European Union FP7, Royal Society, and National Science Foundation.

Background Europe has emerged as a major climate change hotspot, both in terms of an increase in seasonal averages and climate extremes. Projections of temperature-attributable mortality, however, have not been comprehensively reported for an extensive part of the continent. Therefore, we aim to estimate the future effect of climate change on temperature-attributable mortality across Europe. MethodsWe did a time series analysis study. We derived temperature-mortality associations by collecting daily temperature and all-cause mortality records of both urban and rural areas for the observational period between 1998 and 2012 from 147 regions in 16 European countries. We estimated the location-specific temperature-mortality relationships by using standard time series quasi-Poisson regression in conjunction with a distributed lag non-linear model. These associations were used to transform the daily temperature simulations from the climate models in the historical period and scenario period (2006-2099) into projections of temperature-attributable mortality. We combined the resulting risk functions with daily time series of future temperatures simulated by four climate models (ie, GFDL-ESM2M, HadGEM2-ES, IPSL-CM5A-LR, and MIROC5) under three greenhouse gas emission scenarios (ie, Representative Concentration Pathway [RCP]2.6, RCP6.0, and RCP8.5), providing projections of future mortality attributable fraction due to moderate and extreme cold and heat temperatures.Findings Overall, 7•17% (95% CI 5•81-8•50) of deaths registered in the observational period were attributed to nonoptimal temperatures, cold being more harmful than heat by a factor of ten (6•51% [95% CI 5•14-7•80] vs 0•65% [0•40-0•89]), and with large regional differences across countries-eg, ranging from 4•85% (95% CI 3•75-6•00) in Germany to 9•87% (8•53-11•19) in Italy. The projection of temperature anomalies by RCP scenario depicts a progressive increase in temperatures, more exacerbated in the high-emission scenario RCP8.5 (4•54°C by 2070-2099) than in RCP6.0 (2•89°C) and RCP2.6 (1•67°C). This increase in temperatures was transformed into attributable fraction. Projections consistently indicated that the increase in heat attributable fraction will start to exceed the reduction of cold attributable fraction in the second half of the 21st century, especially in the Mediterranean and in the higher emission scenarios. The comparison between scenarios highlighted the important role of mitigation, given that the total attributable fraction will only remain stable in RCP2.6, whereas the total attributable fraction will rapidly start to increase in RCP6.0 by the end of the century and in RCP8.5 already by the middle of the century. InterpretationThe increase in heat attributable fraction will start to exceed the reduction of cold attributable fraction in the second half of the 21st century. This finding highlights the importance of implementing mitigation policies. These measures would be especially beneficial in the Mediterranean, where the high vulnerability to...

No abstract

Despite steady progress in the understanding of El Niño–Southern Oscillation (ENSO) in the past decades, questions remain on the exact mechanisms explaining the heat buildup leading to the onset of El Niño (EN) events. Here we use an ensemble of ocean and atmosphere assimilation products to identify mechanisms that are consistently identified by all the data sets and that contribute to the heat buildup in the western Pacific 18 to 24 months before the onset of EN events. Meridional and eastward heat advection due to equatorward subsurface mass convergence and transport along the equatorial undercurrent are found to contribute to the subsurface warming at 170°E–150°W. In the warm pool, instead, surface horizontal convergence and downwelling motion have a leading role in subsurface warming. The picture emerging from our results highlights a sharp dynamical transition at 170°E near the level of the thermocline.