulting vertical or lateral migration rates of the watercourse. Adapting new outfall design to manage wash-out risks with future weather erosion stresses - Not a Thing. Add changing freeze-thaw cycles to above … even less of a Thing.
Overland drainage sizing (on roadways)	No, these are often very overdesigned for yesterday’s weather. In one example, the ‘Rouge 4A’ subdivision, there were 2 critical overland flow evaluations points in design. At “LP#1” the overland capacity was 3.70 cms, 276% of the 100-year design flow of 1.337 cms. And at “LP#2”, the overland capacity was 1.85 cms, 791% of the 0.234 cms design flow. There is plenty of spare capacity. Adapting new overland drainage systems to prevent spilling / flooding - Not a Thing. The engineers agreed that ‘overland flow climate adaptation’ was “not a thing” to worry about due to existing overdesign in the GTA. But agreed that some considerations should be made in SW Ontario where a saw-tooth road grading pattern keeps all the runoff on the road - some sensitivity analysis toward storage in those systems would be worthwhile to see if freeboards are adequate to store more runoff from higher Weather Zoltar events. What might be a Thing? Check and modify freeboard on-road storage design standards to accommodate future weather.
Storm pumping stations	No, these are not common in the GTA except in SW Ontario they are used to empty ponds or drains when receiving water levels are high. The pump stations operate only under certain conditions. If lake levels or watercourse/municipal drain levels increase, pumps will operate more frequently. Adaptation of new storm pumping station capacity to future weather - Not a Thing. What might be a Thing? Review design hyetographs used for pump storage - currently conservative 6-24 hour Chicago events evaluated. Would a future time series of more extreme weather, presuming lake and watercourse levels could be predicted, such that there is a need to change the size of pumps? Or would longer upstream roadway flooding be acceptable (existing 36 hour drawdown period is used for design). What Is a Thing? Design standards for resilient power supply and back-up capacity should be reviewed, updated as required considering critical features such as some transportation routes, etc (e.g., served by pumping station pumping underpasses).
New Sanitary Systems
Local sanitary sewers	No, sewers are over designed with excessive dry weather peak flow rates and peaking factors and sewers are designed to flow partially full. Limitations with current design include deterministic infiltration and inflow allowances that do not account for extremes. But because extraneous flow stresses in new fully-separated sanitary sewers are limited (100 year I&I rates an order of magnitude below partially-separated system rates) systems have an intrinsic buffer against surcharging and back-up. Also, many municipalities require backwater valves on the sanitary lateral, such that in a rare event, Adapting sanitary sewer capacity for future weather - Not a Thing. What might be a Thing? Thorough review of I&I allowances in design. A doubling of current I&I rates may be in order, at least though a check storm? Some case study subdivisions should be evaluated to confirm if this potentially “Is a Thing”. This is required regardless of the future climate data trends.
Sanitary pumping station	Yes, we found a thing foreshadowing even bigger things in existing systems! Pumping stations are designed for dry and wet weather conditions. Like local sanitary sewers, pumping stations are designed with infiltration and inflow allowances that may not reflect today or future weather’s extraneous flows. The consequences of failure are significant in terms of back-ups / flooding or environmental impacts due to overflows/by-passes. Adapting new sanitary pumping capacity to today’s and future weather - It’s a thing.
Sanitary trunk sewer	No, these are typically deep with no property connections such that existing or future weather extremes do not have consequences in terms of flood or environmental impacts. Dry weather flow rates and peaking factors accumulate in trunk design, resulting in excess capacity compared to conservative design values. Adapting new sanitary trunk capacity for future weather - Not a Thing.
New Water Systems
Local watermain distribution system	No, fire flow scenarios governs watermain size. While future higher temperatures may increase irrigation requirements and peak hour demands, these do not govern design. If governing demands were to increase as a result of irrigation, demand management would be more cost effective than system upsizing especially given the water quality / public safety issues of oversizing systems and reduced chlorine residuals. Adapting new local watermain sizing for future weather and irrigation demands - Not a Thing.

