Case Studies on Climate Change in Flood Mapping

Executive Summary

The objective of this document is to help practitioners design and carry out a climate change assessment for flood hazard mapping and to highlight work that has been performed by flood mapping practitioners across the country to incorporate climate change into flood hazard studies. The document provides a flexible framework for assessing and incorporating climate impacts into flood hazard mapping, even when new information is constantly becoming available.

The guide draws on related documents in Federal Flood Mapping Guidelines Series, on-going research and expertise at Environment and Climate Change Canada (ECCC) and in the Flood Hazard Identification and Mapping Program (FHIMP), and the experience of flood mapping experts across Canada. Practitioners are encouraged to apply and adapt the assessment framework according to the purpose of their project, local factors influencing flood hazards, and available data for the area.

Section 1 introduces different types of uncertainty and their relevance to climate change, and describes how an adaptive approach to understanding flood hazards can help us prepare our communities to be more resilient in the long term.

Section 2 presents the assessment framework, one that practitioners can adjust to suit unique local conditions. The framework is comprised of four guiding principles and a five-step, actionable process. The section also provides examples, scenarios, and special considerations for practitioners to be aware of when designing their own assessments.

Section 3 presents case studies that illustrate different approaches to climate assessments and recent developments in local practices across Canada.

Appendix A presents relevant resources related to flood mapping in Canada.

Table of Contents

• Executive Summary
• Table of contents
  • List of figures
  • List of tables
• Acknowledgements
  • Notice
  • Context
  • Federal Flood Mapping Framework
  • Federal Flood Mapping Guidelines Series
  • Guideline Summaries
  • List of Acronyms
• 1.0 Introduction
  • 1.1 Purpose
  • 1.2 Climate Change and Deep Uncertainty
  • 1.3 How to Use This Guide
• 2.0 The Assessment Framework
  • 2.1 Guiding Principles
  • 2.2 The Assessment Process
    • 2.2.1 Project description
    • 2.2.2 Background review
    • 2.2.3 Conceptual model
    • 2.2.4 Analysis
    • 2.2.5 Recommendation
  • 2.3 Special Considerations
  • 2.4 Conclusion
• 3.0 Case Studies
  • 3.1 Nahanni Butte: Flood Hazard Mapping
  • 3.2 Comox Valley Regional District: Coastal Flood Hazard Mapping
  • 3.3 Cowichan Lake: Shoreline and Lake Level Flood Mapping
  • 3.4 Kootenay Boundary: Flood Mapping
  • 3.5 Nisg̱a’a Nation: River Basin Flood Hazard Mapping
  • 3.6 Rat Creek: Riverine Flood Mapping
  • 3.7 Conservation Halton: Local Waterbody Flood Hazard Mapping
  • 3.8 Durham Region: Riverine Crossings and Roads Flood Hazard Mapping
  • 3.9 Mimico Creek: Erosion-Related Flood Mapping
  • 3.10 Gaspereau Primary Watershed: Municipal Flood Line Mapping
  • 3.11 Nova Scotia Sissiboo-Bear Rivers: Primary Watershed Flood Line Mapping
  • 3.12 Prince Edward Island: Pluvial Flood Mapping Project
  • 3.13 Newfoundland and Labrador: Climate Change Flood Risk Mapping Study
• 4.0 References
• Appendix A: Recommended Resources

List of figures

• Figure 1: Flood mapping framework.
• Figure 2: Guiding principles for the climate assessment.
• Figure 3: The risk triangle, where risk is a product of hazard, exposure and vulnerability (UNDRR, 2015). Hazard is referred to as “dynamic hazard” due to the changing flood hazard from climate change.
• Figure 4: Conceptual model of impacts of climate change on peak flow in a forested plains river system.
• Figure 5: Conceptual model of impacts of climate change on peak flow on a mountainous watershed experiencing rain-on-snow.
• Figure 6: Conceptual model of impacts of climate change on a large lake system.
• Figure 7: Flood frequency analysis results for 30-year block analyses on one climate model trace for the Lillooet River near Pemberton, BC. RP = Return Period. Data source: Pacific Climate Impacts Consortium.
• Figure 8: Non-stationary flood frequency analysis results on one climate model trace for the Lillooet River near Pemberton, BC. RP = Return Period. Data source: Pacific Climate Impacts Consortium.
• Figure 9: Flow chart of a workflow of climate and hydrologic model informed flood hazard assessment. Adapted from Associated Engineering (2021b).
• Figure 10: Case study locations.
• Figure 11: Nahanni Butte project location.
• Figure 12: Sensitivity analysis results for cool/wet climate conditions (South Nahanni River).
• Figure 13: Sensitivity analysis results for cool/wet climate conditions (Liard River).
• Figure 14: Comox Valley Regional District project location.
• Figure 15: Coastal area analysis.
• Figure 16: River analysis process.
• Figure 17: Example of a regulatory flood map from the CVRD (FL = Flood Level).
• Figure 18: Cowichan Lake project location.
• Figure 19: Cowichan Lake and Cowichan River watersheds.
• Figure 20: Average modelled seasonal inflow to Cowichan Lake for past and future climate.
• Figure 21: Sample of simulated lake levels for existing weir and proposed raised weir.
• Figure 22: Project horizontal change in natural boundary location with proposed weir upgrades.
• Figure 23: Kootenay Boundary project location.
• Figure 24: Kettle River watershed and Rock Creek to Midway study reach.
• Figure 25: Hydrologic regime of the Kettle River at Ferry (08NN013).
• Figure 26: Change in annual peak flow (daily average flow) over time from the Raven hydrologic model developed in AE (2021) for the Kettle River near Ferry (WSC 08NN013) shown as a two-dimensional histogram.
• Figure 27: Change in day of year of peak flow over time from the Raven hydrologic model developed in AE (2021) for the Kettle River near Ferry (08NN013) shown as a two-dimensional histogram.
• Figure 28: Nisg̱a’a Nation project location.
• Figure 29: Nass River Basin boundary.
• Figure 30: Monthly annual precipitation for the historical, mid-century, and end-of-century periods.
• Figure 31: Monthly SWE by elevation band for the historical, mid-century, and end-of-century conditions.
• Figure 32: Plot of annual peak for the historic, mid-century and end-of-century periods, showing a transition from spring freshest-dominated peaks historically to winter peaks in the future.
• Figure 33: Rat Creek project location.
• Figure 34: Location of Rat Creek gauge, west-southwest of Drayton Valley, Alberta.
• Figure 35: Rise of Rat Creek stage before loss of transmission.
• Figure 36: Flood frequency analysis.
• Figure 37: Conceptual model depicting the direction of change of factors that contribute to peak flows in the Rat Creek gauge watershed.
• Figure 38: Change in annual maximum flow over several return periods (Dibike et al., 2019).
• Figure 39: Inundated area due to a 100-year event (left), inundated area due to a 20% climate change adjusted 100-year event (middle), and inundated area due to 580 m3/s flow (right).
• Figure 40: Conservation Halton project location.
• Figure 41: Study area and sub-watersheds.
• Figure 42: Durham Region project location.
• Figure 43: Crossings in Frenchman's Bay watershed for existing climate conditions.
• Figure 44: Crossings in Frenchman's Bay watershed for existing and future climate conditions.
• Figure 45: Mimico Creek project location.
• Figure 46: Five-year streamflow volume by month, three GHG scenarios, and two timeframes.
• Figure 47: Change in Ontario summer average temperatures for four RCPs, 1981 through 2100.
• Figure 48: Ontario summer average temperatures for four RCPs, 1981 through 2100.
• Figure 49: Gaspereau Primary Watershed project location.
• Figure 50: Nova Scotia Sissiboo-Bear Rivers project location.
• Figure 51: Sissiboo-Bear study area.
• Figure 52: Flood lines at Haight Brook, Digby Neck, along Highway 217 under current conditions. [5% blue, 1% red AEP.]
• Figure 53: Current conditions 1% AEP riverine flood depth mapping at Weymouth.
• Figure 54: Current conditions 1% riverine flood velocity mapping at Weymouth.
• Figure 55: Current conditions 1% AEP riverine flood hazard mapping at Bear River (m x m/s).
• Figure 56: Prince Edward Island project location.
• Figure 57: An example of culvert representation in Charlottetown and testing results showing water passing through the culverts.
• Figure 58: A comparison of the 100-year return period pluvial flood maps along with the highway in Summerside under the same current climate condition between the island-wide model and the municipal model.
• Figure 59: A comparison of the 100-year return period pluvial flood maps of downtown Charlottetown under the same future climate condition between the island-wide model and the municipal model.
• Figure 60: Newfoundland and Labrador project location.
• Figure 61: Projected sea level rise for Placentia.
• Figure 62: Projected sea level rise for the Carbonear area.

List of tables

• Table 1: The assessment framework: Summary of steps and tasks.
• Table 2: Climate Change Scenarios.
• Table 3: Selected Climate Change Scenarios for Flood Hazard Mapping
• Table 4: Summary of Potential Shoreline Impacts.
• Table 5: Climate Change Adjustments to Design Flow Estimates: Kettle River near Ferry.
• Table 6: Summary of Scenarios Used to Produce Flood Maps.
• Table 7: Current Conditions 1% Riverine Flood Velocity Mapping at Weymouth.
• Table 8: Eight Scenarios for Design of Rainfall Events

Acknowledgements

The Federal Flood Mapping Guidelines Series has been developed under the leadership of the Flood Mapping Committee, a partnership between Public Safety Canada (PSC), Natural Resources Canada (NRCan), Environment and Climate Change Canada (ECCC), National Research Council of Canada (NRC), Defence Research and Development Canada, Department of National Defence, Infrastructure Canada, Indigenous Services Canada, and Crown-Indigenous Relations and Northern Affairs Canada. A Technical Working Group on Flood Mapping formed in 2015 and comprised of parties from federal, provincial, and territorial jurisdictions, as well as the private sector and academia, has also contributed valuable input to the development of the Federal Flood Mapping Guidelines Series documents. Provincial and territorial government representatives also provided essential feedback for this publication.

The documents in the Federal Flood Mapping Guidelines Series are intended as a resource to support municipal, provincial, and territorial agencies and Indigenous communities working on flood hazard mapping. It is acknowledged that flood management in Canada is regulated at the provincial and territorial level of government, and that provinces and territories reserve the right to create their own guidelines specific to their jurisdictions.

Along with the authors at ECCC and Scribe Technical, useful comments on this guideline were provided by the Flood Hazard Identification and Mapping Program (FHIMP) Provincial and Territorial Partners, staff at NRCan and Public Safety Canada, and the consulting engineers who contributed to the case studies section.

Notice

Disclaimer of Liability

This technical documentation has been published by His Majesty the King in right of Canada, as represented by Natural Resources Canada (NRCan). No warranties or representations, express or implied, statutory or otherwise shall apply or are being made by NRCan in respect of the documentation, its effectiveness, accuracy, or completeness. NRCan does not assume any liability or responsibility for any damages or losses, direct or indirect, incurred or suffered as a result of the use made of the documentation, including lost profits, loss of revenue or earnings or claims by third parties. In no event will NRCan be liable for any loss of any kind resulting from any errors, inaccuracies, or omissions in this documentation. NRCan shall have no obligation, duty, or liability whatsoever in contract, tort or otherwise, including negligence.

Additional Information

For more information about this document, please contact the Canada Centre for Mapping and Earth Observation of Natural Resources Canada: geoinfo@nrcan-rncan.gc.ca.

Context

A community achieves an elevated level of resilience when its risks are proactively managed, it is adequately prepared for known and potential disaster events, and it demonstrates an ability to recover after such events have taken place. To become resilient, a community’s mitigation planners must first understand risks and ensure their capacity to manage those risks.

Floods are commonly occurring natural hazards in Canada and account for the largest portion of disaster recovery costs on an annual basis. Mitigating flood risks is key to increasing the resilience of affected communities. By proactively investing in flood risk mitigation activities, a community may improve its future growth and prosperity, reducing the risk of significant disaster recovery costs, productivity losses, economic losses, destruction of non-monetary cultural assets, environmental damage, injuries, and deaths.

Flooding is the temporary inundation by water of normally dry land, and it can occur in coastal and lake areas, along rivers, from stream blockages including ice jams, from failure of engineering works including dams, from extreme rainfall, rapid snow/ice/glacier melt or poor drainage characteristics, high groundwater levels, and other sources. Flood mapping that accurately delineates flood hazards, including those impacted by future conditions due to anticipated development or projected changes in climate, serves as the precondition for mitigation activities and is therefore the first step to increasing community resilience with regard to flooding. Establishing a national approach to flood mapping will facilitate a common national best practice and increase the sharing and use of flood hazard information. Considering the costs of damages and the vulnerabilities of infrastructure to the probability of the hazards provides an estimate of the flood risks, thereby improving the foundation from which further flood risk mitigation efforts can be initiated.

Federal Flood Mapping Framework

The Flood Mapping Framework consists of all the components of the flood mitigation process, from flood hazard identification to the implementation of flood mitigation efforts. Figure 1 illustrates the relationship between these different components.

Federal Flood Mapping Guidelines Series

The following documents are intended to inform individuals and organizations involved with flood management in Canada:

1. Federal Flood Mapping Framework
2. Federal Guidelines for Flood Hazard Identification and Priority Setting
3. Federal Airborne Lidar Data Acquisition Guideline
4. Case Studies on Climate Change in Flood Mapping
5. Federal Hydrologic and Hydraulic Procedures for Flood Hazard Delineation
6. Coastal Flood Hazard Assessment for Risk-Based Analysis on Canada’s Marine Coasts
7. Federal Geomatics Guidelines for Flood Mapping
8. Federal Guidelines for Flood Risk Assessment
9. Federal Flood Damage Estimation Guidelines for Buildings and Infrastructure
10. Federal Land Use Guide for Flood Risk Areas
11. Bibliography of Best Practices and References for Flood Mitigation

Guideline Summaries

1. Federal Flood Mapping Framework

This document provides background and context on flood mapping in Canada, describes a vision and principles for flood guidance, and introduces the Federal Flood Mapping Guidelines Series. It provides a summary of each of the documents in the Series and explains how each document fits into the overall framework, including its role in the flood mapping cycle.

2. Federal Guidelines for Flood Hazard Identification and Priority Setting

This document outlines methods for determining where to conduct flood hazard mapping and how to prioritize flood hazard mapping projects.

3. Federal Airborne Lidar Data Acquisition Guideline

This document is a resource for the acquisition of base elevation data from airborne lidar data undertaken across Canada. This guideline provides technical specifications to federal, provincial, and territorial departments, as well as individuals and organizations in Canada requiring information to understand and plan for airborne lidar data acquisition.

4. Case Studies on Climate Change in Flood Mapping

This collection of documents describes projects from across Canada where climate change was incorporated into the flood mapping process. It provides a framework for designing a climate assessment and examples for practitioners to draw upon and learn from others’ experiences and complements the climate change-related information and resources included in the Federal Hydrologic and Hydraulic Procedures for Flood Hazard Delineation document.

5. Federal Hydrologic and Hydraulic Procedures for Flood Hazard Delineation

This document provides guidance to municipal, provincial, and territorial agencies and Indigenous communities working to produce flood hazard maps. It provides technical information on the types of river and lake flooding, general practices, procedures for hydrologic and hydraulic analyses, procedures for incorporating climate change, and reporting best practices.

6. Coastal Flood Hazard Assessment for Risk-Based Analysis on Canada’s Marine Coasts

This document provides guidance on methodologies for coastal flood hazard assessments using risk-based approaches.

7. Federal Geomatics Guidelines for Flood Mapping

This document contains information on the different types of flood maps and outlines the technical specifications to consider when acquiring, managing, and disseminating these maps and their associated geospatial data.

8. Federal Guidelines for Flood Risk Assessment

This document provides technical guidance on conducting flood risk assessments in Canada.

9. Federal Flood Damage Estimation Guidelines for Buildings and Infrastructure

This document provides guidance on how to evaluate potential economic losses, with a focus on buildings and infrastructure, incurred as a result of flooding.

10. Federal Land Use Guide for Flood Risk Areas

This document provides guidance to the professionals leading and supporting risk-based processes and methodologies for the purpose of land use planning in flood-prone areas.

11. Bibliography of Best Practices and References for Flood Mitigation

This document contains lists of Canadian and international references and case studies pertaining to hydrology and hydraulics, climate change, risk assessment, and flood mapping. The purpose of this document is to provide a consolidated list of reference materials intended as further resources for practitioners involved in flood mapping.

List of Acronyms

AEP

Annual exceedance probability

AOI

Area of Interest

CMIP

Coupled Model Intercomparison Project

DFAA

Disaster Financial Assistance Arrangements

ECCC

Environment and Climate Change Canada

ESM

Earth Systems Model

FFA

Flood Frequency Analysis

FHIMP

Flood Hazard Identification and Mapping Program

GEV

Generalized Extreme Value

GHG

Green House Gas

HEC-RAS

Hydrologic Engineering Center’s River Analysis System

IDF

Intensity-Duration-Frequency

IPCC

Intergovernmental Panel on Climate Change

NDMP

National Disaster Mitigation Program

NHS

National Hydrological Services

NRC

National Research Council

NRCan

Natural Resources Canada

PSC

Public Safety Canada

RCM

Regional Climate Model

RFFA

Regional Flood Frequency Analysis

RP

Return Period

RRP

Risk reduction potential

UNESCO

United Nations Educational, Scientific, and Cultural Organization

WSC

Water Survey of Canada

1.0 Introduction

Climate change is caused by ever-increasing levels of CO2 (and other greenhouse gasses) in the atmosphere since the industrial revolution, leading to globally increased temperatures and more extreme weather events. Climate change is already affecting Canadian life and will continue to do so into the future, even with major and immediate global cuts in CO2 emissions (Calvin et al., 2023). One notable effect is an increase in damage from natural hazards, including the frequency and severity of floods (Public Safety Canada, 2018).

The Government of Canada’s Flood Hazard Identification and Mapping Program (FHIMP) is one response to this issue. Predicting and mapping potential changes in flood hazards into the future will help make flood maps produced under the FHIMP more effective for map users and relevant for a longer time. Provincial and Territorial Governments, city planners, emergency managers, and more can use these maps to better prepare for the future through visualizing how people and infrastructure may be impacted by flooding through the rest of the 21st century.

While the value of making flood hazard maps that account for climate change impacts is clear, the methods are not. In Canada, hydrology, and therefore flooding, is governed by complex, nonlinear interactions between the atmosphere, land, and the cryosphere. Outputs from Earth Systems Models (ESMs) provide valuable information about the earth’s future climate, but not direct information on flooding. Post processing of ESM data, or inputting climate data into a hydrologic model is often necessary to learn about future changes in flooding.

Additionally, an understanding of the processes that cause flooding in a particular area and the potential impacts on these processes from climate change is necessary to predict future flood hazard. For example—when flooding is caused by snow accumulation and melt, flood hazards will not always increase as temperatures increase. A lack of snow on the ground may cancel out increases in snowmelt rate, and the risk of flooding may actually decrease. In some cases, a tipping point may occur where a flood hazard has been increasing for years and then begins to decrease. Further, the impact of the size of a watershed has implications for flood hazards, two nearby rivers can have vastly different drainage basins and therefore experience very different hydroclimatic conditions.

