Evolution and forcing mechanisms of ENSO over the last 300,000 years in CCSM3

• Main Authors: Lu, Zhengyao, Liu, Zhengyu, Chen, Guangshan, Guan, Jian
• Format: Text
• Language: English
• Published: 2018
• Subjects: geo; envir; Pacific; Ice Sheet
• Online Access: https://doi.org/10.5194/cp-2016-128; https://cp.copernicus.org/preprints/cp-2016-128/

The responses of El Niño-Southern Oscillation (ENSO) and the equatorial Pacific annual cycle to external forcing changes are studied in three 3,000-year-long NCAR-CCSM3 model simulations. The simulations represent the period from 300 thousand years before present (ka BP) to present day. The first idealized simulation is forced only with accelerated orbital variations, and the rest are conducted more realistically by further adding on the time-varying boundary conditions of greenhouse gases (GHGs) and continental ice sheets. It is found that orbital forcing dominates slow ENSO evolution, while the effects of GHGs and ice-sheet forcing tend to compensate each other. On the orbital time scales, ENSO variability and annual cycle amplitude change in-phase and both have pronounced precessional cycles (~ 21,000 years) modulated by variations of eccentricity. Orbital forced ENSO intensity is dominated linearly by the change of the coupled ocean-atmosphere instability, notably the Ekman upwelling feedback and the thermocline feedback; and is also possibly affected during ENSO intrinsic developing season by the remote (or extratropical) influences of the short-scale stochastic weather noises. The acceleration technique is found to dampen the precessional signal in ENSO intensity. In glacial-interglacial cycles, additionally, the weakening/strengthening of ENSO owning to a more concentrated/depleted GHGs level leaves little net signal as compensated by the effect coherent change of decaying/expanding ice sheets. They influence the ENSO variability through changes in annual cycle amplitude via a common nonlinear frequency entrainment mechanism while the GHGs effect might has an additional linear part.