So the overall assessment of new development servicing is that storm systems are full of resiliency now and would not require adaptation for future. The municipal engineering manages raised a glass. They decided that system components could be checked with future IDF values, but may not require upsizing if existing safety factors, such as HGL freeboard in storm sewers are adequate and overland capacity remains excessively conservative - someone should get federal funding and do a detailed model of a new subdivision to show this with the future climate data they all agreed. The acknowledged that some components such as pumping stations and ponds designed using hydrograph methods would benefit from a review of design storm distributions to ensure conservative design.

Their overall assessment of new development sanitary systems suggested that design approaches for considering inflow and infiltration should be reviewed to account for existing extreme weather stresses as well as future ones. But in new developments this seems to be a low risk area - new builds do not have infiltration inputs like old ones, nor the inflow ‘pathways’, like downspouts to foundation drains or unmanaged overland inflows. Again, they agreed someone should apply for federal funding to analyze a new subdivision to confirm this so they could put the adaptation question to bed.

All of a sudden, a giant asian carp leaped out of the channel - they were meeting at the Keating Channel Pub & Grill - and ‘took’ out the littlest manager. BAM! Bonked him right in the head. Since he was from a city with mostly new development it didn’t matter that he was now out cold on the patio for the rest of the discussion because his systems seem to be in good shape with a bit of review of safety factors in existing design approaches - it was noted that the University of Waterloo’s Intact Centre for Climate Adaptation was developing a  Flood Resilient Community Design Guideline to identify high level master planning requirements and local storm, overland and sanitary design objectives to limit flooding - engineers could follow a simple checklist to see if they are following this good standard practice approach in their planning requirements in their Official Plans and engineering design standards. If they do this, there is little need to focus further on climate adaptation in new developments.

So the municipal engineering managers would feel pretty good, except for one. He worked in a city with little new development but a lot of old pre-1970’s development and infrastructure built to early limited design standards. His systems included many storm drainage systems built with no overland drainage system, and many combined and partially separated sanitary sewer systems. The managers all agreed that they spend most of their time and effort on these old limited standard systems - these are the ones where there is ongoing litigation or claims regarding system performance and damages. The ‘old development city’ manager said his city was undertaking basement flood management studies in all these old areas, completing remediation projects to upgrade service levels, and undertake billion-dollar strategies for CSO management and operational improvements. He said he saw a great blog that analyzed the historical floods in 2000, 2005 and 2013 and correlated the flood density to the era of construction - so these is a lot of variability within the subset of older systems, with combined CSO systems generally having lower flood risk due to the surcharge relief while partially separated systems have the highest flood density. The manager with a balance of new and old development echoed this observation on diversity, saying he has litigation ongoing for flooding in one old area -never new areas - and all his ongoing storm flood remediation projects are in pre-1960’s areas. All the sanitary I&I reduction efforts focus on the high extraneous flow old areas too - nothing is done in the new developments because when we get 25-year storm, we don’t get calls from new areas. Climate adaptation in new areas is really ‘Not a Thing’.  Bottoms up!

The SWMinatti found another napkin on the next table and started to evaluate the infrastructure components in old, existing developments to see how many things they could find to adapt given the perfect Weather Zoltar data now in hand:

Old Development System and Component	Will Future Weather Datasets Help? / Why? / How?
Old Storm Systems
Storm sewer system	No, systems may be built to 2 year storm capacity, so there are predominantly exiting weather risks. Upgrades to today’s conservative 100-year storm are often ‘maxed-out’ within right of ways meaning bigger upgrades at not always feasible. Some system upgrades are not cost-effective and do not meet the Council approved threshold for implementation funding. Larger upgrades would introduce more utility constraints and expensive relocations, deeper systems with higher marginal costs due to dewatering requirements or more more costly unconventional construction methods. Small marginal incremental benefits of larger upgrades for future weather would have to be measured against high marginal incremental cost, and low benefit/cost ratios. This cost/benefit analysis should be completed against the backdrop that normalized catastrophic losses are not increasing in Canada considering net written premium growth. Adapting old storm sewer capacity to prevent basement flooding - Not a Thing. Why? Because it’s already a big expensive, constrained thing under existing weather (i.e., when cities upgrade to 100 year level of service for today’s weather). A spike in catastrophic losses - Not a Thing when GDP growth or premium growth are factored in, suggesting no economic driver to address damages beyond those associated with existing extremes. What might be Thing? Review design hyetographs for existing weather - if conservative for existing, adaptation to future weather not expected to be a Thing. If not conservative, further upgrades may be revealed to be constrained physically, financially, or from an incremental benefit/cost sense.
Storm outfalls	No, see outfalls under new development. Old outfall siting intrinsically more susceptible to wash-out under either existing or future weather. Rehabilitation / protection required to address existing risk. Adapting old storm outfalls to future erosion stresses - Not a Thing. Systems are intrinsically highly vulnerable under existing weather due to siting.
Overland drainage sizing (on roadways)	No, see storm sewers above. What might be Thing? Mapping and managing overland flow paths through existing urban areas to guide infill development risk management. Use JBA Risk 2D overland mapping (GRID format) or readily available provincial conditioned DEM overland drainage features (vector format).
Old Sanitary Systems
Local sanitary sewers	No, see old development storm sewers. Sanitary systems are constrained like storm. Upgrades consider a 25-50 year historical storm design standard. Future weather will not change the historical standard. What might be a Thing? Review historical design hyetographs for existing weather. Complete cost-benefit analysis to determine if alternative design standard can be justified for more extreme existing weather or future weather.
Sanitary pumping station	No, see local sanitary sewers above. Standard practice for sanitary pumping station design in an old development with high extraneous flows would include the evaluation of overflow / by-pass devices, and I&I reduction. Peak flows under existing or future weather would not necessarily be accommodated in the pumping station. Adapting old pump station capacity in a high extraneous flow system - Not a Thing.
Sanitary trunk sewers	No, old development trunk sewers in valleys are often highly susceptible to natural erosion processes, downcutting and lateral migration of watercourses. Significant investments in remediation and protection are ongoing. Key existing challenges are access for ongoing operation and maintenance and lifecycle replacement of features in constrained valleys (property constraints, topography constraints, environmental constraints). Adapting old sanitary trunk sewers for future watercourse erosion stresses - Not a Thing.
Regulator weirs	Yes, the operation of regulator weirs could be greatly optimized to minimize CSO’s and/or limiting basement flooding with minute by minute weather predictions. Well not quite. The city’s hydrologic models even with perfect rainfall inputs, predict peak flows within a range of -10% to +25%, so CSOs could be better managed but perhaps not optimized. Trade-offs between environmental impacts (aquatic habitat, beach closings) and flooding impacts would have to be made, with one objective satisfied at the expense of the other. And some regulators would not be adjustable remotely or in real time. Adapting sanitary system operation to minimize CSO’s, and/or limit basement flooding could optimized for some components having real time control capabilities is a Thing. Don’t forget though that Weather Zoltar with minute by minute rain predictions is not a Thing, this is fiction.
Real time CSO controls	See above.
CSO management strategy	No, the city’s strategy is already build on a historical continuous period rainfall record that virtually eliminates CSO’s. Modifying the strategy to account for some other future extreme years, seasons or weeks would add considerable expense with marginal benefit compared to baseline CSO elimination with existing weather. Adapting a CSO strategy that already eliminates CSO’s to future weather - Not a Thing.
Wastewater treatment plant	Yes, perhaps. Presumably, the future gridded 5 minute rainfall could be put in a calibrated model to transform it into precise wet weather flow at the plant to support optimized operations to maximize treatment efficiency and minimize by-passes. But then again, perfect rain data will not yield perfect flow data at the plant anyway - there is so much scatter in the long term GWI flow response to precipitation, and the short term RDII flow response, plus uncertainty with the macro-scale groundwater systems and foundation drainage (aka mysterious “urban karst”) and surface drainage/wastewater system hydraulic interaction driving inflows during extreme weather, or micro scale interactions between surcharged foundation drains, leaky floor slabs and sanitary floor drains. WWTP management could be marginally improved with perfect precipitation data, but that perfect data will then reveal the uncertainties in the other complex processes (precipitation-extraneous flow transformations and processes that we don’t even have terminology to describe.

The municipal engineering managers put the old development napkin notes down and scooped up the last nachos. The new development city engineer on the patio was still breathing so that was good - they wondered if they may have to cover his part of the bill - not good.