Beyond the physical considerations, there are statistical issues that complicate our ability to understand flood hazards. Climate change has decreased our ability to rely on past observations to understand current and future flooding. In simple terms, one can no longer assume that the past represents the future. In statistical terms, this is a “loss of stationarity” (Milly et al., 2008). Traditional flood frequency analysis depends on this assumption of stationarity as a basis for developing design flows that feed into a hydraulic model and then into a flood hazard map. When this assumption is no longer valid, new tools and techniques are needed^{Footnote 1}. This document illustrates some of these available tools, and the Case Studies Section shows how various tools and techniques have been applied across Canada.

1.1 Purpose

The purpose of this guide is to assist practitioners with designing and conducting a climate assessment for flood mapping. As part of the Federal Flood Mapping Guidelines Series, this guide is focused on flood hazard mapping in Canada; however, it also may be useful for a broad range of water-related projects where flood hazard is a concern.

The term “practitioners”, in this case, refers to hydrotechnical and civil engineers, geoscientists, and hydrologists. It is not expected that practitioners have advanced climatological knowledge (for example, an understanding of the inner workings of ESMs or types of climate/weather model downscaling). However, this type of knowledge may be necessary in some cases, and this guide will assist practitioners in recognizing where it is necessary.

In the case of flood hazard mapping and other flood-related hydrotechnical projects, “climate impacts on a project” typically refers to the impacts of climate change on the hydrologic conditions that lead to flooding. As most flood hazard studies rely on design flows or water levels (for example, the 1:200 year or 0.5% annual exceedance probability), the guide focuses on approaches for estimating changes in these design flows and levels from the present day into the future.

1.2 Climate Change and Deep Uncertainty

The Federal Hydrologic and Hydraulic Procedures for Flood Hazard Delineation (Natural Resources Canada, 2025b) classifies uncertainty into four types.

• Natural or intrinsic uncertainty: arising from the inherent randomness of natural processes, which is variable over time and space. This natural uncertainty is difficult to reduce and quantify as the data is irreproducible.
• Data uncertainties: from measurement errors, instrumentation errors, inconsistencies and non-homogeneity of the data, data handling, and inadequate representativeness of data over time and space. This data uncertainty may be reduced with better or increased measurements.
• Calculation uncertainty: from the inability of a mathematical technique or model to accurately represent the true physical behaviour of the natural world, since the technique or model is poorly or incompletely specified, or the phenomena modelled has instabilities and non-linearities not reflected in the modelling approaches.
• Parameter uncertainties: from inaccurately assessed parameter values in the test or calibration data, due to limited numbers of observations, and statistical imprecision.

Uncertainty about the future climate of our planet contains, to some extent, all these types of uncertainty. Marchau, et al (2019) describes this as “deep uncertainty”. While the path that we are currently on is clear, it is not possible to know the specific efforts that society will make (or not make) to reduce greenhouse gases (GHGs) over the next century. This is the reason that outputs from climate models are usually described as “scenarios” rather than “forecasts.” Climatologists across the globe have developed a range of possible futures that society may take, and the GHG emissions that they result in. A suite of independently developed ESMs (also with world-wide development) then uses the GHG emission scenarios as input to predict how the climate will respond. The combination of GHG scenarios and ESMs produces an “ensemble” of climate scenarios. This ensemble of model outputs is useful for exploring what our future might look like, but it is not yet possible to make probabilistic predictions on which of these scenarios are most likely to occur.

Further, it is possible that completely unanticipated events will happen before 2100. Marchau (2019) referred to these as “black swans”. In the case of climate change this means events that cannot be predicted before they happen but have significant impact on the global climate. Past events fitting the black swan description include the industrial revolution, supervolcanoes, and asteroid impacts. Our limited ability to predict black swan events and the future actions of society that affect the earth’s climate make climate change a deep uncertainty problem. Even with more knowledge, there are still many plausible futures, many of which have not yet been envisioned.

Fortunately, the deep uncertainty of climate change does not mean there is nothing that can be done to prepare for the future. An agile, flexible approach is needed to prepare ourselves and our communities to be more flood resilient in the future; that is, an “adaptive” approach to mapping flood hazards:

• Accept future uncertainty and be willing to make changes: Realize that it is not possible to know the future perfectly. Some things we think we know now will turn out to be wrong, but this is ok. By planning for uncertainty, we can be prepared to pivot.
• Regularly reassess: Climate impacts on flood hazards cannot be assessed once and then never again. Reassessment on a regular cycle (for example, every 5–10 years or when a major event happens) will help us determine if any course correction is needed.

For further information on adaptive approaches to flood hazards (not limited to mapping), see Ebbwater (2023).

1.3 How to Use This Guide

This guide presents practitioners with a climate change assessment approach that is adaptable to unique local conditions. The assessment framework presented in Section 2 is designed to be adapted according to the purpose of a project, the availability of information, and practitioners’ skillsets and knowledge. Section 2 presents four guiding four principles and provides an actionable process in five steps. It also provides examples, scenarios, and special considerations for practitioners to be aware of when designing their own assessments.

This guide also highlights successful work by flood mapping practitioners from across Canada. The case studies presented in Section 3 illustrate the varied approaches to climate assessments and recent developments in local practices in 13 jurisdictions across the country.

2.0 The Assessment Framework

This section provides a description of the recommended assessment framework for practitioners who want to integrate climate change considerations into engineering-scale flood hazard mapping. It presents:

• Four guiding principles to steer all stages of development (Section 2.1)
• Five assessment steps, with examples and scenarios (Section 2.2)
• Special considerations to take into account when making recommendations (Section 2.3).

2.1 Guiding Principles

The four guiding principles (Figure 2) described in this section are intended to help practitioners

1. plan a climate assessment that is appropriate to the project, and
2. make appropriate recommendations in the context of deep uncertainty around climate change.

Each principle may apply to more than one stage of the process. Practitioners should remain aware of the principles throughout.

Proportionality

The climate assessment should be proportionate to the hydrotechnical project. Not every project requires the same level of effort, due to both time and budget, and because the risks of over- or under-predicting climate impacts on flooding can be different depending on the project (see “Focus on key vulnerabilities” for more on risk).

In practice: A good rule of thumb for a proportionate assessment is that the approximate percentage of effort spent on a climate assessment should remain approximately constant for a range of project sizes. For example—spending a year developing a hydrologic model and producing an ensemble of hydrologic model outputs may be appropriate when developing a flood map for a densely populated city. However, this is excessive when performing a climate assessment for sizing a culvert on a forestry road. In that case, a simpler literature review and conceptual understanding (performed over a few days) is more appropriate.

The proportionate risk should also be considered, as the risk of being incorrect can vary widely. For example—if the risk of under-predicting climate impacts could present a substantial risk to lives and property, this should be an obvious indication that spending more time (including a higher percentage of total project budget) on the climate assessment, or simply recommending very conservative (large) increases in expected flooding for the future, is appropriate. See the “Focus on key vulnerabilities” principle for more on risk.

Start conservative and refine

A simplistic climate assessment should tend toward conservative recommendations (for example, larger anticipated increases in design flows), while a detailed climate assessment should describe why it is acceptable to decrease these recommendations. Starting with conservative recommendations and refining from there also helps tailor a climate assessment (and the resulting recommendations) to a specific project.

In practice: In some cases, it could be easier, and ultimately less effort, to simply prepare for large increases in peak flows by over-predicting climate effects. For example—in an undeveloped area, assuming a large increase in peak flows and mapping a very conservative (i.e., large) floodplain to limit development may be an appropriate method. Future development would be safely kept outside of this floodplain.

Conversely, in the case of a large urban area that is already inhabited, many lives would be impacted if flood hazard maps vastly over-predict the hazard. It is likely useful to spend a substantial amount of time justifying smaller (or no) increases in expected flooding in an area by 2100. Further, the costs of building protective infrastructure or moving people under this scenario would outweigh the cost or time of a detailed climate assessment on the changing flood hazard.

Focus on key vulnerabilities

Focusing on key vulnerabilities reminds practitioners to keep risk in mind. Maintaining a risk-based perspective is valuable, given that flood hazard is a key contributor to flood risk, alongside exposure and vulnerability (Figure 3).

Though hazard is meant to be an independent side of the risk triangle, it is inappropriate to disregard risk when working in a situation with deep uncertainty. Because a range of potential futures could occur, practitioners must focus recommendations on the exposure and key vulnerabilities of the specific project. This means thinking about outcomes and working backward to identify the plausibility of very consequential outcomes that might occur in the future. This approach is often referred to as a “bottom up” approach (Kucharski et al., 2024). Considering this, at least conceptually, is a useful complement to the more standard approach of starting at the present and forecasting the future (“top down” approach).

In practice: Typically, flood preparedness is based on an assumption that floods occur at a specific time of year, during a “flood season.” In Western Canada, this often means during the spring snowmelt freshet. In other parts of the country, this might mean during the summer thunderstorm season. If floods start occurring outside the anticipated flood season, this may be more consequential than a fractional increase in the size of a flood that still occurs during the standard flood season. All the standard preparedness systems in place (for example, more staff on call, dams drawn down to hold flood waters) would not be available. It may be that a project location is more vulnerable to changes in the timing of flooding than it is to changes in the magnitude. A detailed understanding of the location and flood processes is necessary to understand this kind of vulnerability.

Establish multiple lines of evidence

Ideally, a climate assessment uses multiple approaches to lead to an appropriate recommendation. Whenever possible, a practitioner should assess climate impacts on a project in different ways (for example, literature review, conceptual model, assessment of hydrologic model output), which would individually and together point to the same recommendations. This increases confidence and lends credibility to the recommendations.

Multiple methods may not always point to the same results. Along with putting more weight towards more advanced approaches (for example, trusting hydrologic model output over simple transfer from an Intensity-Duration-Frequency rainfall curve), practitioners should consider the three prior principles and consider: “What is the downside of being wrong?” Being wrong, in this context, means either being overly conservative, or not conservative enough. A sensitivity analysis, where different climate change predictions are mapped to see the difference in flood inundation, is a useful approach to visualize the real impacts on the flood map.

In practice: This approach is also applied numerically and usually described as a “model ensemble” (for example, see the Coupled Model Intercomparison Project (CMIP) used in Intergovernmental Panel on Climate Change (IPCC) reports). Independently developed climate models are evaluated as an ensemble, and the ensemble of results is used to produce overarching recommendations.

2.2 The Assessment Process

This section is intended to assist practitioners with developing an understanding of the potential impacts of climate change on their hydrotechnical project. The recommended approach can be adapted to suit unique local and project-specific conditions. The four principles outlined in the prior section should be considered throughout the five-step process outlined in Table 1. Specific cases when they are relevant are pointed out in each section.

Table 1: The assessment framework: Summary of steps and tasks.

Step	Main Tasks
1. Project description	Define the area of interest Define the purpose and objectives of the project Describe the watershed(s) Define the flood-generating processes
2. Background review	Conduct a literature review Collaborate with local experts, as needed Identify available data
3. Conceptual model	Sketch out the understanding of the system Use the conceptual model to inform the analysis
4. Analysis	Design an analysis proportionate to the project Analysis Options Special considerations
5. Recommendation	Synthesize Steps 1–4 Provide climate-informed safety factor as a value or range Special considerations

2.2.1 Project description

Summary of main tasks:

1. Define the area of interest
2. Describe the purpose and objectives of the project
3. Describe the watershed(s)
4. Describe the flood-generating processes

Overview

Developing a complete project description is the crucial first step in the assessment process. This allows practitioners to organize their understanding of flooding at the site and provides evidence that the flood hazard assessment team has sufficient background knowledge to effectively assess climate impacts on the area of interest (AOI).

i) Define and describe the area of interest

The project description should include a thorough written discussion of AOI. The AOI is the area in which the project is undertaken, and by extension, the area for which the flood hazards need to be identified. For a flood mapping project, this may include a town or city, or it may more specifically include certain reaches of a river or stretches of a shoreline. A high-level description of the project risks (vulnerable people, assets, infrastructure, etc.) should be included in the AOI description.

ii) Describe the purpose and objectives of the project

The reason that the project is being undertaken may be to create regulatory flood maps, to define flood conveyance structures or protections, or a host of other reasons. It is likely that the purpose, as well as the objectives and AOI, have largely been defined by the project client rather than the consultant practitioner. The practitioner should have a succinct statement of purpose and AOI that defines the underlying reasons for the project, and keep them in mind throughout the climate assessment. The project purpose should also include a definition of the risks of over- or under-estimation of the impact of climate change on flooding for the project. Laying these risks out helps to shape the appropriate analysis.

iii) Describe the watershed(s)

The scale of the particular watershed(s) affecting the AOI is a critical factor in understanding the flood processes. In large river systems, the weather and climate impacts at the AOI may play only a minor role in site-specific flooding, as the upstream conditions in the watershed’s headwaters (or other runoff-generating regions) are the dominant influence on downstream flooding.

For example—if creating flood maps near Saskatoon (SK), the hydrology of the South Saskatchewan River would have very different flood processes than local creeks that flow into the river or nearby sloughs.

Along with scale, the physiography of the watershed (elevation, river morphology, etc.), along with the climate, vegetation, and landcover of the watershed all can have an impact on peak flow processes and should be understood and described.

iv) Describe the flood-generating processes

It is expected that the practitioner has (or will obtain) an understanding of the processes that cause flooding in the AOI. The project description should include a thorough discussion of the mechanisms by which flooding can occur at the site. Understanding the flood mechanisms that affect the AOI is essential for creating accurate models (conceptual or numerical), identifying flood hazards, and developing effective flood-prevention measures.

Canada is a large country with a diverse climate and terrain; there is a wide range of flood-generating processes that can be experienced throughout. For a primer on flood types in Canada, refer to Watt et al. (1989) and Chapter 6 of Canada’s Changing Climate. Examples of flood processes that can be experienced include the following:

• Severe rainfall can result in rapid surface runoff, overwhelming drainage systems and causing flooding separate from the standard waterway (referred to as pluvial flooding).
• Coastal areas and regions near large water bodies face the threat of storm surges and high lake levels during extreme weather events, leading to flooding of shorelines and adjacent areas.
• Coastal flooding due to sea-level rise.
• In colder regions, a sudden temperature rise can trigger rapid snowmelt (either during the spring freshet or mid-winter), elevating river flows and contributing to flooding.
• Ice jams in rivers can create temporary barriers that obstruct water flow, causing upstream flooding.
• Prolonged periods of rain or snowmelt can lead to elevated groundwater levels, saturating the soil and causing surface flooding when the land's water retention capacity is exceeded.

2.2.2 Background review

Summary of main tasks:

1. Conduct a literature review
2. Collaborate with local experts, as needed
3. Identify available data

Overview

Conducting the background review involves undertaking a literature review, collaborating with local experts and residents to strengthen the understanding of the study area, and identifying available hydroclimatic data.

i) Conduct a literature review

It is important to examine any work that has already been done in the AOI, which can vary significantly by region. For well-studied major rivers, such as the Fraser River (BC) or the Red River (MB), numerous published journal articles and engineering studies are likely to be available. However, this may not be the case for smaller or less studied watersheds.

Relevant literature may include:

• Scientific or engineering journal articles
• Reports by engineering firms
• Provincial or federal information papers
• Newspaper or news magazine articles
• University research (graduate theses, etc.)
• Past flood projects documented in the FHIMP historical flood events data layer

Research should emphasize learning about the dominant flood processes (examples in prior section) that have caused past flooding. There may be more than one process at the same location that has caused floods; in these cases, all flood generating processes should be assessed.

ii) Collaborate with local experts, as needed

While journal articles and studies are valuable, they may not be directly applicable to a particular AOI. Incorporating local knowledge (i.e., from experts in the area or long-time residents) can strengthen the understanding of the study area and is crucial where little to no existing data and research are available.

iii) Identify available data

Hydroclimate observations, including weather and streamflow data, are fundamental for understanding a region's current conditions. Climate projections are also important to anticipate future changes, and existing hydrologic models can provide a valuable starting point. Rainfall intensity projections, often available from various data sources, are useful for understanding potential flood risks.

In addition to these hydroclimatic factors, other future changes in the watershed must be considered (for example, urban development, forest harvesting, and riverine or coastal erosion). While not always possible to quantify, they can significantly impact the watershed and should be acknowledged to provide a comprehensive understanding of the evolving environmental context area of the area. Depending on the project location, geomorphic factors may be impacted by climate change and should be considered. For further information on geomorphic considerations in flood mapping see Natural Resources Canada (2025b).

For key data see:

• Historical climate data: Historical Climate Data Portal
• Water level and flow data: Water Level and Flow Portal
• Climate projections: ClimateData.ca
• Rainfall intensity projections: ClimateData.ca
• Hydroclimatic model output:
  • Power Analytics and Visualization for Climate Science: Ouranos
  • Pacific Climate Impacts Consortium
  • Ontario Climate Change Data Portal
  • Hydroclimate Atlas of Southern Quebec

2.2.3 Conceptual model

Summary of main tasks:

1. Sketch out the understanding of the system
2. Use the conceptual model to inform the analysis

Overview

The third step in the assessment process involves developing a conceptual model of how climate change is expected to impact the AOI.

i) Sketch out the understanding of the system

A conceptual model allows practitioners to compile and map out all potential forces of change at a high level. The purpose is to develop a baseline qualitative understanding of flood generating processes and climate impacts on these processes. It serves as a foundational step that guides the quantitative analysis to follow. Most commonly, the conceptual model is drawn out in a flowchart-like diagram that shows inputs and outputs to the AOI, such as weather, snow, wind, river flows, and illustrates the most important ways that the flood generating processes work at present and may change in the future. The understanding of this system and ability to sketch the conceptual model are developed in steps 1 and 2 of the climate assessment.

ii) Use the conceptual model to inform the analysis

If a conceptual model indicates a relatively straightforward system, then a simplified quantitative analysis may suffice. Conversely, a complex conceptual model indicates that a multifaceted analysis is likely needed to explore the interactions and relative impacts of the different processes affecting the system.

The baseline understanding from the conceptual model can also serve as a “reality check” for the subsequent quantitative analysis, enabling the identification and investigation of unexpected results from the analysis. If the general understanding derived from the conceptual model and the results of the quantitative analysis are inconsistent, practitioners may not have considered certain potential climate impacts in the conceptual model, or the quantitative analysis may not fully capture the most critical climate impacts on the AOI. In either case, a discrepancy indicates that some reassessment of the process is needed.

Conceptual models are intended to guide practitioners in decision-making and recommendations. Practitioners should develop their model as is most appropriate to their specific project and not limit themselves to any one particular approach (see Case Study 3.6). Further, a conceptual model does not need to be static. Practitioners should be revising and updating their “mental model” of flooding on a system as they learn more, more information is collected, and as climate change proceeds forward.

This section provides examples of simple conceptual models for three locations:

• a forested plains river system,
• a mountainous watershed experiencing rain-on-snow, and
• a large lake system.

Note that the illustrations in these examples were polished with the assistance of a professional designer; however, this is not necessary in practice. The practitioner developing the model should use whatever method they feel is most effective to sketch out their conceptual model of the AOI.

Conceptual Model 1: Forested Plains River System

In a forested plains river system, such as the Benchlands region in Alberta (Alberta Geological Survey, 2021), climate change is expected to increase both air temperature and precipitation (Shrestha et al., 2017).

Precipitation that traditionally fell as snow is likely to shift to rainfall. While the exact effect (and timing of that effect) on streamflow remains uncertain, changes in the frequency and magnitude of large events are expected. Additionally, rain-on-snow events may become more frequent, particularly during the early stages of warming, when a substantial snowpack still exists (Dibike et al., 2019). With an increase in temperature, an increase in atmospheric evaporative demand is expected to occur (Pike et al., 2010), which will reduce water availability in the watershed during hot periods of the year. Moreover, the potential loss of tree cover in the basin, due to forest fires and deforestation, could reduce rainfall interception from trees, resulting in increased runoff entering the stream (Nasirzadehdizaji & Akyuz, 2022).

