How does the Interdecadal Pacific Oscillation affect design floods in Australia?

Abstract

[1] Previous studies have shown significant dependence of annual maximum floods in eastern Australia on the Interdecadal Pacific Oscillation (IPO), a low-frequency mode of natural climatic variability. However, the causative factors behind differences in flood risk observed during contrasting phases of the IPO remain poorly understood. In particular, an important question that stems from a practical flood hydrology perspective is: Does maximum precipitation exhibit a similar level of dependence on the IPO as that observed in the floods? If not, then what are the factors responsible for this disparity, and are there significant ramifications for the conventional approach of estimating design floods through a predefined design rainfall? This paper investigates the possible reasons for the disparate relationship of IPO on rainfall and streamflow design events; that is, why do flood characteristics for contrasting IPO phases differ significantly compared to less notable changes in the corresponding rainfall intensities? We hypothesize that this difference in flood characteristics as a function of the IPO is a result of catchment antecedent or wetness conditions prior to the design storm. This hypothesis is tested using data from 166 high-quality daily rainfall stations across Australia along with catchment-averaged rainfall and resulting flows across an additional 128 catchments from the same region. In addition, 35 subdaily rainfall stations with long records in east Australia were also used to support our arguments. The results of these tests suggest that catchment antecedent conditions prior to design storm events were found to vary significantly across the two IPO phases, leading to the significant differences in flood characteristics across the two phases.

How streamflow has changed across Australia since the 1950s: evidence from the network of hydrologic reference stations

“…They found that the reduction in runoff for the study region is likely to continue under projected future climates. Pui et al (2011) detected changes in annual maximum flood data of 128 stations in NSW according to multiple climate drivers. Ishak et al (2010Ishak et al ( , 2013 presented trend analysis in annual maxima flood series data from 491 stations in Australia, and suggested much of the observed trend may be associated with the climate modes on annual or decadal timescales.…”

Flood triggering in Switzerland: the role of daily to monthly preceding precipitation

“…APIs have been commonly used in hydrology for decades (see e.g., Kohler and Linsley Ray K., 1951;Pui et al, 2011). We follow the method of Baillifard et al (2003):…”

The frequency of major flooding in coastal southeast Australia has significantly increased since the late 19th century

“…Much of the interannual variability is due to the El Niño-Southern Oscillation (Chiew et al 1998, Power et al 1999, Kiem et al 2003, Verdon et al 2004, Kiem et al 2001, 2003, Pui et al 2011, Power and Callaghan 2015. Much of the decadal/multidecadal variability is due to the Interdecadal Pacific Oscillation (IPO; Chiew et al 1998, Power et al 1999, Franks 2002, Kiem et al 2003, Verdon et al 2004, Franks and Kuczera 2002, Henley et al 2015, Micevski et al 2006, Christensen et al 2013, Kiem and Verdon-Kidd 2013, Power and Callaghan 2016.…”