Integrative Frequency Analysis

The developed model will allow for the estimation of flood levels corresponding to fixed levels of risk, which will be used by the DEH to draw up a new map of flood zones in southern Québec.

Project details
Scientific program
2020-2025 programming
Theme(s) and priority(s)
Extreme Events
Start and duration
June 2020 • 21 months
Project Status
In closing
Linked project
Support for INFO-Crue
 
Principal(s) investigator(s)
Jonathan Jalbert
Polytechnique Montréal

Context

Expected climate changes in southern Québec include a decrease in winter snow cover and an increase in spring precipitation, which have opposing effects on spring floods. The intensity of the resulting changes varies with time and latitude. Furthermore, historic flow observations are rare: flow rates are not measured for the vast majority of rivers.

This project consists of developing a methodology for estimating expected floods over the coming decades that will integrate uncertainty related to the use of pseudo-observations in the absence of flow measurements and uncertainty related to the use of climate simulations, hence the title Integrative Frequency Analysis. These flood estimates will be used to draw up new flood maps for Québec.

This project is part of the INFO-Crue initiative set up by the MELCC.

Objective(s)

  • Develop a non-stationary frequency analysis method to estimate the level of future floods on gauged rivers (with observations) and non-gauged rivers (with pseudo-observations provided by another INFO-Crue project);

  • Implement the developed method on 18,000 reaches provided by the Direction de l’expertise hydrique du Québec (DEH) to estimate flood levels occurring on average once every 100 years, once every 350 years and once every 400 years over the next 50 years;

  • Develop a method to approximate water levels in confluence zones.

Methodology

  • Develop a frequency analysis model for pseudo-observations that integrates their uncertainty;

  • Develop a non-stationary frequency analysis model for simulated flows from a large set of climate simulations;

  • Assemble the two models to integrate the sources of uncertainty;

  • Utilize the model to estimate flood levels.

Expected results

Once completed, the project will have led to the development of a frequency analysis method integrating various sources of uncertainty and non-stationary behaviour, and to the estimation of flood levels occurring on average once every 100 years, once every 350 years and once every 400 years over the next 50 years for the 18,000 reaches studied.

Benefits for adaptation

Benefits for adaptation

The developed model will allow for the estimation of flood levels corresponding to fixed levels of risk, which will be used by the DEH to draw up a new map of flood zones in southern Québec.

However, the methodological developments resulting from this project could also be applied to a broader context than river flow. The model could apply to the calculation of return levels for any environmental variable in a changing climate with multiple sources of uncertainty. The results therefore have a large number of potential users.

Funding

This project was funded by Ouranos as part of the support for the MELCC's INFO-Crue initiative

Related projects

button back to top