They would sum up the old system adaptation needs observing that old systems have existing capacity limitations intrinsic in their design. They can have very low levels of service - CSOs can occur many times a year, sewer back-ups in most chronic areas can happen for small return periods, never mind extreme events. Fortunately there are large scale programs and projects aimed aimed at remediating existing issues and improving level of service. Unfortunately these improvements also have physical, financial and environmental constraints.  As a result the Weather Zoltar future climate details will not change how these existing issues are managed. Low cost measures like inflow reduction through cost effective downspout disconnection will continue whether future rainfall is more or less extreme. High cost measures, like upgrading storm sewers to convey 100 year events will continue regardless of future weather, and these upgrades will be constrained physically and financially. The CSO management strategy that will virtually eliminate CSOs will continue whether future seasonal rainfall patterns are more or less variable, or have a few more extreme in some years - it is needed for operational purposes as well. Erosion protection for intrinsically vulnerable features like sanitary systems in valleys will continue as well - design is based on conservative practices like taking the recommended armouring size and doubling it due to inherent uncertainties in current design practice and past experience with wash-outs.

The municipal engineering managers agreed that the fundamental drivers for climate adaptation should be reviewed. They questioned the common belief from their Environment Office staff that storms are becoming more intense or occurring more frequently - Environment Canada’s own Engineering Climate Dataset version 2.3 and their regional analysis of short duration rainfall shows no detectable trends and in some regions the statistically significant trends are downward. There seems to be an ‘availability bias’ in the media and among those with limited scientific background to list a few extreme events and cite this as statistically relevant information to act upon - the engineers agreed that the plural of anecdote was not data and that data-driven, evidence based policies are needed. To test this our they asked the bus boy if he thought storms and flooding were getting worse - he said that he worked at this Keating Channel pub for 4 years and there are a lot of floods on the local roads so yes, it rain must be getting worse - and look at the flood they just had in Windsor. One of the engineers joked that flooding has been happening since the 70’s .. the 1870’s based on the flood inquiry report for the system - the bus boy didn’t get it. They would have to tip well to make up for harassing him. They agreed that someone should get some federal funding to communicate the historical and regional trends data so engineers know if they are in a higher change zone or if their raw weather data needs any safety factors applied to account for intrinsic biases (short uncertain records with sample bias, raw data not corrected/adjusted for daily measurements). That would support engineers in deciding if they need to update IDF curves. One engineer suggested that this could be a reality check for some Environment Office staff who keep citing now-discredited insurance industry rain trend claims - that storms that happened every 40 years are happening every 6 years - that has been shown to be a theoretical shift in a bell curve and not real Environment Canada data as cited.

The municipal engineering managers agreed that the future climate Weather Zoltar data took away one important aspect of uncertainty in infrastructure planning, analysis and design. What it exposed was all the other sometimes more significant uncertainties that go into infrastructure planning, analysis and design like transforming rainfall to runoff given variable antecedent conditions, or transforming rainfall into sanitary inflows or moderate or slow-response groundwater infiltration responses, or water levels in infrastructure and the routing and storage of peaks. Precise exact future climate data allows engineers to put a sharp point on one part a big, blunt instrument called hydrologic and hydraulic modelling. Maybe someone should get some federal funding and demonstrate these uncertainty factors and how they all work together and show that climate uncertainty should be considered along with all the other uncertainties - for what system components is climate important and how can it be readily addressed. After all we only have exact future climate data available to 2100 and after that we have to make some assumptions about how to handle later uncertainties.  

Lastly, the managers agreed that what is really needed is a review of the economic drivers for climate adaptation. Since US data shows tropical storms are less severe and frequent and there is local data on decreasing intensity trends, like from convective storms, there must be other drivers for increasing catastrophic losses. The growth and intensification of urban areas could be reviewed - one engineer suggested his initial GIS analysis showed watershed urbanization increasing 100% a decade compared to research showing peak rain intensities increasing 1-2% per decade. Those drivers have to put into perspective so that remediation actions can be prioritized. Analysis in the US showed flood losses normalized for GDP growth were decreasing - increases could be explained by growth and more insurance market penetration. This means the issues of flooding is still significant, but it is not ballooning out of control such that the investment in existing flood mitigation or future flood adaptation should baloon as well. One engineer shared that he normalized the Canadian catastrophic loss trends from 1990 to 2015 using person property premium totals and there is no normalized upward trend. The economic drivers could be considered as part of comprehensive cost-benefit analysis for infrastructure planning and could help refine budget thresholds currently being placed on rehabilitation projects.

