Multi-decadal Variability of Flood Risk in Southwestern Canadian Prairie Rivers as Characterized by the PDO and ENSO

Abstract:

The 2013 floods across southern Alberta, Canada, are considered to be one of the worst natural disasters in recent Canadian history. This region is highly vulnerable to flooding during spring as the frozen ground restricts infiltration and the melting snow directly contributes to streamflow. Studies have concluded that the 2013 floods in Alberta were a result of heavier snowpack from winter precipitation and higher amounts of spring precipitation as rain over the eastern slopes of the Rockies. Although this flood is considered to be less than the 100-year flood of the region, the effects were economically devastating. The return periods of floods are generally determined under the assumption that the annual peak flow series are independent and identically distributed (i.i.d.). However, researchers have demonstrated that this assumption is not valid in Australia and that the i.i.d. assumption can lead to under- or over-estimation of the true long-term flood risk. The Pacific Decadal Oscillation (PDO) and the El Niño-Southern Oscillation (ENSO) have a strong impact on western Canadian hydroclimate via teleconnections. The negative phase of the PDO and La Niña typically produce heavier snowpack across the prairies compared to that during the positive phase of PDO and El Niño. In this study, we explore the connections between the PDO, ENSO and the peak annual streamflow in southwestern Canadian prairie rivers. Daily averaged annual peak flow records from 22 rivers were stratified according to the PDO phases and ENSO states and fit to the Log-Pearson III (LP3) distribution. We determined that the flood risk is significantly higher in the negative phase of the PDO and is enhanced during La Niña episodes within the negative PDO phase. To ensure these results were not due to sampling error or unequal record lengths, a regional index approach was also employed, which confirmed these results. Our results are important for the optimal planning and design of flood control structures, transportation infrastructure, and water distribution systems, etc.