Search

Increased flash drought awareness in recent years has motivated the development of numerous indicators for monitoring, early warning, and assessment. The flash drought indicators can act as a complementary set of tools by which to inform flash drought response and management. However, the limitations of each indicator much be measured and communicated between research and practitioners to ensure effectiveness. The limitations of any flash drought indicator are better understood and overcome through assessment of indicator sensitivity and consistency; however, such assessment cannot assume any single indicator properly represents the flash drought "truth." To better understand the current state of flash drought monitoring, this study presents an intercomparison of nine, widely used flash drought indicators. The indicators represent perspectives and processes that are known to drive flash drought, including evapotranspiration and evaporative demand, precipitation, and soil moisture. We find no single flash drought indicator consistently outperforms all others across the contiguous United States. We do find the evaporative demand- and evapotranspiration-driven indicators tend to lead precipitation- and soil moisture-based indicators in flash drought onset, but also tend to produce more flash drought events collectively. Overall, the regional and definition-specific variability in results supports the argument for a multi-indicator approach for flash drought monitoring, as advocated by recent studies. Furthermore, flash drought research-especially evaluation of historical and potential future changes in flash drought characteristics-should test multiple indicators, datasets, and methods for representing flash drought, and ideally employ a multi-indicator analysis framework over use of a single indicator from which to infer all flash drought information. SIGNIFICANCE STATEMENT: Rapid onset or "flash" drought has been an increasing concern globally, with quickly intensifying impacts to agriculture, ecosystems, and water resources. Many tools and indicators have been developed to monitor and provide early warning for flash drought, ideally resulting in more time for effective mitigation and reduced impacts. However, there remains no widely accepted single method for defining, monitoring, and measuring flash drought, which means most indicators that are developed are compared with other individual indicators or conditions and impacts in one or two flash drought events. In this study, we measure the state of flash drought monitoring through an intercomparison of nine, widely used flash drought indicators that represent different aspects of flash drought. We find that no single flash drought indicator outperformed all others and suggest that a comprehensive flash drought monitor should leverage multiple, complementary indicators, datasets, and methods. Furthermore, we suggest flash drought research-especially that which reflects on historical or projected changes in flash drought characteristics-should seek multiple indicators, datasets, and methods for analyses, thereby reducing the potentially confounding effects of sensitivity to a single indicator. [ABSTRACT FROM AUTHOR]

The annual Australian monsoon pattern includes an onset, or the much anticipated first active monsoon period of the season, but defining the monsoon onset has proven to be problematic. Since the first Australian monsoon onset definition by Troup in 1961 there have been many others presented. There appears to be no universally accepted method to define the Australian monsoon onset, and therefore, we present here an analysis of the methods that have been proposed. The aim of this paper is to systematically review the different methods used to define the Australian monsoon onset, adding to the work that has been done by other reviews for monsoon systems around the world. For the first time, we identify the 25 different methods that have been published for the Australian monsoon/wet season onset and compare them to identify how well they align. When considering the 57 seasons where more than one onset definition is provided, the range of dates within the season can range over several months, with an average range of 44 days and the largest range within a season of 78 days. Thus, we show that different onset definitions are capturing different events altogether and pin the 'onset' to different dates throughout the progression of the north Australian wet season. Some capture a 'wet season onset' while others capture the dynamical overturning of the atmosphere (i.e. the monsoon). In conclusion, our analysis finds that there is still a lack in real-time monitoring or prognostic capabilities of monsoon onset dates as well as limited operational applicability despite a plethora of definitions. [ABSTRACT FROM AUTHOR]