The Hydrogeologists

The hydrogeologist (just one - they don’t have friends to hang with) looked at the infographic that came with the future weather datasets and laughed. “What am I going to do with this? So what if I have the rainfall and temperature data. Yes I want to refine my recharge estimates that drive my groundwater model, but the biggest part of the water balance, the evapotranspiration, is still a great big hairy unknown. My empirical equations are really really rough as it is now. Anyway, before Weather Zoltar, we did a sensitivity analysis on future climate for a really stressed groundwater water supply system. We found that most scenarios gave us more recharge because of more precipitation and less frozen soil and that resulted in increasing groundwater levels at the municipal well. The assumption that climate change would automatically decrease groundwater levels, starve baseflows and necessitate costly infiltration BMPs is really unfounded. The Weather Zoltar data confirms it. I’m always amazed at these municipal engineers - they have these cartoonish representations of water budgets for cities and claimed in the past that urbanization would decrease baseflows - what do we see in our watersheds? - increasing baseflows. The same was predicted for baseflows with climate change early on and now we show that is unfounded. If groundwater levels are going to drop - and I don’t know if they will, I don’t have an evapotranspiration Zoltar, or a perfect representation of the groundwater systems - then just monitor the situation and drill a deeper well if you ever need to. It would be way too expensive to change what we are doing now based on uncertain impacts, even with this perfect Weather Zoltar data. Done.

***

So what do the hydrologists, municipal engineers, and hydrogeologists do when they ‘catch the car’? When they have perfect future climate data to use? They find that they have significant design safety factors in new systems and take the opportunity to look at weather closely and test assumptions on design storms. Some further study is required to confirm these safety factors. They check if their current local or regional IDF data has been trending higher and needs updating. And they realise they could do these things even without the Weather Zoltar data. The also realise they have a lot of other outstanding uncertainties in any planning, analysis and design. Even perfect future climate data does not help address uncertainties in processes resulting from climate data (e.g., water balance losses, runoff and infrastructure flows). In old systems there are significant physical and financial constraints under existing conditions such that future weather may not change their strategies significantly. Robust cost-benefit analysis as part of multi-objective risk-based decision making is therefore required to guide any adaptation measures that would increase infrastructure investment considering climate impacts. This analysis should consider some decreasing regional trends in extreme rainfall and relatively ‘flat’ trends in losses relative to economic growth and insurance premium growth.

(c) R.J.Muir, Toronto - 2017


PS - today there is a considerable amount of effort prognosticating about future intensity-duration-frequency curves - what will they be? Unfortunately this is not converging. The questions we should be asking is what if you knew what they would be exactly ... or better yet what all future rain patterns would be? The possible answers in the Weather Zoltar story show us that once you know the exact future rain, you would have to face the reality that you have wide uncertainty on the next steps in applying rainfall data, whether in deriving a synthetic 'design' storm from that IDF data (rain statistics become simulated storm), or a hydrology transformation (simulated rain storm becomes runoff) or a hydraulic simulation (simulated runoff becomes infrastructure flow), and that a fulsome economic framework does not exist for decision making related to infrastructure investments as a function of system performance (e.g., flood damage losses, environmental issues, etc.... i.e., simulated flow becomes flood depths and potential damages/losses) what is the benefit/cost, what is the ROI, whose benefit? whose cost?. The good news is that most municipalities have a couple decades of obvious remediation work to do based on what we clearly know already about today's rainfall - they should get on with it - the needs are largely in pre-1980's subdivisions with design limitations related to riverine flood risk management (encroachment/enclosure of large channel/watercourses), wastewater system design (high extraneous wet weather flow stresses from foundation drains etc), and storm drainage system design (no explicit major overland flow design, limited minor system / sewer capacity).

Make a wish. All future rainfall details .... if your wish is granted then you'll need a few more arcade machines to predict
how the rain becomes runoff, how runoff becomes infrastructure flow, how flow becomes flood depths, how flood depths
becomes losses, and how to determine the appropriate economic investments in infrastructure to address the issues ... oh,
and another machine to predict future interest rates to support the discount rate in the economic analysis.