|Thursday, June 03|
Statistical Estimation of Probable Maximum Precipitation for Small Watersheds
* Daniel Erl, Stantec, Canada
Ammar Taha, Canada
The Probable Maximum Precipitation (PMP) is utilized by hydrological professionals in determining the magnitudes of reasonable extreme floods throughout the world. In Canada it is often used to determine the flood for some important structures as well as in dam studies as an input in determining the Probable Maximum Flood (PMF). The Canadian Dam Association (CDA) requires that two PMPs be utilized when analyzing floods to estimate the PMF. These include the Winter/Spring PMP and the Summer autumn PMP. There are many methods in determining the PMP. Many of these methods require complex meteorological data such as persistent dew points at different geopotential heights to determine maximum precipitable moisture. Much of this data is not available for many locations in Canada. This presentation will focus on the statistical method (Hirschfield Method) as presented in the World Meteorological Organization’s Manual on the Estimation of Probable Precipitation and how to apply it to the readily available historical precipitation data available from Environment and Climate Change Canada for the purpose of estimating the seasonal PMPs. Seasonal separation of the precipitation data and application of the statistical method to the separated data will be discussed. This method of estimating the PMP is suitable for watersheds of less than 1000 km2 . The statistical method provides a useful, quick estimate of PMP without the need for lengthy complex analysis and very specific meteorological data.
Extreme events do not take vacation - A case study of the Christmas 2020 rain-on-snow event in southern Quebec
* Alexis Bédard-Therrien, Université Laval, Canada
Geneviève Beaudoin, Canada
Antoine Thiboult, Canada
Benjamin Bouchard, Canada
François Anctil, Canada
Sylvain Jutras, Canada
Jean-Michel Lemieux, Canada
Charles Malenfant, Canada
Richard Turcotte, Canada
Daniel Nadeau, Canada
Historically, floods resulting from rain-on-snow events have been particularly devastating, but due to the complex nature of the phenomenon, have proven difficult to account for in flood forecasting models. Snow being a heterogeneous and porous media, its properties vary greatly at the basin scale, which complicates the task of accurately predicting the snowmelt runoff. As winters get warmer due to climate change, the frequency of such events should increase as well, which underlies a growing need to analyze them in depth. This study focuses on the exceptional rain-on-snow event that occurred in southern Quebec on 24-25 December 2020. By region, up to 110 mm of rainfall fell, concurrently breaking heat records for this time of year, and causing flooding in many places. The goal of this study is to present a detailed analysis across multiple sites, with a focus on runoff delay and snowpack liquid water storage capacity. This case study is made at a point scale with an emphasis on the vertical flow of water. For this purpose, 6 sites across a 300 km transect in southern Quebec, in varying environments and conditions, have been equipped to measure snowpack water equivalent (2 of which have continuous measurements), snowpack temperature, and precipitation. Of those sites, the open-field Sainte-Anne-de-la-Pérade site (46°34'N, 72°13'W) provides soil water measurements and was also equipped with a disdrometer to assess precipitation phase. Data from experimental site NEIGE, situated in a forest clearing in Montmorency Forest, are also used and provide a longer dataset than at the other sites. Observations of snowpack water equivalent are compared to simulations from Hydrotel’s snow module, used in provincial flood forecasting. The simulated runoff is also compared to soil water measurements to assess its timing and the snowpack’s water storage capacity. The analysis presented aims to provide the detailed behavior of the snowpack during winter rain-on-snow events, and to suggest model modifications for an accurate representation of such events.
Climate change and rainfall IDF (Intensity, Duration, Frequency) curves: overview of science and guidelines for adaptation
* Jean-Luc Martel, École de technologie supérieure, Canada
François Brissette, Canada
Magali Troin, Canada
Philippe Lucas-Piche, Canada
Richard Arsenault, Canada
One of the most important impacts of a future warmer climate is the projected increase in the frequency and intensity of extreme rainfall. This increasing trend in extreme rainfall is seen in both the observational record and climate model projections. However, a thorough review of the recent scientific literature paints a complex picture in which the amplification rainfall extremes depends on a multitude of factors. While some projected rainfall indices follow the Clausius-Clapeyron relationship scaling of ~7 % per 1 °C of warming, there is substantial evidence that some indices follow a larger scaling, even in conditions where the mean annual precipitation is decreasing. This larger scaling can reach up to a Super Clausius-Clapeyron scaling for the most extreme events. This amplification as a function of increasing return period is now well documented at the daily scale, but less clear at the sub-daily scale. In recent years, climate models simulations at a finer spatial and temporal resolution, including convection permitting models, have provided more reliable projection of sub-daily rainfall. Results indicate that the rainfall scaling may also increase as a function of duration, so that for short duration, large return period events will likely see the largest precipitation increases in a warmer climate. This has broad implications on the use of rainfall Intensity-Duration-Frequency (IDF) curves, for which both an increase and a steepening can now be predicted. This paper also presents an overview of measures that have been adopted by various governing bodies to adapt IDF curves to the future climate. Current measures are ranging from multiplying by a simple constant percentage, to increasing the correction factors based on return periods, to scaling them to the Clausius-Clapeyron relationship based on projected temperature increases, and are all currently inadequate or incomplete. None of the adopted measures recognize the likely Super-Clausius-Clapeyron scaling of extreme rainfall, and perhaps more importantly, the increasing scaling towards short duration rainfall and the more extreme rainfall events, which will significantly impact stormwater runoff in cities and in small rural catchments. This presentation will also discuss the remaining scientific gaps and offers technical recommendations for practitioners on how to adapt IDF curves to improve climate resilience.
Flood Forecasting Across Canada: A Fluid Landscape
* David Casson, University of Saskatchewan, Canada
"Flood forecasting is an essential element of disaster risk reduction, needed to protect public safety and infrastructure. Prediction of incoming floods is challenging in practice, requiring a strong scientific, institutional and technological basis. This practice of flood forecasting in Canada is diverse. It is complicated by large domains, complex cryosphere processes such as ice jams and a variety of institutional mandates. Current national, provincial and territorial initiatives aim to improve the practice of flood forecasting. Global Water Futures is advancing scientific best practices for flood prediction, building on the groundwork of FloodNet. The recent 2nd Annual Canadian Flood Forecasting Forum aimed to foster relationships between Environment and Climate Change Canada and its partners towards “National Flood Forecasting Community of Practice”. In this evolving landscape, partnerships and collaboration are essential to bring innovation into practice. This presentation will provide an overview of the current state of flood forecasting, and where new technologies can be harnessed to help to improve predictions. This will include highlights of current practice of from the national, provincial, territorial and municipal scale with innovations in large domain modelling, remote sensing of ice and cold regions process understanding. Outstanding challenges will be framed as questions to the flood forecasting community. "