The conceptual model points towards the total system having an overall likelihood of increased flooding due to climate change. If it is necessary to turn this qualitative understanding into quantitative numbers, hydrologic modelling is necessary.

Conceptual Model 2: Mountainous Watershed Experiencing Rain-on-Snow

A conceptual model of climate impacts on a mountainous watershed experiencing potential atmospheric river flooding (for example, in the Coast Mountains of British Columbia) is shown in Figure 5.

High-intensity rainfall events are expected to become more frequent (Kharin et al., 2007), resulting in more runoff entering the streams. Additionally, atmospheric rivers are expected to bring significantly increased rainfall to mountainous regions (Rhoades et al., 2020). An increase in air temperatures will reduce average annual snowfall throughout the watershed, especially at higher elevations; this precipitation is instead expected to occur as rainfall (Musselman et al., 2018). This effect is expected to result in increased instances of rain-on-snow events.

These impacts are expected to increase the overall likelihood for large peak flow events and flooding. The impacts of long-duration rainfall, high humidity, and high winds on peak flows in the watershed remain unclear but could contribute to further hydrological variability. The total system has an overall likelihood of increased flooding due to climate change and qualitatively indicates a more severe increase than in Figure 4. Further, a shift from mostly snowfall related flooding to primarily rainfall related flood may be possible may be possible if snowfall decreases to near zero.

Conceptual Model 3: Large Lake System

Climate change is expected to intensify wind patterns (Eichelberger et al., 2008), leading to increase wave effects on lakes (Figure 6).

This is expected to increase storm surges, bringing water higher up on the shores, which could increase impacts on shorelines and lead to flooding of any present infrastructure, and may result in erosion of the shore. With increasing wind, evaporation will also increase on the lake (Woolway et al., 2020), which will potentially lower water levels and slightly reduce flood risks. However, the increase in storm surges is expected to outweigh the increase in evaporation, resulting in an overall likelihood of increased flooding potential (Seglenieks & Temgoua, 2022).

2.2.4 Analysis

Summary of main tasks:

1. Design an analysis proportionate to the project
2. Analysis Options

The analysis step in the assessment begins by ensuring the analysis is proportional to the project, compiling historical climate and flow data. Then, the practitioner moves to actually conducting analyses of climate and hydrologic models, and if necessary, developing and using a model (for example, Case Studies 3.5 and Case Studies3.9).

There is already a large body of literature on the technical aspects of climate model output, climate downscaling, and hydrologic model linkage to climate model outputs. For a technical primer on climate change and flood hazards in Canada, see Khaliq (2019).

i) Design an analysis proportionate to the project

While the first three steps in the climate assessment framework are consistent across all project sizes, the level of effort required for the analysis may vary widely. This is due to basic logistical concerns such as the time or budget available to do the work, but also the specific needs of the project and the information that is available. The technical analysis undertaken can thus vary widely, from applying design flow changes obtained through the literature review at the simple end to developing and operating a custom hydrologic/climate model at the complex end, and every option in between.

As discussed in Section 2.1, it is recommended to follow the guiding principle of starting conservative and refining from there. This means that in some cases it may be easier to make simple yet conservative assumptions about climate impacts on flooding; that is, to assume a large increase in peak flows or water levels at the AOI and plan accordingly. In other situations, where over-estimating climate impacts is undesirable, it is worth more effort and time to close in on a more specific, and potentially less drastic, recommendation for changes in the design flood magnitude (or level).

It is likely obvious that a climate assessment would be very different when assessing the flood hazard for a city of one million people than when designing culverts for a single forestry road. There is a substantially greater number of people put at risk for the former, and thus a much greater total effort to be put into the project. The larger a project is, and the greater impact it may have, the greater the importance of the principal of multiple lines of evidence. Using more than one analysis approach lends credibility to any conclusions or recommendations drawn from the analysis.

ii) Analysis options

Our recommended technical analyses for climate impacts on flood hazards can generally be broken into three categories:

1. Direct transfer of climate model output to flood impacts
2. Use of available model outputs
3. Purpose-built model for the area of interest

Because there are no observations for the future, analysis of climate impacts on flooding primarily revolves around the use and assessment of models. Models can include earth systems models and regional climate models (ESMs and RCMs), but also hydrologic models, routing models for lake levels, and wave and storm surge models for lakeshores.

An ideal situation is that local hydrologic (or the appropriate flood hazard) model output, which has been run with climate model output, is already available for a project’s specific location. This is the case for some areas of Canada: several regional climate modelling centres have hydrologic model output that end users can download and post-process for their AOI.

When these model outputs are unavailable, or not of appropriate level of detail for an AOI, other approaches must be taken. Whether those approaches are simpler or more complex depends on the needs, budget, and risks of the project (see Section 2.1).

This section provides a conceptual description of three of the most common analysis categories.

Option A: Direct transfer of climate model output to flood impacts

In some situations, it may be appropriate to directly apply numbers from ESMs or RCMs to estimate changes in flooding. This category of assessment is only applicable in cases where the variables available in ESMs or RCMs are directly correlated with potential changes in flooding, and where a single flood hazard type exists (see the conceptual model developed in Step 3 of the assessment process). Appropriate uses of this option include:

• Using projected changes in Intensity-Duration-Frequency (IDF) curves to directly infer changes in peak flows. This is potentially applicable in small urban catchments where peak flows and runoff are directly (or nearly directly) correlated. Practitioners can also direct IDF curve changes into stormwater calculation methods, such as the rational method or into simplified event-based hydrologic models. The ECCC recommended method for climate impacts on IDF curves is described by the Canadian Centre for Climate Services.
• Using sea level rise estimates directly for a coastal location (see Case Study 3.1).
• Application of future changes in wind speed directly to a wave run-up/storm surge calculation. Refer to the Coastal Flood Hazard Assessment for Risk-Based Analysis on Canada’s Marine Coasts (Section 5.6) for further reading.

This direct transfer approach is inappropriate in cases where nonlinear relationships between climate variables and flooding exist, or when multiple flood hazards affect the same area of interest (an understanding developed while making a conceptual model). This is the situation in many parts of Canada, where cold region hydrology and an abundance of surface freshwater complicate flood generation processes. Snow accumulation and melt, with further complication through rain-on-snow, create nonlinear responses of flooding to climate change. River-ice-related processes can complicate flood response even further. Additionally, frozen ground and permafrost, and areas with multiple flood hazards, such as the confluence of two rivers, or where a river meets a lake, add further complications to a process.

Option B: Use of available model outputs

Where Option A is inappropriate, use of a model is necessary. The appropriate model is not always the same; it may include hydrological models, river ice models, and hydraulic or hydrodynamic models that are driven with downscaled output from ESMs. The technical aspects of these models are beyond the scope of this guideline, but have been covered elsewhere (Khaliq, 2019). Practitioners should use the conceptual model of climate impacts on flooding to identify when these complex interactions exist and use the background research (Step 2 of the assessment process) to determine if model results already exist for the AOI, or if applicable model results from a similar site nearby can be applied to the AOI.

Questions that practitioners should ask when determining if it is appropriate to use output from a model developed by others for their AOI include:

• Does the model represent all the flood hazards identified in the conceptual model?
• If the model was not directly developed for the AOI, does the area that it was developed for:
  • Experience very similar flood hazards?
  • Experience a very similar climate?
  • Have a very similar geographic scale?
• Did the model use a full and recent sample of climate model and carbon scenario outputs (different ESMs and shared socioeconomic pathways)? This is referred to as ensemble output, where multiple time series are produced, covering the same period.

The availability of climate-projected hydrologic model outputs is variable across Canada and is expanding rapidly. The appropriateness of these available models should still be assessed for the project at hand. Hydrologic (or other) models are not always fit for the same purposes; assessing the relevance to flood studies before use is necessary. Potential sources that practitioners can investigate include:

• Pacific Climate Impacts Consortium;
• PAVICS;
• Hydroclimatic Atlas of Southern Quebec;
• Research model outputs from academic research in hydrology and hydraulics in the AOI; and
• Data-sharing agreements between hydrotechnical consultants working in the same region.

These sources are continually expanding, including model outputs from Environment and Climate Change Canada and initiatives started in the Federal Hazard Identification and Mapping Program (FHIMP) and through Canada’s National Adaptation Strategy.

If an appropriate model output is available for the AOI, there is still substantial post-processing required to apply the model results to a project. In the case of using climate-driven hydrologic model output to adjust for future design flows on a river, the following steps are usually necessary:

1. Calculate the present-day design flows from streamflow observations (for example, from WSC^{Footnote 2} data). See NRCan (2025b) for further information.
2. Extract maximum streamflow (annual or other) from the ensemble of model outputs, separately by ESM and carbon scenario, at the AOI.
3. Perform a frequency analysis on ensemble climate model outputs, on each separate time series, to assess change from present to future periods. The frequency analysis on model output can be done by:
   • A comparison of shorter “pseudo-stationary” periods: for example, a frequency analysis calculated on 30-year periods of record (Figure 7). Rolling frequency analyses (for example, on a rolling X-year window) have also been proposed (Kharin & Zwiers, 2005). Or;
   • Non-stationary frequency analysis: Several approaches are possible. Figure 8 shows results for one approach using the year as a covariate for the generalized extreme value (GEV) distribution’s location and scale parameters, via the “extRemes” package (Gilleland & Katz, 2016) for the statistical software R. Other approaches include that by Vidrio-Sahagún et al. (2023, 2024) (with free and open source software), Villarini et al. (2009), and El Adlouni et al. (2007).
4. Summarize the changes from the present to the future periods for all future ensembles. Consider changes from the present day to the end of the model outputs (such as 2100 with the current generation of ESMs) but also consider changes that happen in the intermediate periods, such as decadal changes.
5. Apply the changes in the modelled frequency analyses to the design flows calculated with observations. This step is necessary as it is unlikely that the model output completely matches the observations in the overlapping present-day period. There are two options:
   • Bias correcting the model output to match the observations prior to the FFA. This would be done prior to the calculation in Step 3.
   • In a simpler approach, the calculated percent change of modelled design flows (present to future) can be applied to the observed design flows.

Option C: Purpose-built model for the area of interest

While the availability of hydroclimatic model output in Canada is increasing, there are still many situations where there is no available model output, or the available model output is inappropriate for a particular project. The course of action taken in these cases should be guided by the principle of designing an assessment proportionate to the project (see Section 2.1).

For small-scale projects, the best approach may be to combine the literature review and qualitative understanding to recommend changes that account for climate change impacts on flooding, with a mind toward over-preparing. NRCan (2025b), Chapter 8, provides some options for simplified storm surge estimation methods in lake flooding situations. However, in very large projects, over-designing is less feasible, and it is more appropriate to design and build a model specifically for the project.

Where a purpose-built model is necessary, planning is required from the initial project scoping phase. Technical expertise in hydrologic modelling, climatology, ice effects, wave effects, and other specialized fields must be available for the project, and the budget and schedule must be increased accordingly.

The detail and expertise required to develop a purpose-built hydrologic model is beyond the scope of this guideline. Following the example in Option 2 (above), the steps for developing a customized hydroclimate model to account for climate impacts on flooding in a riverine system are:

1. Configure an appropriate hydrologic model for the watershed of interest.
2. Calibrate the model for the historical period using observations (for example, weather and hydrometric data).
3. Downscale (or obtain downscaled) ESM output so that it statistically matches historical weather observations in the historical period.
4. Run the calibrated model using the downscaled climate model outputs.
5. Post-process the model output in the same manner as in Option 2 (above).

For this “maximum” analysis level, Figure 9 illustrates where and how custom hydrologic model development and climate scenario modelling can integrate into the full workflow of flood hazard mapping.

Not recommended: extrapolation

An approach for quantitative climate analysis that is not recommended is trend analysis of observations. Though a trend assessment is a worthwhile part of assessing data used for generating present-day design flows, there are multiple reasons why it is inappropriate to extrapolate these trends into the future:

• The principle of a loss of stationarity due to climate change also applies to trends. Just because a trend has occurred in the past does not mean one can assume this trend will continue.
• There are many nonlinear and threshold-based processes that relate to climate impacts and flood hazards. Extrapolation of a historical trend in flood hazard into the future discounts all these complex interactions.
• It is easy to mistake natural variability for climate change. There are many reasons that hydrometric data may exhibit a trend, such as fluvial geomorphologic factors affecting the stream channel, land use change, water management changes, and shorter-term climate variability (for example, Pacific Decadal Oscillation, El Niño Southern Oscillation).
• Many hydrometric station records are too short to provide trend analysis results with suitable confidence to conclude that the trend is caused by climate change impacts. Figure 7 illustrates how a 30-year record is often considered short enough to be called “pseudo-stationary” and analyzed as though it were a stationary record. Yet, most WSC gauges have a published peak flow record shorter than 30 years. In the 16 Jan, 2026 version of HYDAT, the median instantaneous annual peak flow record length of WSC gauges (active or discontinued) was only 20 years.

2.2.5 Recommendation

Summary of main tasks

1. Synthesize Prior Steps
2. Provide climate-informed safety factor as a value or range

Overview

Step 5 involves synthesizing Steps 1 through 4 and providing the result as a value or range of how climate change will impact peak flows in the AOI. Special considerations for this step are provided in Section 2.3.

i) Synthesize prior steps

The recommendations phase is a synthesis of the four prior steps to provide a best approach for design flow (or water level) estimation for the particular AOI. Practitioners should use all the results to make a recommendation that is most appropriate, given the understanding of climate impacts and the needs of the study.

ii) Provide climate-informed safety factor as a value or range

This final step is primarily where the key principles of assessment should be considered, particularly Principle 3 (focusing on key vulnerabilities) and Principle 4 (establishing multiple lines of evidence), to provide the result as a value or range of how climate change will impact peak flows in the AOI. For example— “it is recommended to increase the 200-year design flow by XX% to account for increased flood hazard due to climate change”.

Ideally, through Steps 3 and 4 (the conceptual model and analysis), the practitioner has explored a few different ways of considering climate impacts on the AOI (for example, the conceptual model was informed by a literature review and the assessment incorporated climate and hydrologic model outputs). Optimally, these approaches all point to similar future outcomes for the project site. If this is the case, recommendations for how design flows or levels should be adjusted into the future should be fairly straightforward.

Where various analyses are not in agreement, focusing on key vulnerabilities becomes essential. In these situations, the recommendations should focus on results and outcomes that are most critical to the specific project, and results are best presented as a range.

If one analysis points to stable flood hazard and another points to increasing flood hazard, and they are equally trustworthy methods, the conservative engineering approach would be to recommend planning for increased flood hazard. There is usually less disadvantage to planning for increasing floods that do not happen than there is for the opposite. There may be, however, a limit to being conservative in recommendations, for example, when major changes to a populated floodplain are implicated. In those cases, further analysis (either through new models or an independent assessment) may be warranted.

2.3 Special Considerations

The following special considerations should be considered, especially in Steps 4 and 5 of the assessment process.

A. Maximum climate impacts on flooding before 2100 or decreasing peak flows:

Recommendations for changes in design flows or lake levels should always be based on the highest hazard between the present day and the end of the analysis. Practitioners may assume that this highest hazard occurs in the last year of a climate projection (2100 in most cases^{Footnote 3}), but this is not always the case.

Projects with detailed analyses of model output may show that flood hazard is expected to increase for some decades and then taper off. This can be the case in snowmelt-dominant systems, where increases in melt rate increase flood hazard for some time, but are eventually overwhelmed by decreases in snowpack. The recommendation should be based on the time with the highest hazard. We have to live through 2050 to get to 2100, so if the flood hazard is higher in 2050, this should be the design recommendation.

Further, flood hazard may decline for some watercourses in Canada. For example—a watershed that will experience a complete disappearance of snow would no longer experience a spring freshet. If all other processes are staying the same, the recommendation could be for “no change” to design flows to account for climate. However, practitioners should consider Principles 2 and 4. The burden of evidence is higher for a smaller climate change adjustment, and that multiple lines of evidence should point to the “no change” conclusion.

A decrease to design flows for climate change should never be recommended; we still need to prepare for a flood today.

B. Changes in the dominant flood type:

Change in the dominant flood type or season is a particularly difficult situation to anticipate. For example—rather than a monotonic increase in spring flood hazard, the flood hazard may change from a spring freshet flood to a winter rainstorm flood. This type of change likely requires thoughtful and sophisticated physics-based hydrological modelling and a detailed conceptual understanding of the system.

Along with the difficulties in prediction, this type of change presents substantial logistical difficulties. Municipalities and other governments prepare for flooding based on knowing the approximate time of year when it can occur. If the typical flood season changes, a reorganization of the flood response is required and will be more difficult to deal with if not planned for ahead of time (see Case Study 3.4). This means that a modelled 15% increase in the 200-year flood, for example, is substantially more problematic if it is also expected that it could happen in a different season. In these cases, recommendations should go beyond numbers to also describe how the timing of flooding could change as well.

C. The worst carbon scenarios do not always result in the worst flooding

It is typically assumed that the worst carbon scenarios (i.e., most warming) will result in the most severe increases in flooding. While this is usually the case, it may not always be true. When “mild” climate scenarios result in a decrease in snowpack, but “severe” carbon scenarios result in a complete loss of snowpack over a watershed, hydrologic modelling may indicate greater flood hazard for the moderate scenario. This possibility should be considered during the conceptual model phase and evaluated (via assessment of multiple carbon scenarios) during the analysis whenever possible.

2.4 Conclusion

The assessment process described in this section is intended to guide, rather than prescribe, how practitioners can integrate climate change considerations into local flood hazard mapping projects. Practitioners are encouraged to make adjustments according to the purpose of their project, the availability of information, and their skillsets and knowledge.

The case studies presented in Section 3.0 illustrate recent approaches to climate assessments and recent developments in local practices across the country.

3.0 Case Studies

The climate assessment framework in the prior section described why a one-size-fits all approach to climate assessment of flooding is not feasible. The assessment must be tailored to the project. The case studies in this section reflect engineering developments in incorporation of climate impacts into flood hazard assessments across Canada. Practitioners from across Canada contributed their approaches, solutions, and experiences to this guide.^{Footnote 4} Though they were carried out before release of the climate assessment framework, the contributed case studies cover a variety of communities and their unique flood hazards, a range of mapping products and methodologies, and a diversity of outcomes.

3.1 Nahanni Butte: Flood Hazard Mapping

Community Profile

Nahanni Butte, traditionally Nahɂą Dehé Dene Band, is a small, remote Indigenous community in the southwest corner of the Northwest Territories (NT). The community is located in the Dehcho region, downstream of the Nahanni National Park Reserve, a UNESCO World Heritage Site.

Nahanni Butte is situated on a relatively flat river terrace on the south bank of the South Nahanni River, near its confluence with the Liard River. The total drainage area contributing to the community is approximately 264,900 km². Within this, the South Nahanni River watershed (36,000 km²), is a key tributary. The watershed is a mountainous and forested basin in the Mackenzie Mountains, with the community itself on the eastern edge of the mountains.

Local flood hazards

The community is susceptible to open-water riverine flooding, which is a key distinction from most other NT communities that face ice jam flooding risks. The primary flood drivers are large rainfall events within the mountainous South Nahanni River basin, particularly when coinciding with a late spring snowmelt from a substantial snowpack.

