Weather and Climate Extremes (Dec 2016)
Variability and long-term change in Australian temperature and precipitation extremes
Abstract
Risks from weather and climate extremes to governments, industries and communities are increasing and, at present, are not well quantified. In the presence of climate variability and long-term change, it may not be appropriate to base an assessment of the likelihood of climate hazards on the long-term averages. Many weather and climate extremes have increased in frequency and/or intensity in recent decades with climate model projections showing that several trends are likely to continue. The analysis of historical records is complementary to the use of climate models in understanding the changing nature of extremes. In this study, we apply a consistent methodology to examine the modulation in the probability of extremes in temperature and rainfall as they are influenced by climate change over the past century and natural climate variability and present summary tables and charts for a comprehensive comparison. The daily temperature data – maximum (Tmax) and minimum (Tmin) air temperature – are from a ‘high-quality’ dataset developed at the Australian Bureau of Meteorology – the Australian Climate Observations Reference Network (ACORN) Surface Air Temperature (SAT) dataset. In the absence of a comparable dataset for daily rainfall, we analysed rainfall data at ACORN-SAT locations. Our analyses of extremes are based on annual maxima (annual minima for Tmin) of daily time series from 58 Australian sites over the period 1910–2009. We found statistically significant long-term increases in extreme maximum temperatures but with marked regional and seasonal variations. The increase in the lowest minimum temperature extremes typically exceeds the increase in the extremes of maximum temperature. Daily precipitation extremes rarely exhibit long-term change over the century but are strongly modulated by the El Niño Southern Oscillation (ENSO). The relative importance of long-term change and climate variability therefore depends on the variable or index. We conclude that in assessing the likelihood of climate hazards, one needs to consider the modulation of climate extremes due to both long-term change and climate variability. Our findings imply that when planning for adaptation, different emphasis needs to be given to changing temperature and precipitation extremes.
Keywords
