The 2015/2016 El Niño, which was generated with a smaller westerly wind burst in 2015 than in 1997, but had a similar strength with the famous extreme 1997/1998 El Niño, has attracted wide attention in the scientific community. Especially, it was following a failed 2014 El Niño event, which was considered to be a strong El Niño under strong westerly wind burst at the beginning of 2014. Study of the formation mechanisms of 2015/2016 El Niño is critical to advance our scientific understanding and enhance our prediction capability of extreme El Niño. Although the previous study has addressed the role of heat accumulation during failed El Niño in 2014 for the extreme 2015/2016 El Niño, they explained the heat accumulation by mainly used the recharge oscillator theory for the whole equatorial Pacific region, and was unable to show the detailed variation and contribution of three-dimensional advection and surface heat flux in different divided physically-relevant regions. The detailed spatiotemporally variable physics and their quantitative assessments along the equatorial Pacific for the formation of 2015/2016 El Niño still remains largely unclear.

Based on analyses of heat balance in the three divided regions along the equatorial Pacific Ocean in different El Niño development phases, and on the contrasting comparison with the extreme 1997/1998 El Niño, we identify the physical processes invoked in the formation of 2015/2016 El Niño. We found that the vertical temperature advection (WADV) played an important role for heat distribution in the western equatorial Pacific. The central and eastern equatorial Pacific were dominated by both zonal temperature advection (UADV) and vertical temperature advection (WADV) in all the three phases, but their relative significance varied in different phases. The effect of meridional temperature advection (VADV) was relatively weak for all the three divided regions. The WADV was determined by the variation of horizontal current convergence ∂u/∂x(UDIV) and ∂v/∂y(VDIV), and the importance of these two components were different in distinct regions and phases. The heat accumulation during the prior-El Niño phase was determined by the unique zonal wind variation in the central equatorial Pacific during 2014, which was absent in 1997/1998 El Niño. This prior-El Niño heat accumulation helped to compensate the heat loss around January 2015 and maintain sufficient heat content in the eastern equatorial Pacific for El Niño. We found that the favorable coupled atmosphere-ocean conditions for the heat accumulation in the central equatorial Pacific during the prior-El Niño phase provided important physical hub to nourish the extreme 2015/2016 El Niño, because this accumulated heat could provide 17% heat content for the whole heat increasing in the central and eastern equatorial Pacific during the development phase. This study identified the unique characteristics of the 2015/2016 El Niño and quantified the associated heat distribution, accumulation and transport over the equatorial Pacific in response to the coupled atmosphere-ocean system. This study suggested that the concerned spatiotemporal scales in the coupled atmosphere-ocean processes for El Niño may be much larger than we previously thought.