High water levels on the larger Liard River can exacerbate these events by creating a backwater effect that impedes the flow of the South Nahanni River. The most significant recorded flood event in recent history occurred in June 2012.

Purpose

The objective of this study was to develop flood hazard maps for Nahanni Butte to support future land use planning, emergency preparedness, and climate resilience. The study focused on modelling inundation extents for multiple open-water flood scenarios. These scenarios included:

• The 2012 observed flood;
• Flood with 1% AEP (1:100-year flood);
• Flood with 0.5% AEP (1:200-year flood); and
• Two future climate scenarios.

Methodology

A complete hydrotechnical assessment was conducted, beginning with a background review and a field survey, followed by hydrologic and hydraulic modelling. The field survey was undertaken to obtain current bathymetric and topographic information, and to identify high water marks based on community input and recollection from the 2012 flood. The bathymetric survey used a single ping depth sounder mounted on a boat to cover 15 km of the Liard River and 10 km of the South Nahanni River. The bathymetric survey was able to capture the braided channels in the area, and the topographic survey used a drone to capture banks and exposed sand bars that were not possible to survey using the boat.

Subsequently, a flood chronology was developed for the June 2012 flood to better understand the dynamics and factors leading to the event. Hydrologic modelling was then conducted using Raven (Craig et al., 2020) to identify flow hydrographs for inclusion in the 2-D HEC-RAS (Hydrologic Engineering Center’s River Analysis System) model. The resultant flood inundation extents, velocities, and depths were used to develop flood inundation and hazard maps.

Hydrologic modelling

Hydrologic modelling used a published Raven model, originally developed and calibrated for the Liard River basin. For this component of the project, Dillon partnered with Heron Hydrologic, which completed the Raven modelling. Following recalibration and inclusion of updated hydrometric and meteorological data, the model was used to simulate continuous daily streamflow over historical and future periods. The total drainage area in the model was approximately 264,900 km², encompassing both the South Nahanni River and Liard River watersheds. Regional Flood Frequency Analysis (RFFA) was also used to inform flood magnitudes using data from gauges within a 150 km radius because there was no hydrometric gauge with sufficient data in Nahanni Butte. Peak flows for five scenarios were extracted from this modelling:

• 2012 historical flood: The peak flow for the 2012 flood was obtained from the simulated Raven model output, as no hydrometric station existed at the community during that event.
• 1% AEP (1:100-year) and 0.5% (1:200-year) floods: Peak flows were estimated using flood frequency analysis (FFA) on the historic simulated flow time series from the Raven model. These were compared with peak flows estimated using the RFFA. Since the RFFA results yielded higher peak flows than those derived from the Raven FFA, the RFFA results were selected for use in the hydraulic model.
• Two climate change scenarios: Peak flows were estimated using flood frequency analysis on forecasted flow time series simulated from the Raven model, driven by the forcing data from the respective climate change scenarios’ data. Details on the selection process of the climate change scenarios are described in the following section.

For each of the scenarios, the shape of the daily-scale flow hydrograph was based on the 2012 historic flood hydrograph and scaled according to their respective peak flows relative to the 2012 flood. The hydrographs were used as upstream boundary conditions on both the South Nahanni and Liard Rivers for the hydraulic model.

Climate change datasets were obtained from Environment and Climate Change Canada (ECCC) and were based on an ensemble of 26 statistically downscaled CMIP6 (Coupled Model Intercomparison Project Phase 6) models. The datasets for the model ensemble were downscaled using the multivariate statistical downscaling method CanDCS-M6. The downscaled datasets covered the period 1950–2100.

Two climate change scenarios, SSP2-4.5 (representing a moderate-emissions scenario) and SSP5-8.5 (representing a high-emissions scenario), were chosen for modelling and mapping upon review of the initial modelling results. For the model ensemble, percentiles for the temperature and precipitation projections were used to summarize the range of results from the multi-model ensemble as follows:

• Cool and wet conditions: 25th percentile temperature and 75th percentile precipitation (represented here as T25P75);
• Median conditions: 50th percentile temperature and 50th percentile precipitation (represented here as T50P50); and
• Extreme cool and wet conditions: 5th percentile temperature and 95th percentile precipitation (represented here as T5P95).

The model runs included in the hydrologic assessment of climate change are summarized in Table 2.

Table 2: Climate Change Scenarios

Scenario Name	Emission Pathway scenario	Model	Modelling Period	Temperature Percentile	Precipitation Percentile
Moderate Emission—Median Conditions	SSP2-4.5	Ensemble	2022–2100	50	50
Moderate Emission—Cool/Wet Conditions	SSP2-4.5	Ensemble	2022–2100	25	75
Moderate Emission — Extreme Cool/Wet Conditions	SSP2-4.5	Ensemble	2022–2100	5	95
High Emission — Median Conditions	SSP5-8.5	Ensemble	2022–2100	50	50
High Emission — Cool/Wet Conditions	SSP5-8.5	Ensemble	2022–2100	25	75
High Emission — Extreme Cool/Wet Conditions	SSP5-8.5	Ensemble	2022–2100	5	95

The output of the hydrologic assessment for each scenario was a continuous time series of flows in the South Nahanni and Liard Rivers from 2022 to 2100. Based on the results, the median conditions (T50P50) appeared to underestimate flows by approximately an order of magnitude. The extreme cool/wet condition (T5P95) was overly conservative, and produced peak flows that were approximately twice as large as historic estimates. The climate change sensitivity analysis demonstrated that the cool/wet climate change scenarios (T25/P75) were most reflective of increased but reasonable maximum daily flows for both rivers. This suggests a potential for increased snowpack and higher annual peak flows in the future.

Having chosen the cool/wet climate change scenario, the continuous results must be translated into flood hazard maps by understanding the statistical relationship of projected annual peak flows. To account for the inherent non-stationarity in the climate change results, the results were examined in three segments:

• Near future (2022–2050);
• Mid-century (2050–2075); and
• End of century (2075–2100).

Two combinations of the following conditions were selected from the sensitivity analysis to represent the climate change scenarios:

• Annual Exceedance Probability: for example, 1% (1:100-year) or 0.5% (1:200-year).
• Emissions scenario: for example, SSP2-4.5 or SSP5-8.5.
• Time period: for example, 2022–2050, 2051–2075, or 2076–2100.

The respective peak flow magnitudes for each of the cool/wet scenarios are presented in Figure 12 and Figure 13 for the South Nahanni River and Liard River, respectively. Note that the vertical axis has been modified to demonstrate the variability between scenarios.

The figures present the historical peak flows from the RFFA and the future peak flows from FFA results for the climate change sensitivity analysis. In the Liard River (Figure 13), the peak flow appears to increase as climate change progresses, while the South Nahanni River (Figure 12) flows appear to stabilize or decrease in response to climate change.

The two climate change scenarios were selected in consultation with the Government of the Northwest Territories, Environment and Climate Change Canada, and Natural Resources Canada, and based on the following considerations:

• Annual Exceedance Probability: The hydraulic assessment of historical conditions indicated that flooding will happen during both 1% and 0.5% AEP historical flood events, and the difference between the extent of the two events was minor. As such, the 0.5% AEP was selected for the climate change scenario. This aligns the outcomes of this study with external initiatives (for example, the DFAA, which uses 200-year).
• Emission scenarios: SSP2-5.5 and SSP5-8.5 were selected for climate change scenarios to represent one moderate and one extreme emission scenario.
• Time period: The near future from 2022 to 2050 was selected to create flood hazard mapping under climate change scenarios as it is more useful in the short term for planning considerations.

The two selected scenarios to be used in flood mapping under climate change are summarized in Table 3. For clarity, they will be referred to by the appropriate emission pathway scenario, which is what differentiates the two, in the following sections.

Table 3: Selected Climate Change Scenarios for Flood Hazard Mapping

Flood Scenario ID	AEP (%)	Climate Change Scenario Name	Emission Pathway Scenario	Model	Modelling Period	Temperature/Precipitation Percentile
Scenario 1: Climate Change to 2050, SSP2-4.5	0.5	Moderate Emission—Cool/Wet Conditions	SSP2-4.5	Ensemble	2022–2050	25th/75th
Scenario 2: Climate Change to 2050, SSP5-8.5	0.5	High Emission—Cool/Wet Conditions	SSP5-8.5	Ensemble	2022–2050	25th/75th

Results indicate that climate change will differentially affect the South Nahanni and Liard Rivers, with modelled flows on the South Nahanni River anticipated to decrease and flows on the Liard River anticipated to increase. The near-future statistical results for the 0.5% AEP (1:200-year) were chosen for SSP2-4.5 and SSP5-8.5.

Hydraulic modelling

A 2D HEC-RAS model was developed using lidar-derived terrain and bathymetric survey data of the Liard and South Nahanni Rivers. The 2D model used daily hydrographs from Raven as upstream boundary conditions on the Liard and South Nahanni Rivers. The downstream boundary condition was used as a calibration mechanism to align the flood elevations with surveyed high-water marks from the 2012 flood. The calibrated hydraulic model generated depth, velocity, and water surface elevation rasters for each of the modelled scenarios.

Outcomes

The project produced a set of detailed flood inundation and hazard maps for future community land use and emergency planning. The 2012 flood of record was produced as a flood inundation map, and four flood hazard maps were produced:

• 1% AEP (1:100-year) historical flood,
• 0.5% AEP (1:200-year) historical flood,
• 0.5% AEP (1:200-year) Climate change scenario 1 (T25/P75)—SSP2-4.5 for 2022–2050, and
• 0.5% AEP (1:200-year) Climate change scenario 2 (T25/P75)—SSP5-8.5 for 2022–2050.

The flood hazard maps define floodway and flood fringe areas, with the floodway primarily determined by water depths greater than 1 metre and/or velocity greater than 1 m/s. Due to the community's undulating topography, flood depths vary significantly across the area, and few places are expected to remain outside the flood fringe.

Conclusion

The Nahanni Butte study used updated Raven hydrologic modelling and 2D HEC-RAS hydraulic modelling, supported by ground surveys and bathymetric data, to simulate flood scenarios including historical (2012 flood), floods of 1% AEP and 0.5% AEP (1:100-year and 1:200 year), and future climate change scenarios.

Climate change analysis was based on downscaled CMIP6 datasets, and two emissions scenarios (SSP2-4.5 and SSP5-8.5) were evaluated using a cool/wet percentile combination (25/75). The climate change scenarios representing the 0.5% AEP (1:200-year) for the near-future horizon (2022–2050) were selected as a conservative planning scenario.

Five flood maps were developed, of which two quantified projected changes as a result of climate change scenarios. The final flood maps were shared with the community at the completion of the project.

Back to Case Studies

---

3.2 Comox Valley Regional District: Coastal Flood Hazard Mapping

Community Profile

Comox Valley Regional District (CVRD) encompasses a 1,725 km² area on the east coast of Vancouver Island, British Columbia (BC). It includes the incorporated communities of the Town of Comox, the City of Courtenay, the Village of Cumberland, and many smaller unincorporated communities. It is the ancestral home of the K’omoks First Nation, which has reserve lands on Comox Harbour adjacent to the CVRD.

The CVRD is situated on approximately 150 km of coastline on the Strait of Georgia in the Salish Sea, and includes the Oyster River and Courtenay River systems, as well as many smaller watercourses. Most of the approximately 67,000 residents live near the Strait or these river systems.

Local flood hazards

The CVRD has a history of flooding in its coastal and river areas. The severity of flooding is projected to increase under future climate change conditions. This project was the first step in its multi-year process to understand and prepare for flood hazards.

To create a variety of mapping products that illustrated current and projected future flood exposure, Kerr Wood Leidal Associates (KWLA) conducted detailed analysis and modelling of coastal flood and river hazards. KWLA considered coastal and river area flood conditions in parallel to better understand how water levels in coastal areas might influence river flood dynamics.

Purpose

KWLA produced flood mapping products with a set of technical memoranda describing the approach to modelling and mapping, including:

• Region-wide base maps that provided the platform for modelling and mapping this project.
• Regulatory flood maps that showed flood inundation extents for an extreme storm event under a specific set of future climate change conditions, as recommended by established guidelines (0.5% annual exceedance probability [AEP] storm event with 1 metre of sea level rise) (see Figure 12).
• Digital mapping data, including maximum water levels, maximum water depths, maximum flow velocity, flood hazard rating, maximum wave heights in the coastal floodplain, sea level rise planning areas, and setbacks for a range of AEP storm events with climate change.

CVRD is using the mapping products to inform flood hazard area land use management, risk assessment, emergency planning, and engagement with the community. For example, the regulatory flood maps have been used to create a public communication and education tool—a web-based mapping application on CVRD’s website.

Methodology

KWLA developed the flood mapping products according to established guidelines, including:

• the BC Provincial Flood Hazard Area Land Use Management Guidelines (PDF, 1.90 MB) (2018) (“BC provincial guidelines”),
• the Federal Flood Mapping Framework (PDF, 418 KB) (Natural Resources Canada & Public Safety Canada, 2018) including relevant guidelines,
• the Engineers and Geoscientists BC Flood Mapping Guidelines (PDF, 5.90 MB) (2017), and
• the Fraser Basin Council Floodplain Mapping Guidelines and Specifications (2004).

For the analysis, KWLA considered flood levels and depths for a wide range of storm events, ranging from 0.2% AEP (500-year return period storm event) to 10% AEP (10-year return period storm event), with allowances for climate change included in both sea levels and river flows.

Coastal analysis process

KWLA estimated coastal flood levels and setbacks using the probabilistic method from the BC provincial guidelines (Figure 15). They determined extreme static ocean water levels, including tides and storm surge, through analysis of historical water level data collected in the Strait of Georgia and CVRD area.

KWLA incorporated climate change projections in coastal flooding projections by considering sea level rise. They evaluated a range of sea level rise conditions from current conditions to 2.0 m of relative sea level rise and used sea level rise of 1.0 m for regulatory flood mapping, as per the BC provincial guidelines. Adjustments to windstorm intensity and frequency were considered but not implemented, as the climate literature in this area is still evolving and currently provides no consensus. (Most studies show little change in the vicinity of Vancouver Island).

KWLA used a high-resolution, two-dimensional deep water spectral wave model to estimate wave conditions in the Salish Sea. Windstorm events from each cardinal direction were modelled using spatially and temporally varying wind fields generated from data collected during historical storm events. To drive the wave model, wind speeds for each storm event were scaled to extreme values.

For nearshore wave modelling, KWLA performed a one-dimensional “transect-based” approach using its SHORLAX custom software for high-resolution regional wave effect analysis. The entire CVRD shoreline was divided into 233 transects, selected to be representative of the surrounding coastal “zone” in terms of both topography/bathymetry and wave exposure. At each transect, the SHORLAX model was forced with water levels and wave conditions from the two-dimensional deep water spectral wave model, and for each storm scenario, calculated the wave run-up. Secondary effects of sea level rise were captured in these simulations, as with deeper water, waves can travel further into shore before breaking, resulting in a larger wave run-up.

At each transect, the influence of tides, storm surge, wave effects, local land uplift or subsidence, sea level rise, and freeboard were combined to estimate the flood level and setbacks. The flood levels were converted into two-dimensional mapping based on the assumption that the flood level is constant within a given coastal zone.

River analysis process

KWLA designed the flood hydrology for each river system based on its analysis of the historic flood frequency data, available guidance, and climate modelling projections. KWLA applied a factor of +15% to the peak design flood discharge, to account for projected climate change impacts on peak floods to mid-century (nominally the year 2050) and end of the century (nominally the year 2100). A factor of +30% was applied to the peak design flood discharge to account for projected climate change impacts on peak floods for the next century (nominally 2200) (Figure 16).

KWLA completed hydrologic and hydraulic modelling for the Oyster River and Courtenay River systems. They used one- and two-dimensional techniques to model and estimate flood elevations, depths, and flow velocities along the river channel and over the floodplain under a range of river storm intensities and climate change impacts. To ensure that the full range flood scenarios were captured, KWLA modelled current and proposed flood protection works and potential dike failure events.

Conclusion

An example of a regulatory flood map resulting from this process is shown below (Figure 17). Key findings from this flood mapping project include the following:

Back to Case Studies

---

3.3 Cowichan Lake: Shoreline and Lake Level Flood Mapping

Community Profile

Cowichan Lake is the second-largest lake on Vancouver Island, British Columbia (BC), with a surface area of approximately 62 km². It is the headwaters of the Cowichan River, a nationally recognized Heritage River, which flows 47 km through the Cowichan Valley to the ocean at Cowichan Bay (Figure 19).

Local flood hazards

Cowichan Lake drains into the Cowichan River through a weir and flow control gate system. Constructed in the 1950s, the system retains water in the lake during the wet spring months and captures spring snowmelt to maintain summer baseflows in the river downstream. This supports habitat for important salmon species, maintains the ecological health of the river system, and supplies water withdrawals for the Crofton Pulp Mill.

The system manages lake levels to maintain river flow rates above established regulatory and environmental targets. A steady decline in late spring and summer (June—September) lake inflows has been observed since the 1960s and climate change is expected to cause further declines into the future, making it increasingly difficult to meet the target flow rates during dry summer months.

Upgrading the weir to facilitate climate adaptation on the Cowichan River has been discussed since the 1990s. The current proposal is to store more water through the wet spring season by raising the weir and initiating flow regulation earlier in the season to help secure water for the river during the dry summer months. The weir and gates would be raised by 0.7 metres, but the gates would only be operated during the spring and summer months, leaving flows and lake levels to be unregulated during the fall and winter. Consequently, the overall water level range in the lake would largely remain the same, while the timing of high and low water levels would be shifted. This approach is discussed in detail below.

Purpose

The goal of this project was to provide a detailed assessment of the potential impacts to the Cowichan Lake shoreline as a result of the weir upgrade and climate change. The assessment will support a future water licence application for the increased storage at Cowichan Lake. The project objectives were to:

• Assess, map, and document the current shoreline conditions;
• Forecast changes to the lake level and shoreline conditions; and
• Identify the potential impacts to lakeshore properties at a lot-by-lot scale.

Methodology

The natural boundary and lake level frequency were used by Kerr Wood Leidel Associates (KWLA) as the primary metrics for assessing shoreline impacts. The natural boundary was selected because of its legal significance in BC, and because it embodies several shoreline characteristics, including erosion, in a single variable. Lake level frequency was selected for its clear impacts on land use and flooding.

KWLA conducted an extensive field program and used subsequent desktop analysis to survey the natural boundary and the physical characteristics of the shoreline (vegetation, sediment, slope) around the entire lake.

Inflow

Cowichan Lake levels are primarily a function of inflow, river discharge, and evaporation. Continuous monitoring of discharge and lake levels since 1953 allowed for back-calculation of historic inflows. To project how inflow might change into the future, a hydrological model (UBC Watershed Model) was used. The inputs to the model included topographic, soil, and land cover parameters, as well as a continuous time series of temperature and precipitation data. KWLA calibrated and verified the model by comparing the model results over the historic period to inflow derived from observational data.

To assess inflows to the lake under climate change conditions, KWLA ran the hydrological model using the daily temperature and precipitation time series from a set of ten downscaled global circulation models (ESMs) from the Coupled Model Intercomparison Project 5 (CMIP5), using the business-as-usual Representative Concentration Pathway (RCP) 8.5 emissions scenario for 1981 to 2099 (Pacific Climate, accessed 2019). While the more recently completed CMIP6 climate projections were available, KWLA selected the CMIP5 data to be consistent with climate data sets used in other climate projection studies in the CVRD.

