BAMOS Vol 31 No.2 June 2018 | Page 10

Article Natural drivers of interannual to decadal variations in surface climate Nicola Maher 1,2 1 Climate Change Research Centre, University of New South Wales, NSW, Australia. 2 ARC Centre of Excellence for Climate System Science, University of New South Wales, NSW, Australia. email: My thesis aims to understand drivers of fluctuations in surface climate that could result in hiatus decades and periods of accelerated warming, specifically focusing on the Indo-Pacific region. I demonstrate the role of the Inter-decadal Pacific Oscillation (IPO) and volcanic eruptions in driving hiatus periods. I further identify a multi-model mean response of the tropical Pacific Ocean to large volcanic eruptions, which may act to further cool the surface ocean. Finally I demonstrate an increase in Indo-Pacific heat content associated with the recent hiatus, which is not reversible even when the IPO returns to its positive phase. 1. Drivers of decadal hiatus periods in the 20th and 21st Centuries While the long term trend in surface air temperature (SAT) is one of warming, there was no significant warming trend in SAT over the period 2001-2014. Decades where the surface of the Earth does not warm significantly are known as hiatus decades. 1 (a) The role of anthropogenic forcing The early 2000s hiatus is not unique in the historical SAT record or within climate model simulations. I used SAT output from a single ensemble member of 31 climate models taking part in Coupled Model Intercomparison Project 5 (CMIP5) (Maher et al., 2014). Here, I considered both the historical (1860-2006) plus two future scenarios RCP4.5 (moderate warming) and RCP8.5 (strong warming), which extend to 2100. Hiatus periods are identified in three categories, (i) those due to volcanic eruptions, (ii) those associated with negative phases of the Interdecadal Pacific Oscillation (IPO), where accelerated warming is associated with the IPO positive phase, and (iii) those affected by anthropogenic aerosols in the mid 20th Century. Figure 1a shows that the likelihood of hiatus periods is sensitive to the rate of change of anthropogenic forcing, when decades influenced by large tropical volcanic eruptions are excluded. While I showed that large volcanic eruptions have dramatically increased the likelihood of hiatus periods in the historical record pre-1980 (77-93% likelihood), the background warming has acted to partially offset the cooling effect of (b) Probability of a hiatus occuring in the future historical (non−volcanic) historical (volcanic) future projections 0.9 10 BAMOS Jun 2018 0.8 RCP4.5 RCP8.5 RCP4.5 + volc (−3.3 W/m −2 ) RCP4.5 + volc (−1.5 W/m −2 ) RCP8.5 + volc (−3.3 W/m −2 ) RCP8.5 + volc (−1.5 W/m −2 ) 0.7 0.6 probability = 50% pro 0.5 0.4 0.3 0.2 0.1 0 0 0.2 0.4 0.6 2 Gradient in anthropogenic forcing (W/m /Year) 2010 2030 2050 Year 2070 2090 Figure 1. a) Probability of a hiatus period versus the gradient in anthropogenic forcing in CMIP5 models, b) probability of a hiatus in two future scenarios, with and without hypothetical volcanic eruptions added in years 2032 and 2087.