Drought severity estimation under a changing climate

Abstract

This paper describes an approach to estimate drought severity for drought response planning and urban water management considering the impacts of climate change and variability. Low flow frequency analysis was used to estimate drought severity (eg. 1-in-100-year average recurrence interval) of different drought durations from several months to years. Traditionally this was done using available historical streamflow record. However, recent research including the South Eastern Australian Climate Initiative has indicated that the prolonged drought experienced in southeastern Australia including Melbourne in the recent decade since 1997 has been unprecedented in historical context, and is potentially part of a long-term trend associated with global warming. This raises the issue of the practicality of the assumption of hydrologic stationarity. To account for the potential for more severe and frequent drought events, an adaptive approach is needed to adapt to the drier future in a changing and variable climate by considering experience from the recent 1997–2009 drought and the latest climate change projections. In this paper, drought severity for the Melbourne system is estimated based on historical streamflow data with monthly flow prior to 1997 adjusted using a flow duration curve decile method to reflect the recent dry conditions of 1997–2009. The approach is consistent with recommendations in the Victorian Guidelines for the Development of a Water Supply-Demand Strategy (DSE, 2011). The results indicate that drought severity and frequency increased by an order of magnitude under a changed climate based on a “return to dry” scenario. This shift in severity and frequency highlights the need for adaptive planning methods to address changes in hydrologic conditions under a variable and changing climate. However, it also raises many challenges for drought planning including uncertainties in climate change projections, attribution of recent drought to climate change and variability, and the appropriate hydro-climate baseline for applying climate change projections.