Lake levels

KWLA used the Cowichan Lake Operational Model to simulate past and future lake levels under both the current and upgraded weir scenarios. As a mass balance model, it uses lake inflow and river flow (either controlled by the gates or uncontrolled) to simulate lake levels throughout the year. The model accounts for seasonal operation of the gates, including spring and summer flow targets and minimum flow thresholds based on environmental targets and water licence requirements.

To understand the variability in future lake level and to select a single representative lake level time series for impacts assessment, KWLA used the modelled lake inflow for each of the ten downscaled ESMs as input to the operational model to generate ten daily lake level series for Cowichan Lake. They were generated for both the past climate and the future projected climate with the existing weir and proposed 0.7 m raised weir conditions.

Waves

Wind waves are thought to be the primary source of erosive energy on the Cowichan Lake shoreline. As a result, quantification of wave conditions was critical in determining shoreline impacts. A high-resolution computational wave model was used to estimate shoreline wave conditions under lake levels with the current and proposed weir.

Natural boundary

The natural boundary is intuitively understood as the line dividing the land from the lakebed. Using the extensive dataset of shoreline physical characteristics (vegetation, sediment, slope, waves, water levels), KWLA developed a statistically based equation that predicts the natural boundary elevation for any given shore segment based on its physical characteristics. The equation is based on the average wave power incident on the shoreline above the natural boundary and was used to estimate the change in the natural boundary elevation with the proposed weir around the entire lake. The estimated vertical change in the natural boundary was used to calculate the horizontal change.

Shoreline properties

Using the estimated location of the future natural boundary and the change in frequency of lake levels for the proposed raised weir scenario, KWLA assessed potential impacts to the 846 shoreline properties at Cowichan Lake.

Outcomes

KWLA considered inflow, lake levels, waves, the natural boundary, and shoreline properties in its assessment.

Inflow

The suite of modelled inflow results indicates that:

• The annual average inflow to the lake will increase 2.1% by the 2050s and 8.3% by the 2080s. Most of the increase will be seen in the fall and winter because of increased precipitation and higher temperatures, resulting in less precipitation falling as snow at high elevations.
• Inflow will decrease during the spring and summer; summer inflow is expected to decrease by 30% by the 2050s and 40% by the 2080s (see Figure 20). A reduction in summer inflow is expected due to reduced precipitation, reduced snowpack (and therefore reduced spring snowmelt runoff), and increased evaporation due to the higher temperatures expected in the future.

As the increase in inflow is expected to occur in the fall, when the gates at the weir are fully open and there is no control of lake level and river flow, they may not result in increased water availability during the critical summer months. Instead, water availability will be most impacted by the reduction in inflows expected during the spring and summer.

Lake levels

Figure 21 compares simulated water levels from the representative time series for the existing weir and the proposed raised weir over a four-year period.

Waves

The difference in average wave power at the shoreline between the two weir scenarios is negligible in the deep areas of the lake but becomes larger close to shore. For most of the shoreline, the difference is less than 10%; however, differences of up to +/-30% were observed in some locations. The largest changes were predicted to occur at sheltered shorelines, or shorelines with a shallow foreshore, where the existing average wave power is very low.

Natural boundary

Figure 22 shows Gordon Bay Provincial Park, the estimated current and projected future natural boundary, as well as the contour lines associated with the current and proposed weir crest.

Shoreline properties

A selection of assessed impacts on shoreline properties is summarized in Table 4.

Table 4: Summary of Potential Shoreline Impacts.

Impact	Description	Findings
Flooding	Potential change in flood risk to primary residences.	Raising the weir would result in a relatively small increase in flood lake levels. Modelling indicates that the mean annual flood lake level could increase by about 0.1 m (4 in) while flood levels up to the 50-year return period could increase by no more than 0.05 m (2 in). Higher peak lake levels (greater than the 50-year flood) are not expected to have any change.
Inundation	Will the raised weir result in additional water being stored within the property boundary?	Approximately half of the shoreline properties (414) have some portion of the lot area with ground elevations below that of the proposed weir crest, such that water stored by the proposed raised weir will cover a portion of these lots.
Estimated change in natural boundary	Estimated horizontal change in natural boundary along lake frontage.	The projections of changes in the natural boundary indicate 708 properties (84%) could have a change of less than 3 m due to the proposed raised weir. There are 13 properties (2%) where the horizontal shift in the natural boundary is estimated to be between 10 m and 22 m.

Conclusion

The largest change in water level regime, and consequently the largest change in natural boundary, would occur in the few years following the completion of the weir and control gate system upgrades. By the 2080s, KWLA projected that decreases in spring and summer inflows, combined with the storage of the upgraded weir, would result in a water level regime similar to current conditions. Consequently, it is expected that the natural boundary would migrate back toward its current location.

Should the weir not be raised, all shoreline properties will have some impact in years when storage is depleted, and lake levels drop below the lowest observed lake levels. Modelling indicates that by the 2080s, lake levels will drop below the lowest observed lake levels about ten times more often with the existing weir than the proposed raised weir.

Back to Case Studies

---

3.4 Kootenay Boundary: Flood Mapping

Community Profile

The Kootenay Boundary Regional District serves more than 33,000 residents in the southeastern BC interior, stretching along the Canada-US border. As the mountainous region confines communities to the valley bottoms, they have been subjected to floods and fluvial geomorphic risks, including a catastrophic flood in 2018 in the western portion of the Boundary.

Purpose

The purpose of this study was to develop flood maps for four regions:

• Kettle River from Rock Creek to Midway (24 km)
• West Kettle River from Carmi to Beaverdell (16 km)
• Kettle River from Cascade Falls to the international boundary near Laurier
• Christina Lake and Christina Creek.

The Kettle River from Rock Creek to Midway study reach is shown in Figure 24. In 2018, the City of Grand Forks experienced considerable freshet flooding; the highest streamflows on record were recorded by Water Survey of Canada (WSC) hydrometric stations.

Figure 25 presents the hydrologic regime for the Kettle River at Ferry (08NN013). To support flood mapping, historic floods were reviewed, and frequency analysis was completed on key WSC gauges to support development of hydraulic model inflows. Future conditions were developed using hydrologic model outputs developed by Associated Engineering (AE) (2021b) and by assessing the skew of the frequency analysis curves for mixed regime watersheds from NHC (2021). Results were applied to the study.

Note: Red line represents the median flow (1929–2018), grey bands represent the 25th—75th percentile, 10th—90th percentile and maximum/minimum, respectively according to receding shade.

For the purpose of describing the climate change approach undertaken for this study, only the Rock Creek to Midway reach is presented. However, the same approach was applied to all study reaches.

Local flood hazards

Flooding in the study area can be riverine and lake induced (including waves), with climate change projected to increase river peak flows and lake levels. Projected increases to river peak flows and lake level rise have potential knock-on effects on the fluvial geomorphology, in some cases increasing the potential of debris flows and avulsions, and potentially accelerating bank erosion, lateral channel migration, sedimentation, and riverbed aggradation.

Secondary impacts were considered. Forest harvesting and disturbance (for example, due to wildfires, insect infestation) has the potential to impact the hydrology of Kettle River watershed over the next century. Chernos et al. (2020) investigated climate change impacts for the Kettle River watershed. They found an increase of 3–6% for the 100-year peak flows for the Kettle River.

Methodology

A climate change assessment was mandatory for this project, based on professional practice guidelines and client requirements. Guidelines that were applied were:

• Flood Mapping in BC—Professional Practice Guidelines (PDF, 5.9 MB) (APEGBC, 2017)
• Legislated Flood Assessments in a Changing Climate in BC—Professional Practice Guidelines (PDF, 3.20 MB) (Engineers and Geoscientists BC, 2018)
• Federal Flood Mapping Guidelines Series (Natural Resources Canada, 2025a)
• Flood Hazard Area Land Use Management Guidelines (PDF, 1.92 MB) (BC Ministry of Forests, Lands, Natural Resources Operations and Rural Development, 2018).

The range of future time horizons and carbon concentrations considered was RCP 4.5 and 8.5, 2050 and 2100.

NHC relied on previous hydrologic modelling completed for the Kettle River watershed, which were then compared to Pacific Climate Impacts Consortium (PCIC) VIC-GL model output. The data sources and climate projections are described below.

A hydrologic modelling study of the Kettle River was carried out by AE (2021a), which extended a model developed in the Raven hydrologic modelling platform (Craig et al., 2020) to the entire Kettle River watershed (Chernos et al., 2020). The extended model used the BCCAQv2—PNWNA met bias-corrected Downscaled Global Climate Model Dataset to simulate projected flows for the Kettle River near Ferry (WSC 08NN013) from 1949 to 2099 using the six different climate models and two different carbon pathways (RCP 4.5 and 8.5).

In addition, modelled streamflow data under 12 climate scenarios was available from PCIC (2020) for the Kettle River near Ferry (WSC 08NN013). NHC compared the PCIC VIC model data to the AE Raven model results and found they aligned. However, as the spatial discretization of the AE Raven model includes the West Kettle River as a subbasin, which is of direct relevance to this study, only the AE Raven results are presented here.

Approach used for assessing climate impacts

The approach to assessing climate impacts was completed as follows:

1. NHC first assessed hydrologic model output from AE (2021a) and examined the relative changes from the baseline to changes in magnitude and timing of future annual peak flows. The AE (2021a) model results did not show an increase in future annual peak flows for the Kettle River near Ferry (WSC 08NN013). Model results did however show the possibility of a future regime shift with the occurrence of winter peak flows on the Kettle River.
2. The second step was to decide how to develop a climate change adjustment to account for a regime shift in future peak flows for the Kettle River. NHC examined the skew of the distribution for mixed regime watersheds that were similar in size to the Kettle River near Ferry (WSC 08NN013). The approach and rationale are as follows:

Watersheds that can experience peak flows in both winter and during the spring freshet tend to have higher skew values than more arid interior BC watersheds. NHC (2021) found that the most extreme floods tended to be winter rain-on-snow floods, a result that agrees with the understanding of different types of flooding in hydrology research (McCabe et al., 2007). In practice, adjusting skew rather than adjusting by a constant percentage will result in smaller changes to lower return periods (for example, 5-, 10-, 20-year) and larger changes to the higher return periods. This result is aligned with the modelling results, where spring peak flows may decrease, while very extreme winter peak flows may become possible.

To account for future conditions, NHC reinvestigated the skew results from NHC (2021) to determine an appropriate skew adjustment. They selected watersheds that had a:

• Similar size as of the Kettle River near Ferry (WSC 08NN013), and
• Similar ratio of winter to spring peak flows (15 to 25%) as the future modelled projections (from 2050–2100) for the Kettle River near Ferry.

From this subset of results, they found a (rounded) median skew value of 0.5 for the 1-day peak flow distributions. They then replaced the log Pearson Type III distribution with a skew value forced to 0.5 for the Kettle River near Ferry (08NN013) to account for future conditions.

Outcomes

The results from the assessment of the hydrologic model output from AE (2021a) are as follows:

• The skew of the distribution for mixed regime watersheds from NHC (2021) that were similar in size to the Kettle River near Ferry was 0.5 (median skew value of all reference gauges reviewed, for 1-day peak flow). NHC used this skew to develop a climate change adjustment to account for a regime shift in future peak flows. They replaced the log Pearson Type III distribution with a skew value forced to 0.5 for the Kettle River near Ferry (08NN013) to account for future conditions. The recommended design flow adjustments for 08NN013 using a skew value forced to 0.5 are shown in Table 5. Skew adjustments that resulted in future design flow changes less than 10% were capped at a 10% minimum, as recommended by EGBC (2018).

Table 5: Climate Change Adjustments to Design Flow Estimates: Kettle River near Ferry.

Return period	08NN013
	Increase (%)	Design flow
2-year	10	388
5-year	10	483
10-year	10	541
20-year	10	593
50-year	10	656
100-year	11.7	713
200-year	15.4	783
500-year	20.6	880

Note: Peak flow increases of less than 10% from the skew adjustment were left at 10%, the minimum increase recommended by EGBC (2018)

Back to Case Studies

---

3.5 Nisg̱a’a Nation: River Basin Flood Hazard Mapping

Community Profile

The Nass River flows 380 km from the Coast Mountains southwest to Nass Bay and has three major tributaries. This project focused on the areas of three Nisg̱a’a Villages — Gitwinksihlkw, Lax̱g̱alts’ap, and Ging̱olx — as well as the coastal road between Lax̱g̱alts’ap and Ging̱olx.

The Nass River basin is the location of the first modern-day treaty settlement in BC, between the provincial government and the Nisga'a Nation.

Local flood hazards

Flooding in the study area (Figure 29) can be due to riverine and coastal (wave, tsunami) sources, with climate change projected to increase river peak flows and sea levels. These projected increases to river peak flows and sea level rise have potential knock-on effects on the fluvial geomorphology, in some cases increasing the potential of debris flows and avulsions, and potentially accelerating bank erosion, lateral channel migration, sedimentation, and riverbed aggradation.

Purpose

The purpose of this study was to outline flood hazards in the Nisg̱a’a Nation, with consideration for climate change, focused on the three Nisg̱a’a Villages of Gitwinksihlkw, Lax̱g̱alts’ap, and Ging̱olx, and the coastal road between Lax̱g̱alts’ap and Ging̱olx.

The project deliverables consisted of flood hazard (depth and velocity), geomorphic hazard, and Flood Construction Level mapping developed through the following project components: bathymetric and topographic surveys, hydrometric monitoring, climate change impacts assessment, hydrological modelling, geomorphological assessment, hydraulic modelling, coastal flood hazard assessment, and flood mapping.

To understand if peak flows were sensitive to future deglaciation, with the volume of glacier ice within Western Canada expected to shrink by 70% by end of century (Clarke et al., 2015), the hydrological model was run with glacierized areas decreased and replaced with open barren land.

It was found that the impact to peak flows for the Nass River (6% glacierized) and smaller tributaries (up to 27% glacierized) was less than 5%. Given the uncertainty in deglaciation, current glacier coverage was used for the determination of design flows. It was highlighted that glacier loss may impact peak flows through potential mechanisms that were not simulated, such as transitional phases of loss that lead to sedimentation and reduction in-channel conveyance, or the formation of meltwater lakes with the potential for outburst floods.

Methodology

The guidelines that were followed that are relevant to a climate change assessment were:

• Flood Mapping in BC—Professional Practice Guidelines (PDF, 5.90 MB) (APEGBC, 2017)
• Legislated Flood Assessments in a Changing Climate in BC—Professional Practice Guidelines (PDF, 3.20 MB) (Engineers and Geoscientists BC, 2018)
• Federal Flood Mapping Guidelines Series (Natural Resources Canada, 2025a)
• Flood Hazard Area Land Use Management Guidelines (PDF, 1.92 MB) (BC Ministry of Forests, Lands, Natural Resources Operations and Rural Development, 2018).

The client also preferred that a comprehensive assessment of climate change impacts be included.

NHC investigated four greenhouse gas global emissions scenarios associated with four different shared socioeconomic pathways (SSPs) as part of the climate change analysis: SSP1-2.6, SSP2-4.5, SSP3-7.0, and SSP5-8.5. For each of these climate scenarios, climate projections from three global climate models were used for mid-century and end-of-century time horizons, resulting in 24 different climate change simulations. The models were selected on the basis of capturing a wide range of projections in temperature and precipitation (Cannon, 2015).

At the time of the study, downscaled CMIP6 projections were not yet available from provincial sources. CMIP6 projections were statistically downscaled using quantile mapping for the project from the following ESMs:

• CNRM-ESM2-1, developed by Centre National de Recherches Météorologiques (CNRM) and Centre Européen de Recherche et de Formation Avancée en Calcul Scientifique (CERFACS)
• EC-EARTH3, developed by EC-Earth Consortium (European community)
• ACCESS-ESM1-5, developed by Commonwealth Scientific and Industrial Research Organisation (Australia).

A hydrological model of the Nass River basin and tributaries to Nass Bay was developed in the Raven hydrological modelling framework to understand the impact of climate change on the hydrological processes within the basin, including the peak flows (Figure 30 and Figure 31). The model was first developed and calibrated to historical data prior to using future climate projections. During calibration, specific attention was given to simulating the historical flow regime and dominant flooding processes.

Outcomes

The model proved to have adequate performance in calibration and validation to assess the relative change between present day and future climate scenarios. Importantly, the model was able to adequately simulate both freshet and winter peaks, which provided confidence in the simulation of future scenarios where their distribution is expected to change.

The model was also able to simulate other internal processes such as snow accumulation and melt, which led to greater confidence in the ability of the model to simulate scenarios outside of the current range of precipitation and temperature.

Conclusion

Back to Case Studies

---

3.6 Rat Creek: Riverine Flood Mapping

Community Profile

Rat Creek is a tributary to the Pembina River near Drayton Valley, southwest of Edmonton, Alberta (AB). The Water Survey of Canada (WSC) operates active hydrometric gauges in the area. The Rat Creek gauge is located 32 km west-southwest of Drayton Valley (Figure 34).

Local flood hazards

In June 2023, a WSC gauge at Rat Creek near Cynthia gauge (07BA002) was destroyed in a flood, due to a heavier than average rain event when large flows in Rat Creek destroyed the gauge infrastructure. The road and bridge across Rat Creek were also destroyed. It is unknown how high the flow got during this event; the upper limit of the rating curve is 100 cm (3.38 m stage), but stage continued to rise for another 1.5 m to 4.82 m before loss of transmission (Figure 35).

Purpose

WSC completed a hydrology and hydraulic assessment at the gauge location to determine if it should be rebuilt in a different location. This project was minimal in scale as it was an internal investigation, not a full flood hazard assessment. Therefore, the following assessment is proportional in scale to this project.

Potential climate impacts typically have not been taken into account when determining locations for new WSC gauges or repairing gauges. However, integrating climate impacts into infrastructure decisions will help the WSC network better withstand extremes, so it can provide more consistent data collection that captures extreme events. More measurements (rather than focusing on gauge damage) during extreme floods will also make more data on these events available to users.

Methodology

To create a flood hazard map for the gauge location, WSC undertook a hydrologic analysis, climate change assessment, and hydraulic analysis.

A flood frequency analysis was completed with historic gauge data in HEC-SSP using Bulletin 17 C (Figure 36).

The flood frequency analysis indicates that the 100-year flow is 340 m³/s. However, at this site, a much larger flow occurred. In July 1986, a flow of 580 m³/s was found in the historic record. This value appears to be based on calculations from slope area analysis, as indicated by physical field notes from the event. This calculation has a large degree of error, but it was taken to be accurate as evidence of a large rainfall event and large flows from this time exist. News articles, ECCC weather station precipitation readings, and readings from a nearby downstream gauge (07BA001) indicate there was a large flood at this time.

To begin the climate change assessment for the area, a conceptual model (Figure 37) was created to better understand the area, including causes of floods and how climate change might affect floods. (The conceptual model visually represents how peak flow and floods are expected to be impacted in a changing climate, informing the rest of the assessment process.)

Figure 37 shows the effect that a given climate change impact will have on peak flow, including:

• Temperature is expected to increase (Shrestha et al., 2017), which may result in:
  • Precipitation that may have fallen as snow will fall as rain. It is unclear the exact effect this will have on streamflow, but it may change the frequency and magnitude of large events.
  • Increased instances of rain-on-snow events, which will result in a flashier system exhibiting higher peak flows.
  • Increased atmospheric evaporative demand, which will result in less water available in the watershed during hot times of the year.
• There is the possibility of the loss of tree cover in the basin due to possible forest fires and deforestation. This will reduce rainfall interception from trees, resulting in increased runoff entering the stream.
• Precipitation is expected to increase (Shrestha et al., 2017).

