Francis Zwiers

University of Victoria, Pacific Climate Impacts Consortium

“Probable maximum precipitation estimation and uncertainty quantification using bivariate extreme value analysis”

Date: Thursday, October 19, 2017

Probable Maximum Precipitation (PMP), and Probable Maximum Flood (PMF), which is derived from PMP, are important parameters for dam safety and civil engineering purposes. Operational procedures for calculating PMP seek to obtain a plausible upper bound for precipitation that is unlikely to be exceeded. These calculations do not produce accompanying uncertainty estimates. We propose a probabilistic framework for the interpretation of PMP when it is calculated via the so-called moisture maximization approach, which is the approach that is used most frequently operationally. The proposed framework, which is based on a bivariate extreme value distribution, (i) enables assessment of the sensitivity of the PMP value to individual, possibly erroneous, observations; (ii) allows evaluation of the uncertainty of PMP values; and (iii) provides clarification of the impact of the assumption that a PMP event occurs under conditions of maximum moisture availability.

The frame work is used to estimate PMP across North America based on a 50-year CanRCM4 regional climate model simulation of the period 1951-2000. Results suggest that operational PMP estimates obtained via moisture maximization are highly uncertain and indicate that the assumption that PMP events have maximum moisture availability may not be valid. Specifically, in the climate simulated by CanRCM4, the operational approach applied to 50-year data records produces a value that is 15% larger on average over North America than the maximum that is likely to be observed over such a period, which may have serious implications in engineering design.

This is joint work with M.A. Ben Alaya and X. Zhang.


Congratulations to our October, 2018 graduates: Jenna Tichon (Ph.D.), Kanika Grover (M.Sc.), Mohammed Kamso (M.Sc.), Inesh Munaweera Arachchilage (M.Sc.), Joynob Siddiqua (M.Sc.), Shamsia Sobhan (M.Sc.).

Important Date

October 3: Fee Payment Deadline Fall Term

Upcoming Seminars

Statistics seminar: Bo Li: “Spatially Varying Autoregressive Models for Prediction of New HIV Diagnoses” — Thursday, September 20 at 2:45 p.m., P230 Duff Roblin.

Statistics seminar: Mohammad Jafari Jozani: “Quantile Regression with Nominated Samples: An example in Osteoporosis Analysis” — Thursday, September 27 at 2:45 a.m., P230 Duff Roblin.

Statistics seminar: Yuliya Martsynyuk — Thursday, October 4 at 2:45 a.m., P230 Duff Roblin.

Statistics seminar: Darja Barr: “Evidence Driven Best Practices for Teaching the Mathematical Sciences” — Thursday, October 11 at 2:45 a.m., P230 Duff Roblin.

Where are they now?

Taraneh Abarin, Ph.D. (2009)

Yanqing Yi, Ph.D. (2008)