Development of a methodology to evaluate probable maximum precipitation (PMP) under changing climate conditions: Application to southern Quebec, Canada Alain N. Rousseau a,⇑ , Iris M. Klein a , Daphné Freudiger a,b , Patrick Gagnon a , Anne Frigon c , Claudie Ratté-Fortin a a INRS-ETE, 490 rue de la Couronne, Québec, QC G1K 9A9, Canada b ETH Zurich, Department of Civil, Environmental and Geomatic Engineering (D-BAUG), Swiss Federal Institute of Technology, Zurich, Switzerland c Ouranos Consortium on Regional Climate and Adaptation to Climate Change, Montreal, Canada article info Article history: Received 11 July 2014 Received in revised form 17 October 2014 Accepted 21 October 2014 Available online 29 October 2014 This manuscript was handled by Andras Bardossy, Editor-in-Chief, with the assistance of Niko Verhoest, Associate Editor Keywords: Probable maximum precipitation (PMP) Probable maximum flood (PMF) Climate change (CC) Canadian Regional Climate Model (CRCM) Precipitable water Moisture maximization summary Climate change (CC) needs to be accounted for in the estimation of probable maximum floods (PMFs). However, there does not exist a unique way to estimate PMFs and, furthermore the challenge in estimat- ing them is that they should neither be underestimated for safety reasons nor overestimated for econom- ical ones. By estimating PMFs without accounting for CC, the risk of underestimation could be high for Quebec, Canada, since future climate simulations indicate that in all likelihood extreme precipitation events will intensify. In this paper, simulation outputs from the Canadian Regional Climate Model (CRCM) are used to develop a methodology to estimate probable maximum precipitations (PMPs) while account- ing for changing climate conditions for the southern region of the Province of Quebec, Canada. The Kéno- gami and Yamaska watersheds are herein of particular interest, since dam failures could lead to major downstream impacts. Precipitable water (w) represents one of the key variables in the estimation process of PMPs. Results of stationary tests indicate that CC will not only affect precipitation and temperature but also the monthly maximum precipitable water, w max , and the ensuing maximization ratio used for the estimation of PMPs. An up-to-date computational method is developed to maximize w using a non-sta- tionary frequency analysis, and then calculate the maximization ratios. The ratios estimated this way are deemed reliable since they rarely exceed threshold values set for Quebec, and, therefore, provide consis- tent PMP estimates. The results show an overall significant increase of the PMPs throughout the current century compared to the recent past. Ó 2014 Elsevier B.V. All rights reserved. 1. Introduction High-risk dams must be able to withstand earthquakes and floods, and therefore the use of an appropriate design flood is required. Design floods can be determined using either a probabi- listic approach (flood estimation based on a given return period) or a deterministic approach (probable maximum flood, PMF; Chow et al., 1988; Debs et al., 1999; England, 2011). The PMF is defined as the flood that would occur if the most severe probable meteoro- logical conditions were to be observed at a specific place, at a spe- cific time (Chow et al., 1988). In general, PMFs are assessed by specialized consultants, government departments or agencies using a hydrological model of a watershed system that uses as input the probable maximum storm (PMS), the time distribution of the PMP, which is the focus of this paper. Major floods occurring in the present time are often described as 1000 year or 100,000 year floods by media; and have mainly public-relations significance. Contrary to that, the PMF is a clearly defined scientific term, and also is of real scientific interest. In northern regions like the Province of Quebec, Canada, PMFs can either be induced by a combination of spring runoff due to extreme snow cover, melt conditions and rainfall occurring simultaneously, or by summer-fall runoff events produced by an http://dx.doi.org/10.1016/j.jhydrol.2014.10.053 0022-1694/Ó 2014 Elsevier B.V. All rights reserved. Abbreviations: CC, Climate change; CRCM, Canadian Regional Climate Model; GEV, Generalized Extreme Value; PMF, probable maximum flood; PMP, probable maximum precipitation; PMSA, probable maximum snow accumulation; r, maxi- mization ratio; RCM, Regional Climate Model; SWE, snow water equivalent; w, precipitable water; w event , precipitable water corresponding to a certain event; w max , monthly maximum precipitable water; w 100 , 100-year return value of the monthly maximum precipitable water; WMO, World Meteorological Organization. ⇑ Corresponding author. Tel.: +1 418 654 2621; fax: +1 418 654 2600. E-mail address: alain.rousseau@ete.inrs.ca (A.N. Rousseau). Journal of Hydrology 519 (2014) 3094–3109 Contents lists available at ScienceDirect Journal of Hydrology journal homepage: www.elsevier.com/locate/jhydrol