Overall, peak flow magnitude is expected to increase in the watershed. Because quantitative assessment of the impact of climate change on peak flows was not possible due to time and budget constraints, a risk-based approach was completed to assess the impacts of possible shifts in peak flow. A literature review was undertaken to identify a range of expected changes in peak flow (by percentage) due to climate change. As Rat Creek is a small watershed, studies have not been completed on it directly. However, much research has been done in the wider area, such as for the Athabasca River watershed.

A study by Dibike et al. (2019) in the Athabasca near the AOI investigated projected changes (by percentage) in the magnitude of 100-year return period flows under emission scenario RCP 8.5. The study found that these flows will increase between 4–12%, depending on the method of extreme value analysis (Figure 38).

Rat Creek and the Athabasca River will experience similar changes in climate (temperature, precipitation). However, as a smaller basin, Rat Creek is more likely to experience synchronized runoff generation (via rainfall, snowmelt, or a combination of both) than the Athabasca River. Thus, the 4–12% change identified in this study was considered to be the lower end of peak flow increase due to climate change at Rat Creek.

Conversely, in a watershed that is much smaller than Rat Creek with little storage, the maximum percent change in peak flow that can be expected is equal to the percent change in intensity of extreme rainfall events. To estimate the increase in rainfall (and by extension, peak flow) on a very small, flashy watershed in the same region as Rat Creek, a climate change adjusted IDF curve was investigated. It was found that projected change in 100-year 24-hour rainfall intensity is about 20% under SSP2-4.5, which was taken to be the higher end of possible peak flow increases for Rat Creek.

Conclusion

Climate change can potentially increase peak flows in this area by about 4–20%. To complete the risk-based assessment, this range of climate-adjusted flows was hydraulically modelled to analyze the impacts of the increased flows. The hydraulic model of the Rat Creek stream channel near the WSC gauge indicated the inundated areas throughout the range of climate-adjusted flows (Figure 39). Since even the 100-year flood with no adjustment indicated flooding in the AOI it was clear the gauge infrastructure should be relocated to an area less prone to flooding. The 20% climate change adjustment was modelled to identify other areas for the gauge location that are resistant to future flooding.

The right bank to the east of the road was identified as a possible location for the gauge; a location that remained free from flooding during the July 1986 580 cm event. These results were provided to the WSC operations team to finalize the gauge relocation.

Back to Case Studies

---

3.7 Conservation Halton: Local Waterbody Flood Hazard Mapping

Community Profile

Conservation Halton is one of Ontario’s 36 conservation authorities and a local watershed management agency. Its jurisdiction spans the Grindstone, Bronte, and Sixteen Mile Creek watersheds, and 18 smaller urban watersheds that drain to Lake Ontario. This study focused on urban watercourses in the Town of Oakville that drain to Sixteen Mile Creek.

Local flood hazards

The Town of Oakville has experienced significant flooding due to heavy rainfall, like many other southern Ontario municipalities. To understand the magnitude and extent of flood hazards and potential risks to life and property, the project used Ontario’s flood event standards (PDF, 322 KB): the 100-year flood event and the 1954 Hurricane Hazel flood event, both of which are based on heavy rainfall. The Hurricane Hazel event consisted of 285 mm of rain over 48 hours, with 212 mm of rain during the last 12 hours at a peak hourly intensity of 53 mm/hr.

Purpose

The purpose of the project was to update peak flows and develop engineered flood hazard mapping sheets to support land use planning and decision-making for the watercourse contributing flows to the Morrison-Wedgewood Diversion Channel (Figure 41).

Conservation Halton also sought to understand how climate change may increase the frequency and severity of rainfall events, and how it may increase the estimated hydrologic peak flows and associated flood inundation on local waterbodies.

Methodology

The scope of work included an assessment of previous hydrologic models, selection of a modelling platform, and selection of existing and future hydrologic scenarios.

The climate scenario included as part of the hydrologic model was based on the recommendations for future climate change rainfall scenarios (Wood, 2018). The rainfall scenarios were based on Representative Concentration Pathways (RCP) scenario 8.5. The time horizons selected by the Wood study included both 2050 and 2080 rainfall IDF scenarios. The 2080 scenario was selected for assessment due to the expected design life of stormwater infrastructure in the area.

The primary data source for the climate projections was also drawn from Wood. The climate change assessment required by the client included incorporating the recommended approach from the Town of Oakville’s climate change assessment.

The approach was guided by Oakville’s Climate Change Strategy (PDF, 628 KB) (2014) and Review of Future Rainfall Studies (Wood, 2018). In keeping with this study’s recommendations, climate-adjusted IDF projections (based on Western’s IDF-CC Tool [V2]) were applied based on a projected climate year of 2080. A 24-hour Chicago rainfall distribution was applied (per Town standards and findings of the sensitivity analysis), given the absence of long-term flow monitoring data that would support selection of a less conservative distribution.

On August 4, 2014, a series of thunderstorms resulted in significant flooding in a nearby watershed in the City of Burlington. Conservation Halton collected and analyzed available rainfall data and defined the hyetograph associated with the maximum rainfall impacting a 1 km² cell, based on calibrated radar data from the NexRAD station in Buffalo, New York. The maximum rainfall determined for the storm totaled more than 190 mm over seven hours. While the maximum hourly rainfall intensity (56 mm/hr) was similar to the Hurricane Hazel flood event standard, the rainfall hyetograph generated for the August 4th, 2014 storm applied 10-minute time steps (rather than hourly time steps), and so the rainfall distribution included a maximum intensity of 126.7 mm/hr over a 10-minute duration. This maximum cell rainfall from the storm was simulated and compared with flows associated with the regulatory flood hazard (associated with the greater of the 1:100-year or Hurricane Hazel storms) and the climate-adjusted 1:100-year flows.

Conclusion

The climate-adjusted rainfall scenario resulted in the greatest peak flows at four of the six watercourse outlets, with both the Hurricane Hazel flood event and the peak August 4, 2014 flows generating higher peak flows at the remaining two nodes. Only the flows associated with the provincial flood event standards were mapped.

Back to Case Studies

---

3.8 Durham Region: Riverine Crossings and Roads Flood Hazard Mapping

Community Profile

The study area for this project is the riverine regulatory flood plain within the Toronto Region Conservation Authority’s (TRCA) jurisdiction in Durham Region, Ontario. Most of the land in the watershed has undergone significant urbanization but also includes rural agricultural land and golf courses.

There are four TRCA watersheds that are within, or mostly within, the Durham Region boundary:

• Frenchman’s Bay watershed: Approximately 2,700 ha in the City of Pickering, with four sub-watersheds that drain into Frenchman’s Bay and then into Lake Ontario.
• Petticoat Creek watershed: Approximately 2,417 ha, mostly in the City of Pickering, surrounded by the Rouge River, Duffins Creek, and Frenchman’s Bay watersheds.
• Carruthers Creek watershed: An approximate drainage area of 3,971 ha in the City of Pickering and Town of Ajax. The watershed consists of the Main Branch and East and West tributaries, which join with the Main Branch approximately 300 m upstream of Taunton Road East; the creek then drains through Carruthers Marsh into Lake Ontario.
• Duffins Creek watershed: The biggest of the four watersheds, spanning an approximate drainage area of 28,216 ha. Most of the watershed is in the City of Pickering; the watercourse network includes several creeks (Lower Duffins, Main Duffins, East Duffins, West Duffins, and Miller Creek).

Local flood hazards

The intensity, duration, and frequency of extreme precipitation events is linked with how the climate is changing. Climate modelling efforts by the international community demonstrate that as global air temperatures warm, the availability of moisture increases, thereby potentially increasing the amount and intensity of precipitation during extreme weather events. How often extreme rainfall events occur is still an active area of research globally, but numerous academics, municipalities and decision makers are developing approaches to update tools used in infrastructure design to consider future conditions, such as Intensity-Duration-Frequency (IDF) curves.

While it was out of scope to update IDF curves for the Region as part of this work, Savanta and the Climate Risk Institute (CRI) were tasked with providing an approach of “shifting” return periods based on future climate scenarios. This task was guided through the following research question: “What is the expected new return period based on a given historical event (for example, two-year) amount of rainfall?” In other words, this task focuses on “shifting” the annual exceedance probability (AEP) considering climate change, not determining the future amount of extreme rainfall (i.e., in millimetres).

Purpose

This study informed the Region’s Community Climate Adaptation Plan (PDF, 7.9 MB) (2016), focusing on riverine crossings and roads within TRCA’s jurisdiction in Durham Region only, and did not consider flood vulnerability of roads under urban (non-riverine) flooding scenarios.

Methodology

The project team undertook a literature review to identify potential methodologies, analyses, and approaches toward updating return periods in light of future climate change. These included, broadly:

• Temperature scaling;
• Applying a “climate model-derived” safety factor;
• Transposing a design storm from another jurisdiction or altering its application to approximate future conditions; and
• Using a statistical analysis to update annual exceedance probabilities (for example, extreme value or a sensitivity analysis).

Following discussions among TRCA and Durham Region staff, Savanta and CRI confirmed they would undertake a statistical procedure and sensitivity analysis that leveraged future precipitation and storm-related information available across Ontario. More specifically, the project team would conduct an analysis to compare future IDF-related information and interpret the extent to which return periods may be shifting in TRCA watersheds within the region.

The project brought together flood impact information, road criticality information, and hydraulic model information to provide a desktop-based risk ranking of roads at risk of riverine flooding, and to characterize the hydraulic performance of crossings (as reflected in TRCA’s hydraulic models) in comparison to Ontario Ministry of Transportation (MTO) criteria. As part of this study, Savanta and the CRI conducted another study on how to incorporate future climate change in which an approach of “shifting” return periods based on future climate scenarios was developed (i.e., “shifting” the AEP of already-modelled storms considering climate change). The results of this study, namely the projected “shifted return period” tables, were used to add a climate change scenario to the flood vulnerable roads and crossing databases.

Crossing information and results were extracted from TRCA’s current hydraulic models for the four watersheds (Figure 43). The extracted model information was further processed to identify which crossings were pressurized and/or overtopped, and the first return period at which they became pressurized or overtopped (meaning that they could not pass the associated storm event). This provided information on the actual level of service a crossing could provide.

The MTO provides flood design criteria to assign the hydraulic level of service that a crossing is supposed to provide, depending on the road classification, road environment, and crossing span. TRCA compared the actual level of service provided by the crossing and the MTO design level of service, which revealed that several crossings in each of the four watersheds did not provide the current level of service as per MTO design criteria. In addition, the crossing capacity assessment was also undertaken to incorporate future climate change conditions (that is, shifted return periods), which identified more crossings that would not meet MTO criteria in the future, due to the anticipated shifting of return periods for the associated storm flows moving from being less frequent to more frequent.

The flood vulnerability of roads was also assessed by extracting the maximum flood depth for each modelled event over the road for each inundated road segment within the regulatory flood plain. Using consensus-based tables to assign a closure disruption to flood depth, and utilizing daily traffic information for each road segment, the event-based disruption and average annualized disruption (AAD) for the road segments was calculated. AAD is a function of closure time, annual average daily traffic, and AEP. Future climate change scenarios were incorporated in the flood vulnerable road assessment, as reflected by revised AEP in calculations of AAD (Figure 44).

The criticality of road segments was also assessed, based on eight factors: 1) functional classification of roads; 2) average annual daily traffic; 3) designated transit routes; 4) goods movement routes; 5) degree of redundancy; 6) evacuation and disaster recovery—proximity to nuclear hazards; 7) proximity to sensitive receptors; and 8) social equity and justice. Each of the indicators was assigned a score based on a scale from one to five, with an overall criticality score calculated as a sum of the scores for each of the factors for each inundated road segment. The larger the combined score along a road, the more critical a given road segment was. The criticality scores were then converted to a multiplier that was applied to the AAD for a comprehensive flood risk score.

Outcomes

The deliverables from this study include:

1. A geodatabase of crossing capacity information for each watershed, including a crossing point shapefile, and a crossing table for each storm event for both existing and future climate conditions.
2. A geodatabase of road capacity information for each watershed, including the road segment polyline shapefile, the road table including disruption calculation for each storm event for both existing and future climate condition, and the AAD.
3. A geodatabase of criticality ranking for the road segments, including a road segment polyline shapefile for each factor, and one for the criticality score determined by the combined eight factors.
4. An ArcGIS.mxd file with links to each above geodatabase and select map views.
5. A comprehensive report which details methodologies and results, as well as a short executive summary report.

Conclusion

Assessing and reducing climate risks, as well as undertaking comprehensive adaptation efforts, is an iterative process. Infrastructure planners, designers, engineers, and all other municipal staff should expect to revisit their research, results and how they are interpreted, and their implications on a regular basis (for example, every five years) to ensure they can be updated as new science and implicating choices are being made. The following considerations were provided as outputs:

• Always ensure that a fulsome discussion regarding Durham Region’s risk tolerance, associated cost-benefit trade-offs, and the level of regional appetite for upgrading infrastructure exists prior to adapting results contained in this report. Projected shifts in some return periods can be significant and carry large implications at the municipal level.
• Avoid using only the single multi-model ensemble average (“one number”) result in the 2050s- and 2080s-time horizons for future return periods of storms. Rather, consider the range in projected return periods and annual exceedance probabilities as part of future investigations.
• Generally, as these future return periods are applied to public works, additional consideration and/or engineering analysis is warranted to ensure that the results are appropriate for site-specific contexts. Recall that climate scenarios can change, and that there is no single, standard scenario that municipalities can design for. Therefore, enable flexibility wherever possible and examine the range in these results as part of engineering analyses and design criteria.

Back to Case Studies

---

3.9 Mimico Creek: Erosion-Related Flood Mapping

Community Profile

The Mimico watershed is a completely urbanized watershed. More than 30% of its land mass features industrial uses and over 60% of its reach is artificially channelized. Mimico Creek runs 33 km on the west side of Toronto, Ontario, is part of the Great Lakes basin, and is a Lake Ontario tributary.

Local flood hazards

To address erosion-related risks to water infrastructure, more than 200 water risk sites were inventoried across more than 20 km of channel length.

Purpose

The City of Toronto retained Aquafor Beech to undertake the Mimico Creek Geomorphic Systems Master Plan (GSMP) with the intent of identifying, evaluating, and prioritizing erosion-related risks to Toronto water infrastructure in the watershed.

Unique to this study was the integration of climate change modelling to evaluate the impacts of climate change on reach-scale geomorphic processes.

Methodology

Working with EBNFLO, Aquafor Beech developed an innovative modelling process that leverages the outputs of regional climate models as inputs to watershed scale hydrological (HSPF) and hydraulic (HEC-RAS) models to evaluate climate change implications with respect to in-channel erosion indices (velocity, shear stress, stream power).

Results from the climate change assessment were used to refine the evaluation of proposed design alternatives and prioritization of capital works projects. While the emphasis was on understanding the relationship between climate change and in-stream erosion, the assessment also considered flooding implications (i.e., how climate change will impact the seasonality of rainfall and flooding in Mimico Creek and how it will increase the intensity and frequency of existing return period events).

The assessment was a mandatory component of the project scope of work and completed in accordance with the guiding principles defined in Ontario’s Adaptive Management of Stream Corridors guidelines (2001) by the Ministry of Natural Resources (MNR) and the Trent University Watershed Science Centre. It was also designed to apply methods consistent with the MNRF’s Guide for Assessment of Hydrologic Effects of Climate Change in Ontario (2010).

The climate change assessment consisted of two parts requiring different approaches. The first examined the potential impacts of climate change on various components of the hydrological cycle, including air temperature, precipitation, evapotranspiration potential, and total streamflow volume in Mimico Creek. The second focused on the potential impacts of warming climates on the potential severity and frequency of major storm events. The methodologies are discussed in detail below.

Secondary climate change impacts were also considered, particularly with respect to defining climate change impacts on geomorphic processes. The relationship between modelled peak flow rates and stream power was leveraged to evaluate climate change impacts on in-stream erosion rates, taking into consideration reach-scale variability with respect to channel composition (bed and bank materials).

As a response to uncertainty around global climate models (ESMs) and greenhouse gas emission (GHG) scenarios, Aquafor considered six future climate conditions as part of the climate change assessment. They involved three GHG emission scenarios applied at two future time periods, 2050 and 2080. The scenarios and timeframes were as follows:

• RCP 2.6 2050: This low emission scenario (3.0 W/m² of global radiative forcing) results in rising global air temperatures to about the year 2050, followed by a period of temperature stabilization. It examines climate around the year 2050 when global warming would be limited to about 2°C.
• RCP 4.5 2050: This is an intermediate emission rate scenario (4.5 W/m²) that results in higher rates of temperature increase than RCP 2.6. This emission scenario is expected to cause global warming of about 3 °C after stabilizing in the latter part of this century. It examines climate around 2050.
• RCP 8.5 2050: This is a very high emission rate scenario (8.0 W/m²), representing the worst case examined. This emission rate would cause warming greater than 4°C beyond 2100 without stabilization. It represents the climate around 2050.
• RCP 2.6 2080: This scenario/timeframe results in identical climate in 2080 as 2050 due to stabilization of temperature beyond 2050.
• RCP 4.5 2080: This scenario/timeframe represents the RCP 4.5 emission case around the year 2080 when climate stabilization is expected.
• RCP 8.5 2080: This scenario/timeframe represents the RCP 8.5 emission case around the year 2080 as warming continues.

All future climate projections used in support of the Mimico Creek GSMP climate change assessment were sourced from the ClimateData.ca and the Ontario Climate Data Portal.

Assessment of potential impacts for the key factors

As noted above, the Mimico Creek GSMP climate change assessment included two parts requiring different approaches.

Part 1 examined the potential hydrologic impacts of different climate warming scenarios in terms of the resultant runoff and streamflow, including components of the hydrologic cycle in the local study area. The HSPF hydrologic model was used to simulate runoff in the watershed and streamflow in reaches under various climate conditions. The model was calibrated with 2011 to 2016 climate inputs and updated land use data. This simulation timeframe includes some large storms including a significant storm in July 2013. Meteorological inputs, including air temperature, precipitation, dewpoint temperature, and potential evapotranspiration were altered to reflect the various future climate scenarios (Figure 46).

To account for uncertainty around ESMs and GHG emission scenarios, the six future climate conditions were considered involving three GHG emission scenarios applied at two future time periods (2050 and 2080). The reference climate (2011–2016) and the climate of the different climate change scenarios were compared in terms of monthly average, maximum and minimum air temperature, and precipitation volume for each future timeframe. The change or delta values for future climates were determined by averaging the results for 20-year periods. For example, to characterize the results around the year 2050, the simulation results for 2040 through 2059 were averaged. Climate input series data used in hydrologic modelling was synthesized for each climate scenario by factoring the reference period climate time series by the delta factors determined on a monthly basis. Meteorological inputs were synthesized for air temperature, dewpoint temperature, and precipitation in this manner. HSPF was then run for the period 2010–2016, with 2010 treated as a spin-up year for each climate scenario. Model outputs were summarized for 2011–2016 in terms of annual, monthly, and seasonal runoff/streamflow. Streamflow time series were also compared directly across climate change scenarios to explain differences and impacts.

