Oral Presentation AMOS Annual Meeting and International Conference on Tropical Meteorology and Oceanography

The severity of multi-year droughts in palaeoclimate reconstructions and climate models (#150)

Benjamin J Henley 1 , Murray Peel 1 , Rory Nathan 1 , Mandy Freund 1 , Ailie Gallant 2
  1. University of Melbourne, Parkville, VIC, Australia
  2. Earth Atmosphere and Environment, Monash University, Clayton, VIC, Australia

The likelihood of long, severe droughts has not been comprehensively assessed in southeastern Australia. This has greatly limited our capacity to prepare for and adapt to future drought and its impacts. Contextualising observed hydrological extremes, such as the Millennium Drought, using palaeoclimate reconstructions provides water planners with extended estimates of the plausible range of climate variability that is unavailable from instrumental records. With Australia’s history of vulnerability to water scarcity and the high costs of water supply augmentation measures, there is an urgent and practical need to accurately constrain estimates of hydroclimatic variability and quantify the likelihood of long, severe droughts.

 

This study uses a network of high-resolution multi-proxy palaeoclimate records to reconstruct inflows into two key water supply reservoirs in southeastern Australia. Our new reconstructions dramatically augment the observed record and allow us to examine in detail the severity and duration of hydrological droughts over the last 400 years. We find that severe interannual to decadal-scale droughts are a common feature in the region prior to the instrumental period. However, the Millennium Drought is particularly unusual in a multi-century context with regards to its exceptionally long duration and its association with a multidecadal negative trend in reservoir inflows. The likelihood of long, severe droughts is assessed in an ensemble of climate model experiments from the CMIP5 and PMIP3 archive. In the context of some model deficiencies in low frequency hydrological variability and model-projected declines in cool season rainfall in large parts of southern Australia, our palaeoclimate reconstructions reveal the extraordinary nature of surface runoff deficits in recent decades – an observation that is not possible with the instrumental record alone.