Part 2 of the climate change assessment examined the potential impacts of climate warming on the occurrence and intensity of major storms in the study area in the future and their impact upon in-stream flow. The approach developed in this study for assessing the potential hydrologic impacts of future storms on geomorphological systems relies on RCM simulations conducted across North America and focuses on the largest storms typically observed in the study area.

This rational approach relates storm severity in the future to air temperature increase. It follows that atmospheric water holding capacity will increase as global warming worsens and, under ideal storm generation conditions, it is more likely that storms will be able to accumulate more moisture. This worsening of storms should be realized in terms of storm magnitude (intensity and duration) as well as frequency. The World Meteorological Organization has published definitions and procedures for estimating a hydrometeorological expression termed probable maximum precipitation (PMP), defined as the greatest depth of precipitation for a given duration meteorologically possible for a watershed or location at a particular time of year. It has been hypothesized that PMP is temperature dependent and should not be considered constant into the future. Future increases in PMP in a particular area may be directly related to warming. This increase in PMP would be due to the atmospheres’ water holding capacity increasing.

Studies focused on the Great Lakes region, which encompasses the Mimico Creek GSMP study area, project a 4% increase in PMP per °C. Using this relationship, future major storms relevant to the Mimico Creek Watershed (i.e., design storms) were synthesized by adjusting precipitation volumes proportionally to projected increases in temperature. These adjustments were made to reflect future increases in PMP. Hydrologic model runs were then made to simulate each storm response for the present day and future time periods, thereby allowing for an analysis of climate change impacts.

Outcomes

Key conclusions from the first part of the climate change assessment included lower summer precipitation in the future leading to decreased baseflow and the possibility of severe drought (Figure 37). During the colder months (November to April) snowpack development will be significantly reduced and nearly eliminated in the worst-case scenario. With the loss of snowpack and warming air temperatures in the winter, surface waters will be subject to increased evaporation, leading to less winter storage of water across the watershed, which will in turn reduce the likelihood of a typical spring freshet. However, more streamflow will occur throughout the winter due to more precipitation in the liquid form, more frequent snowmelt, and higher precipitation in general. In both the warm and cold months, the risk of severe thunderstorms will increase due to climate change.

The conclusion of the second part of the study was that climate change will increase the potential for more severe storms to occur, in part due to increasing the atmospheric water holding capacity. It is also expected that climate change will increase the severity of major storm events when they do occur. It is estimated for the Mimico Creek GSMP study area that major storm rainfall may increase by factors of 8% to represent 2050 storms and 11% to represent 2080 storms (Figure 48).

In terms of secondary climate change impacts, on average it was projected that lateral (bank) erosion rates would increase by 14% by 2050 and 31% by 2080 when compared to existing climate conditions. Conversely, vertical (channel bed) erosion rates are projected to increase by 38% by 2050 and 70% by 2080.

Conclusion

The implications of these results included shortened expected lifespans for existing erosion control measures and further expected enlargement of the existing Mimico Creek channel form. It is the intent of the GSMP to leverage this understanding of future climate change implications on in-stream erosion to select and develop channel restoration solutions that will be resilient to future climate change impacts, to ensure the long-term protection of at-risk Toronto water infrastructure.

Back to Case Studies

---

3.10 Gaspereau Primary Watershed: Municipal Flood Line Mapping

Community Profile

The Town of Kentville is the most populous community in Nova Scotia’s Annapolis Valley. In 2022, the province granted funding to the Town to engage Dillon Consulting in carrying out flood line mapping in the Gaspereau primary watershed. The project committee includes representatives from the Department of Municipal Affairs (DMA) and the municipalities that intersect with the watershed.

Local flood hazards

The Gaspereau primary watershed drains east towards the Minas Basin, which experiences some of the highest tides in the world with a range up to 16 m. Numerous smaller secondary (shore direct) watersheds discharge directly to the Bay of Fundy; each of these watercourses are tidally influenced with numerous communities and agricultural lands located along the shoreline and within the dykelands. The watershed drains towards the heart of the Annapolis Valley, with surface water travelling through multiple communities, historical sites, tourist centres, and agricultural regions. Historically, flooding has been a fairly regular occurrence in the watershed, with flood events occurring in 2003, 2008, and more recently, in July 2023 and July 2024.

Purpose

In 1999, the province enacted regulations under the Municipal Government Act through a Statement of Provincial Interest (SPI) to set minimum criteria and planning standards within floodplain areas. Between 2018 and 2020, the DMA developed a Municipal Flood Line Mapping (PDF, 1.54 MB) (MFLM) document to provide a set of standards that apply across the entire province, creating consistency for how flood line mapping is carried out. The end goal was to ensure that all municipalities apply the Province’s SPI on Flood Risk to areas in their jurisdiction that are likely to flood, including the Gaspereau primary watershed.

Methodology

Provincial staff created a Climate Change Standard prior to developing the MFLM document. The objective was to develop a consistent framework for the incorporation of future climate changes into riverine and coastal flood mapping in Nova Scotia. The framework was developed with the intention of providing a scientifically defensible, consensus-based approach that is practical to implement. The standard was developed in consultation with experts from multiple universities (Dalhousie, St. Francis Xavier, St. Mary’s). An interdisciplinary working group included researchers in municipal planning, water resources engineering, coastal hydraulics, and climatology.

The Climate Change Standard prescribes minimum analysis scenarios for precipitation and coastal water levels (storm surge and tide) for both current and future conditions. Climate change is layered on existing scenarios. Identified scenarios are to have the rainfall amount and sea levels increased by the respective values, as prescribed by the standard.

In addition to the baseline scenarios for the flood line criteria that apply to existing development conditions, potential future scenarios that are specific to the study area must be considered. These may include modifications to the current state of development of the watersheds or the current state of the drainage system.

Climate change standard requirements included:

Emissions scenarios: The Representative Concentration Pathway of 8.5 W/m² (RCP8.5 [AR5]) is to be used for all climate change projections used in riverine and coastal flood mapping studies.

Time horizon: The future time horizon for mapping riverine and coastal flood boundaries is to extend until 2100. An interim flood boundary should be produced with a time horizon extending to 2050 for interim municipal planning purposes.

Intensity-Duration-Frequency curves: Inputs to the hydrologic model are to include 1:20 and 1:100 rainfall events derived from a future climate IDF relationship. A dual-method approach consisting of estimates from both the statistical IDF_CC tool (Institute for Catastrophic Loss Reduction, Western University) and the semi-physical Clausius-Clapeyron scaling method (C-C) were to be used as per the following:

IDF-CC tool specifications: The complete “Bias-Corrected Ensemble” is to be used to generate future IDF relationships for the time horizon of 2050–2100.

The outputs are to be used to determine the 95th percentile rainfall totals/intensities for the storm durations of interest.

C-C scaling method specifications: Statistically downscaled and/or dynamically downscaled daily maximum temperature projections are to be used. Either or both datasets should be obtained from the following sources:

The Environment and Climate Change Canada (ECCC) Climate Data Extraction Tool should be used to obtain an ensemble of statistically downscaled daily maximum temperature projections.

Dynamically downscaled daily maximum temperature ensemble projections should be obtained from the Coordinated Regional Climate Downscaling Experiment (CORDEX).

The future projection period shall be 2070–2100, and the historical period is to be determined by a qualified professional given the available observational data for the site of closest interest.

The IDF relationship based on observational climate data is to be scaled using the difference in the 95th percentile of average daily maximum temperature between modelled historical and future time periods. Scaling factors are to be used for the C-C scaling calculation, where:

A scaling factor of 7% per 1-degree Celsius rise in temperature is to be used to scale historical IDF relationships for rainfall durations greater than six hours.

A scaling factor of 14% per 1-degree Celsius rise in temperature is to be used to scale historical IDF relationships for rainfall durations of less than or equal to six hours.

The 1:20-year and 1:100-year rainfall totals are to be used in flood mapping and shall be the higher of the rainfall values yielded from both methods (i.e., IDF_CC and C-C).

Coastal flood and downstream riverine boundary conditions

The total sea level to be used in riverine and coastal flood mapping is to be computed as per Equation 1:

To determine the HHWLT, the HHWLT grid from the Hydrographic Vertical Separation Surfaces from the Canadian Hydrographic Service and the Canadian Geodetic Survey was used.

The 95th percentile of the (James et al., 2014) projected global relative sea level rise (RSLR) is to be used, for RCP8.5 and projected to year 2100 for the location nearest to the area of interest. Local effects, such as tidal expansion in the upper Bay of Fundy region, should also be considered.

An additional 65 cm should be added to the RSLR projection to account for the possibility of the melting of the West Antarctic Ice Sheet.

The 1:20-year and 1:100-year historical storm surge is to be included. The methodology used for the estimation of this term should be determined by a qualified professional given the available historical data or model projections.

Outcome

The Nova Scotia provincial climate change standard provides guidance to practitioners on the appropriate approaches for incorporating climate change impacts into flood hazard assessments in the province.

Back to Case Studies

---

3.11 Nova Scotia Sissiboo-Bear Rivers: Primary Watershed Flood Line Mapping

Community Profile

The study is located primarily in the Municipality of the District of Digby, with the eastern side of the Bear River Estuary in the Municipality of Annapolis County, Nova Scotia (NS). Beyond the Sissiboo and Bear Rivers, the coastal extent of the study area extends from within the Annapolis Basin, through the west side of Digby Gut, along the Bay of Fundy, west along Digby Neck, and encompassing Brier Island, as well as the inside portion of St. Mary’s Basin ending near the Village of Weymouth at the mouth of the Sissiboo River (Figure 51).

Purpose

The primary objective of this study was to prepare a set of defensible flood extent and hazard maps for major rivers and coastline within the study area, considering both existing and projected future conditions, to help guide decision-making and planning. Historically, major floods in the area were mainly caused by storm surge combined with the Bay of Fundy’s extreme high tides. In addition, along the study area major rivers, there are specific flood-prone areas above the estuarine zone, such as near the confluence of the East Branch and West Branch of the Bear River.

The impacts of flooding and climate change in these areas can include not only infrastructure damage, but also public safety risks, economic impacts, and severe environmental impacts. Waterfront development is becoming increasingly vulnerable to flooding as development densifies and climate change continues to pose greater risks. Waterfront infrastructure along River Road near the Village of Bear River and Evangeline Trail are examples of valuable assets that provide essential services to residents.

Methodology

This project produced an extensive set of flood maps covering approximately 23 km of river length on the Sissiboo and Bear Rivers and approximately 200 km of coastline along the Annapolis Basin, Digby Neck and Islands, and St. Mary’s Bay. Flood mapping includes the six defined scenarios in the province’s Municipal Flood Line Mapping (PDF, 1.54 KB) (MFLM) specifications, including the 5% and 1% AEP for current, 2050, and 2100 climate conditions. Map products include flood lines, flood depths, flood velocity, and hazard. This range of flood scenarios and flood mapping products provides a rich dataset to help portray flood risk in the study area. Flood lines for all scenarios are provided with the report, as well as hazard mapping of current climate conditions.

The study produced flood maps for six scenarios, shown in Table 6:

Table 6: Summary of Scenarios Used to Produce Flood Maps.

Annual Exceedance Probability (%)	Description	Time Horizons for Climate Condition
5	The 1 in 20-year return period, described as “Floodway” in the Statement of Provincial Interest	Current
		2050
		2100
1	The 1 in 100-year return period, described as “Flood Fringe” in the Statement of Provincial Interest	Current
		2050
		2100

This study followed the draft MFLM specifications (final published 2024). The project progressed in the following major phases:

1. Review of previous studies, reports, and flood events.
2. Water level monitoring and bathymetric surveying of rivers and bridge structures.
3. Stakeholder engagement via questionnaire and discussion with stakeholders.
4. Climate change and sea level rise analysis.
5. Coastal water level analysis.
6. Hydrological analysis including review of historical flood events, development of hydrologic model of the watersheds.
7. Hydraulic analysis and modelling of rivers.
8. Production of flood maps for riverine and coastal water levels.

The Statement of Provincial Interest on Flood Risk Areas establishes the 1 in 20-year return period event as the floodway and the 1:100-year return period event as the flood fringe. This study used the term annual exceedance probability (AEP) to describe the probability of a flood event.

Flood mapping methodology

Flood lines for all six flood scenarios (present, 2050, and 2100 in combination with 5% and 1% AEP) covering the coastline and rivers of the study area were prepared following the MFLM requirements. In addition, flood hazard maps for the rivers were developed based on the MFLM hazard classification (depth x velocity). Flood lines developed for coastlines and rivers were merged into a single continuous flood line to represent the conditions of the AEP of that scenario, as in Figure 52.

Regarding the coastal flood line mapping, note that the maps represent static water levels, without the effect of wave run-up. During a storm event, other dynamic components will contribute to the extreme water levels observed. Wave run-up, for example, can significantly enhance the damage caused. Wave run-up at exposed coastline areas in the Bay of Fundy can be in the order of metres, while more sheltered areas will be limited by the fetch length and water depths near the coast.

Depth, velocity, and hazard mapping was carried out for the 5% AEP and 1% AEP events in current and future climate conditions. These maps contain extensive information that could be useful for emergency management and risk management of people and property, as well as to support land use planning.

In addition to flood lines maps, depths, velocity, and hazard maps can provide rich detail to describe the potential hazards from flooding. The following examples show the depth, velocity, and hazard from riverine flooding under the 1% AEP scenario at the Village of Weymouth. Depth maps show the depth of water over each square metre of land during the peak of the flood and are a helpful representation for visualizing flooding (Figure 53).

Velocity maps are generated based on the 1D velocity profiles at each model cross section and on interpolated data between cross sections. These maps are helpful to identify areas with swiftwater hazards and locations that may be susceptible to erosion damage (Figure 54).

Hazard mapping is created by multiplying flood depth by the flow velocity as a proxy for the level of danger posed to people. Velocity is important to include, as shallow but swift flowing water can pose a high hazard. Based on the MFLM specifications, hazard was defined as velocity (m/s) multiplied by depth (m) and categorized into three classes described in Table 7. Coastal flooding has been included to illustrate the potential extent of flooding and assigned a nominal class of 1. The coastal flood hazard mapping does not represent the actual expected hazard but illustrated the extent of potential flooding. A sample is shown in Figure 55. At this location, several roads, many buildings, and the fire station are within a Class 1 hazard area. Infrastructure including roads, bridges, and coastal harbours, are at risk of inundation from current and future flood events.

Table 7: Current Conditions 1% Riverine Flood Velocity Mapping at Weymouth.

Class	Hazard Value (Depth [m] * Velocity [m/s])	Level of Hazard
0	Below 0.5	Caution
1	0.5 to 1.5	Danger to Some
2	1.5 to 2.5	Danger to Most
3	Above 2.5	Danger to All

Outcomes

The findings and recommendations of this study were based on available information at the time of completion with uncertainties associated such as, gaps in data, unknown effects of dam and reservoir operations, future rainfall, and sea level rise projections under climate change, and more.

Conclusion

This flood mapping exercise demonstrated the presence of flood risk at several locations around the municipality, most notably at the villages of Bear River and Weymouth. Policies and bylaws will require updating to reflect the findings of the flood study. For example, CBCL recommended that the District of Digby update its Municipal Planning Strategy (MPS) and bylaws to acknowledge the risk and enact policies to address it. They further recommended comprehensive consultations with First Nations, stakeholders, municipal staff, and the public to ensure that everyone understands the purpose of any changes and its importance for improving public safety.

CBCL also recommended that the municipality and the provincial Department of Municipal Affairs collaborate with Nova Scotia Power to evaluate the potential influence of hydroelectric facilities on downstream flooding. Further information on the operations of the system during extreme events will help reduce uncertainty in the flood line mapping. As well, it is recommended that the effect of dam operations be considered for flood line mapping.

Additional analysis and modelling were also recommended for coastal flooding, including analysis of wave exposure, wave run-up and overtopping for relevant areas, as well as other related analysis and modelling to enhance the value of the present flood study.

CBCL recommended that the flood maps be reviewed carefully by the municipality and local partners to ensure they are consistent with local knowledge.

Back to Case Studies

---

3.12 Prince Edward Island: Pluvial Flood Mapping Project

Community Profile

Surrounded by the Atlantic Ocean and the smallest province in Canada, Prince Edward Island (PEI) is susceptible to intense and frequent storm surges, coastal erosion, and floods. The province is 5,681 km² and has a population of more than 154,000 (2021).

The project team developed detailed pluvial flood maps for six municipalities, two Indigenous communities, and the entire island of PEI.

Purpose

In recent years, Islanders have been seeing more intense and frequent heavy rainfalls that always cause unexpected floods.

Although the Government of PEI has offered publicly available province-wide online mapping for coastal flooding from storm surge and sea level rise since 2021, the mapping did not include pluvial flood risks associated with intense rainfall. This project aimed to address that gap by modelling pluvial flood risks under both current and future climate conditions. The result is a set of island-wide pluvial flood maps designed to help Islanders plan for flood mitigation and climate adaptation.

Methodology

Climate-scaled Intensity-Duration-Frequency (IDF) curves developed by ECCC were used to support the climate change assessment for this project. Due to the tight project timeline and budget limitation, this project considers one future climate scenario under SSP5-8.5. Based on the recommendation from the technical advisory committee, four return periods (10-, 25-, 50-, and 100-year) were considered for both current climate and future climate.

The HEC-RAS 2D model was selected to conduct pluvial flood modelling for PEI, after a preliminary and comparative assessment of three flood models (i.e., HEC-RAS 2D, PCSWMM, and FloodMapper). It was used to drive pluvial flood maps based on two configurations, including detailed maps for major municipalities and an island-wide level. The model was validated with the flood event data for September 2, 2021 (for Charlottetown) (Figure 57) and July 3–4, 2023 (for three counties across PEI). The rainfall IDF curves with four return periods (10-, 25-, 50-, and 100-year) under current and future climate condition were used to design rainfall events for pluvial flood mapping.

Throughout this study, the project team encountered many challenges, as this was the first attempt to generate island-wide pluvial flood maps for PEI. Challenges included the representation of stormwater systems in cities, of the impacts of ocean tides on those systems, and of hydraulic structures (bridges and culverts), as well as determining the feasibility of the flood model for province-wide flood modelling.

The data used to run the HEC-RAS 2D model (for land use, soil, roadways, watercourses, hydraulic structures, stormwater systems, rainfall IDF curves under current and future climate conditions, flood events) was obtained from existing sources and multiple field visits.

To generate the pluvial maps, the model setup included a hydrological analysis at both municipal and island-wide levels. This involved watershed delineation and identification of its characteristics and parameters. It also involved the design of rainfall events: four design rainfall storms under two climate conditions, to generate eight scenarios (Table 8). The model setup also included analysis of hydraulic structures (and considerations of coastal and downstream boundary conditions), and simulation setup (setting parameters, a warm-up period, and configurations) at both municipal and island-wide levels.

Table 8: Eight Scenarios for Design of Rainfall Events

No.	Climate Condition	IDF Return Period (years)
1	Current Climate	10
2	Current Climate	25
3	Current Climate	50
4	Current Climate	100
5	Future Climate (SSP5-8.5, 2071–2100)	10
6	Future Climate (SSP5-8.5, 2071–2100)	25
7	Future Climate (SSP5-8.5, 2071–2100)	50
8	Future Climate (SSP5-8.5, 2071–2100)	100

Because complete model calibration was not possible (due to a lack of calibration data), a preliminary sensitivity analysis was performed to identify an optimal combination of parameter values for antecedent moisture conditions, Manning’s coefficient, and infiltration rates that yielded satisfactory results compared to observed events.

Considering the constraints in hydrometric data, different flood data was used for validating the flood models (the September 2, 2001 flood for Charlottetown and July 3–4, 2023 flood for the island). In addition to flow rates, the project team also attempted to collect the water level data at these three hydrometric stations to support further comparisons (however, these did not provide necessary information). Additional verification of model parameters was also considered by looking at the infiltration ratio patterns: four test runs were performed and the results were found to be reasonable. The project team also conducted a cross-check of the differences between the model and island-wide models.

To ensure the quality of the developed pluvial flood maps, the project team undertook quality assurance and control measures throughout the HEC-RAS model runs and the post-processing of output maps. In particular, the percent error in water volume accounting was used as a criterion to check the validity of a model run; the output pluvial flood maps from the HEC-RAS were carefully inspected and post-processed to ensure a certain level of consistency for final map presentation.

Outcomes

The resulting pluvial flood maps are available for public viewing through the PEI Climate Hazard and Risk Information System (CHRIS), an interactive map portal where users can browse to visualize different climate hazards that affect PEI communities and properties. These flood maps are available for four return periods (including 10-, 25-, 50-, and 100-year) and two climate conditions (current climate condition and future climate scenario for the period of 2071–2100 under SSP5-8.5) (See Figure 58 and Figure 59).

Conclusion

PEI is facing an increasing risk of compound floods caused by rising sea levels, extreme storm surges, and extreme rainfall. Due to the lack of research on the compounding flood risk in the region, this project only considered normal ocean water level (1.6-year return period) to drive island-wide pluvial flood modelling. However, in the context of climate change, the compound floods may become more frequent. This means that the flood maps might need to be updated in the near future when new information becomes available.

Given the limited scope and the tight timeline of this project, the project team introduced several assumptions to ensure the successful delivery of the project:

• The HEC-RAS 2D model selected for this project is unable to represent the urban stormwater systems in urban areas.
• Ocean boundary conditions are critically important for pluvial flood modelling over low-lying coastal areas like PEI, where people have seen some unprecedented post-tropical storm events in recent years.
• Although the island-wide pluvial flood maps generated from this project can provide useful information for decision-making, there are some limitations (for example, further efforts should be made to better represent the major hydraulic structures in island-wide pluvial modelling).
• Because the IDF curves are based on a very limited number of weather stations in PEI that cannot fully represent the spatial variations of rainfall intensities, the project team recommends using the University of PEI’s Canadian Centre for Climate Change and Adaptation’s weather monitoring network, or using radar or satellite observations.
• Finally, the model calibration and validation in this project are limited due to the lack of high-quality data.

Back to Case Studies

---

3.13 Newfoundland and Labrador: Climate Change Flood Risk Mapping Study

Community Profile

The Avalon Peninsula is a large peninsula that makes up the southeast portion of Newfoundland and Labrador, about 130 km from the capital of St. John’s. Approximately 9,000 km², it is home to more than half of the province’s population. Placentia is located to the southwest, while Carbonear, Victoria, and Salmon Cove are to the northeast.

Communities throughout the province have a pattern of settlement in localized areas such as bays and coves, rather than being evenly distributed along the coastline.

Local flood hazards

Flooding in the region is driven by both coastal and riverine processes. Coastal hazards include storm surge, wave run-up, and sea level rise, while riverine flooding is caused by heavy rainfall and ice jamming. Climate change is expected to exacerbate both mechanisms, increasing flood frequency and severity.

In Placentia, low-lying areas are particularly vulnerable to coastal flooding. In Carbonear and Salmon Cove, inland development mitigates coastal risk, but riverine flooding remains a concern. Victoria, being inland, is primarily affected by riverine flooding.

Purpose

After Hurricane Fiona (September 2022) exposed the flood risks facing hundreds of coastal communities, the province began incorporating coastal flood mapping into its hazard studies. This study marked the first use of coastal flood hazard mapping and the province’s initial step toward a comprehensive coastal flood mapping approach. The new maps that were developed highlight vulnerable areas and inform flood management and infrastructure planning.^{Footnote 8}

The project updated and developed new flood risk and hazard maps for the four communities. It considered both current conditions and projected climate change conditions, including coastal flooding (storm surge, sea level rise, wave run-up), open water riverine flooding (i.e., from heavy rainfall), and ice jam flooding. Detailed hydrologic and hydraulic models, surveys, and climate analyses were used to guide land use planning and disaster mitigation.

The purpose of the project was to develop updated flood risk maps for 1:20 and 1:100 AEP events under both current and projected future climate conditions. It also aimed to assess the impacts of climate change on key environmental factors such as precipitation patterns, sea level rise, and ice processes. In addition, the project evaluated the capacity of hydraulic structures and reviewed flood forecasting strategies to enhance preparedness. The findings are intended to support the Water Resources Management Division and communities in climate adaptation planning efforts.

Methodology

The province’s approach to coastal flood mapping was shaped by expert collaboration and the province’s unique geography and population distribution. This study involved a comprehensive field program, hydrologic, hydraulic and hydrodynamic modelling. The study was completed following the Newfoundland Technical Guidelines for Flood Hazard Mapping Studies (PDF, 2.3 MB) and federal National Disaster Mitigation Program funding criteria.

For the field program,

• 520 river cross sections were surveyed;
• 78 coastal bathymetric transects were completed;
• 100 bridge and culvert crossing sites were surveyed;
• Water level monitoring equipment was deployed at five sites; and
• Flow measurements were taken at seven locations.

For the hydrologic modelling,

• Tools: HEC-HMS, Single Station Frequency Analysis (SSFA), Regional Flood Frequency Analysis (RFFA)
• Inputs: Updated IDF curves, SPOT-6 imagery, National Soil Database data
• Outputs: Flow estimates for 20- and 100-year AEP events under current and future climate scenarios.

For the hydraulic modelling,

• Tools: MIKE21 FM HD (coastal), HEC-RAS (riverine), RIVICE (ice jam)
• Calibration: Based on observed water levels and flow measurements
• Outputs: Flood levels, velocities, depths and hazard classifications.

Outcomes

The climate change impacts were as follows:

On river flows:

• An increase of 14 to 18% in Placentia.
• An increase of 21 to 50% in Carbonear, Victoria, Salmon Cove.

On sea level rise:

• An increase of 0.62 m in Placentia (Figure 61).
• An increase of 0.63 m in the area of Carbonear and Salmon Cove (Figure 62).

Flood risk mapping

Flood hazard maps were developed for 1:20 and 1:100 AEP events under current and future climate conditions. Flood risk maps were developed showing depth, velocity, and risk classification. The new maps highlight vulnerable areas to inform flood management and infrastructure planning and were validated with municipal representatives.

Infrastructure vulnerability

The study identified several culverts and bridges as undersized. In Placentia, the flood wall was determined to be sufficient for current conditions but will be overtopped under 100-year climate change scenario. It also found the Victoria wastewater lagoon dike to be adequate but close to ice jam levels.

Conclusion

A regional approach to coastal flood mapping was not suitable for regulatory level flood mapping and a more localized mapping strategy was required. Additionally, both storm surge and wave impacts are significant concerns for coastal communities. Wave-related flooding is more periodic and shorter in duration when compared with storm surge and riverine flooding and was therefore assessed separately in flood mapping studies.

The results from this study show that climate change will significantly elevate flood risks across all communities, underscoring the urgency of proactive planning. By employing an integrated modelling approach, the project generated robust and reliable flood hazard data and mapping that can inform decision-making. To effectively mitigate future risks, it is essential to pursue infrastructure upgrades, implement policy changes that reflect the realities of climate change and continue to update flood hazard mapping studies, and the approaches applied in this study

4.0 References

• Alberta Geological Survey. (2021). Hydrogeological Regions of Alberta (GIS data, polygon features).
• APEGBC. (2017). Flood Mapping in BC. APEGBC Professional Practice Guidelines (V1.0; p. 50).
• Associated Engineering. (2021a). Flood Mitigation Program Schematic Design and Downtown Detailed Design Hydrology and Climate Change Assessment (p. 51) [Prepared for the City of Grand Forks].
• Associated Engineering. (2021b). Investigations in support of flood strategy development in BC: B-1: Impacts of Climate Change (p. 102) (PDF 4.4 MB). Prepared for the Fraser Basin Council.
• BC Ministry of Forests, Lands, Natural Resources Operations and Rural Development. (2018). Flood Hazard Area Land Use Management Guidelines (p. 84) (PDF, 1.98 MB). Province of BC.
• Calvin, K., Dasgupta, D., Krinner, G., Mukherji, A., Thorne, P. W., Trisos, C., Romero, J., Aldunce, P., Barrett, K., Blanco, G., Cheung, W. W. L., Connors, S., Denton, F., Diongue-Niang, A., Dodman, D., Garschagen, M., Geden, O., Hayward, B., Jones, C., … Péan, C. (2023). IPCC, 2023: Climate Change 2023: Synthesis Report. Contribution of Working Groups I, II and III to the Sixth Assessment Report of the Intergovernmental Panel on Climate Change [Core Writing Team, H. Lee and J. Romero (eds.)]. IPCC, Geneva, Switzerland. Intergovernmental Panel on Climate Change.
• Cannon, A. J. (2015). Selecting GCM scenarios that span the range of changes in a multimodel ensemble: Application to CMIP5 climate extremes indices. Journal of Climate, 28(3), 1260–1267.
• Chernos, M., Macdonald, R. J., & Green, K. (2020). The hydrological effect of forest disturbance on the Kettle River Watershed, B.C. (p. 44) [Prepared for the B.C. Ministry of Forests, Lands, Natural Resource Operations, and Rural Development]. MacDonald Hydrology Consultants, Ltd. and Apex Geoscience Consultants, Ltd.
• Clarke, G. K. C., Jarosch, A. H., Anslow, F. S., Radić, V., & Menounos, B. (2015). Projected deglaciation of western Canada in the twenty-first century. Nature Geoscience, 8(5), 372–377.
• Craig, J. R., Brown, G., Chlumsky, R., Jenkinson, W., Jost, G., Lee, K., Mai, J., Serrer, M., Snowdon, A. P., Sgro, N., Snowdon, A., & Tolson, B. A. (2020). Flexible watershed simulation with the Raven hydrological modelling framework. Environmental Modelling & Software, 104728.
• Dibike, Y., Eum, H.-I., Coulibaly, P., & Hartmann, J. (2019). Projected Changes in the Frequency of Peak Flows along the Athabasca River: Sensitivity of Results to Statistical Methods of Analysis. Climate, 7(7), Article 7.
• Ebbwater Consulting, Inc. (2023). Adaptive Flood Management: From Fragility to Flexibility—Final Report (p. 20) (PDF 7.70 MB). Prepared for Environment and Climate Change Canada.
• Eichelberger, S., Mccaa, J., Nijssen, B., & Wood, A. (2008). Climate change effects on wind speed. North American Windpower, 7, 68–72.
• El Adlouni, S., Ouarda, T. B. M. J., Zhang, X., Roy, R., & Bobée, B. (2007). Generalized maximum likelihood estimators for the nonstationary generalized extreme value model. Water Resources Research, 43(3), 1–13.
• Engineers and Geoscientists BC. (2018). Legislated Flood Assessments in a Changing Climate in BC.” Guideline Version 2.1. Professional Practice Guidelines (PDF, 3.2 MB). Engineers & Geoscientists British Columbia.
• Fraser Basin Council & BC Ministry of Water, Land and Air Protection. (2004). Flood Hazard Map User Guide (PDF, 1.22 MB).
• Gilleland, E., & Katz, R. W. (2016). extRemes 2.0: An Extreme Value Analysis Package in R. Journal of Statistical Software, 72(8).
• James, T. S., Henton, J. A., Leonard, L. J., Darlington, A., Forbes, D. L., & Craymer, M. (2014). Relative sea-level projections in Canada and the adjacent mainland United States (No. 7737; p. 7737).
• Khaliq, M. N. (2019). An Inventory of Methods for Estimating Climate Change-Informed Design Water Levels for Floodplain Mapping (p. 119). National Research Council Canada.
• Kharin, V. V., & Zwiers, F. W. (2005). Estimating extremes in transient climate change simulations. Journal of Climate, 18(8), 1156–1173.
• Kharin, V. V., Zwiers, F. W., Zhang, X., & Hegerl, G. C. (2007). Changes in Temperature and Precipitation Extremes in the IPCC Ensemble of Global Coupled Model Simulations. Journal of Climate, 20(8), 1419–1444.
• Kucharski, J., Steinschneider, S., Herman, J., Olszewski, J., Arnold, W., Rahat, S., Maendly, R., & Ray, P. (2024). Bridging the Gap Between Top‐Down and Bottom‐Up Climate Vulnerability Assessments: Process Informed Exploratory Scenarios Identify System‐Based Water Resource Vulnerabilities. Water Resources Research, 60(11), e2023WR036649.
• Marchau, V. A. W. J., Walker, W. E., Bloemen, P. J. T. M., & Popper, S. W. (Eds.). (2019). Decision Making under Deep Uncertainty: From Theory to Practice. Springer International Publishing.
• McCabe, G. J., Clark, M. P., & Hay, L. E. (2007). Rain-on-snow events in the western United States. Bulletin of the American Meteorological Society, 88(3), 319–328.
• Milly, P. C. D., Betancourt, J., Falkenmark, M., Hirsch, R. M., Kundzewicz, Z. W., Lettenmaier, D. P., & Stouffer, R. J. (2008). Climate change: Stationarity is dead: Whither water management? Science, 319 (5863), 573–574.
• Musselman, K. N., Lehner, F., Ikeda, K., Clark, M. P., Prein, A. F., Liu, C., Barlage, M., & Rasmussen, R. (2018). Projected increases and shifts in rain-on-snow flood risk over western North America. Nature Climate Change, 8(9), 808–812.
• Nasirzadehdizaji, R., & Akyuz, D. E. (2022). Predicting the potential impact of forest fires on runoff and sediment loads using a distributed hydrological modeling approach. Ecological Modelling, 468, 109959.
• Natural Resources Canada. (2025a). Federal Flood Mapping Guidelines Series. Natural Resources Canada.
• Natural Resources Canada. (2025b). Geomorphic Considerations in Flood Mapping (No. 159e; Federal Flood Mapping Technical Bulletins Series, p. 62). Natural Resources Canada.
• Natural Resources Canada & Public Safety Canada. (2018). Federal Flood Mapping Framework (No. 112e; version 2.0, p. 112e).
• Northwest Hydraulic Consultants, Ltd. (2021). Bulletin 2020-1-RFFA: British Columbia Extreme Flood Project, Regional Flood Frequency Analysis. Prepared by Northwest Hydraulic Consultants Ltd. For the former BC Ministry of Forests, Lands, Natural Resource Operations and Rural Development.
• Pacific Climate Impacts Consortium. (2020, February). VIC-GL BCCAQ CMIP5 RVIC: Station Hydrologic Model Output.
• Pike et al., R. G. (2010). Chapter 19: Climate Change Effects on Watershed Processes in British Columbia. In Compendium of Forest Hydrology and Geomorphology in British Columbia, Volume 1 (PDF, xxx MB).
• Public Safety Canada. (2018, December 21). Disaster Financial Assistance Arrangements.
• Rhoades, A. M., Jones, A. D., & Srivastava, A. (2020). The Shifting Scales of Western U.S. Landfalling Atmospheric Rivers Under Climate Change.
• Seglenieks, F., & Temgoua, A. (2022). Future water levels of the Great Lakes under 1.5 °C to 3 °C warmer climates. Journal of Great Lakes Research, 48(4), 865–875.
• Shrestha, N. K., Du, X., & Wang, J. (2017). Assessing climate change impacts on fresh water resources of the Athabasca River Basin, Canada. Science of The Total Environment, 601–602, 425–440.
• Town of Oakville. (2014). Climate Change Strategy—Technical Report (V1.1) (PDF 2.8 MB).
• UNDRR. (2015). Sendai Framework for Disaster Risk Reduction 2015–2030 (p. 32). United Nations.
• Vidrio-Sahagún, C. T., He, J., & Pietroniro, A. (2023). Nonstationary hydrological frequency analysis using the Metastatistical extreme value distribution. Advances in Water Resources, 176, 104460.
• Vidrio-Sahagún, C. T., Ruschkowski, J., He, J., & Pietroniro, A. (2024). A practice-oriented framework for stationary and nonstationary flood frequency analysis. Environmental Modelling & Software, 173, 105940.
• Villarini, G., Smith, J. A., Serinaldi, F., Bales, J., Bates, P. D., & Krajewski, W. F. (2009). Flood frequency analysis for nonstationary annual peak records in an urban drainage basin. Advances in Water Resources, 32(8), 1255–1266.
• Watt, E. W., Lathem, K. W., Neill, C. R., Richards, T. L., & Rouselle, J. (1989). Hydrology of floods in Canada—A guide to planning and design (p. 245). National Research Council of Canada.
• Wood. (2018). Review of Future Rainfall Scenarios. Town of Oakville.
• Woolway, R. I., Kraemer, B. M., Lenters, J. D., Merchant, C. J., O’Reilly, C. M., & Sharma, S. (2020). Global lake responses to climate change. Nature Reviews Earth & Environment, 1(8), 388–403.

Appendix A: Recommended Resources

Canada:

• Federal Flood Mapping Guidelines Series
  1. Federal Flood Mapping Framework
  2. Federal Guidelines for Flood Hazard Identification and Priority Setting
  3. CSA W229.1:25 Airborne Lidar Data Acquisition Guideline
  4. Case Studies on Climate Change in Flood Mapping
  5. Federal Hydrologic and Hydraulic Procedures for Flood Hazard Delineation
  6. Coastal Flood Hazard Assessment for Risk-Based Analysis on Canada’s Marine Coasts
  7. CSA W229.2:25 Geomatics Guidelines for Flood Mapping
  8. Federal Guidelines for Flood Risk Assessment
  9. Federal Flood Damage Estimation Guidelines for Buildings and Infrastructure
  10. Federal Land Use Guide for Flood Risk Areas
  11. Bibliography of Best Practices and References for Flood Mitigation
• Canadian Climate Data Portal

British Columbia:

• Flood Mapping in BC—Professional Practice Guidelines (PDF, 5.9 MB)
• Legislated Flood Assessments in a Changing Climate in BC—Professional Practice Guidelines (PDF, 3.2 MB).
• Flood Hazard Area Land Use Management Guidelines (PDF 1.92MB)

Newfoundland and Labrador:

• Newfoundland and Labrador Atlas of Storm Surge and Wave Climates
• Technical Guidance for Flood Hazard Mapping Studies

Northwest Territories:

• Northwest Territories Ice Jam Flood Mapping Guideline

Nova Scotia:

• Minimum Planning Requirements in Nova Scotia: Guidebook on Implementing the “Statements of Provincial Interest” (PDF, 2.04MB)
• Municipal Flood Line Mapping (MFLM) Technical Specifications for Flood Mapping in Nova Scotia (PDF, 1.54 MB)

Ontario:

• Adaptive Management of Stream Corridors (PDF, 87.2 KB)
• Guide for Assessment of Hydrologic Effects of Climate Change in Ontario (PDF, 24.8 MB)
• Ontario Climate Data Portal
• Ontario Climate Change Data Portal
• Ontario Flood Hazard Mapping Program Guide (Ministry of Natural Resources) (PDF, 626 KB)

Prince Edward Island:

• Climate Hazard and Risk Information System
• PEI Coastal Property Guide
• PEI Flood Guide: A Resource for Local Land Use Planning