The asymmetry of ENSO is a measure of its nonlinearity, and may be a key ingredient in climate variability on the decadal and longer time-scales, particularly for the Pacific sector. Understanding its causes and ensuring accurate simulation of it by climate models is a key issue facing climate modelers who strive to make reliable forecast/projections of climate changes over the coming decades.
The proposed research attempts to address this issue by analyzing existing model runs as well as through conducting specially designed experiments. Our working hypothesis— is that the nonlinearity in deep convection is an important cause of the asymmetry in ENSO. Specifically, an increase of the nonlinearity of tropical convection will lead to an increase in the asymmetry of zonal wind stress and therefore an increase in the asymmetry of subsurface signal, favoring an increase of ENSO asymmetry.
We plan to analyze the coupled runs as well as the corresponding AMIP runs from the latest NCAR and GFDL models, including the diagnosis of the NCAR model runs with different convection schemes and with different model resolutions and the GFDL model runs with different convection schemes. We not only examine ENSO asymmetry in the surface fields such as SST, surface heat flux, and precipitation but also its subsurface manifestation. To understand how the changed wind stress associated with changed convection scheme will affect the subsurface asymmetry and thereby the SST asymmetry, we will perform forced experiments with the NCAR Pacific basin model, the POP global ocean models (the ocean component of CCSM4/CESM1) and the MOM4 (the ocean component of the GFDL coupled model) using the winds from the AMIP runs of NCAR and GFDL models. We will compare the results from the forced runs driven by observed winds. In addition, the forced runs will be perturbed by warm anomaly, cold anomaly, and the residual of wind stress from observations and model simulations. Experiments especially designed to understand the relative importance of the nonlinearity from the atmosphere and the nonlinearity from the ocean dynamics will also be conducted.
To further validate the effect of changes in convection schemes and model resolutions on the simulation of ENSO asymmetry we will find from NCAR and GFDL models, we also plan to examine the coupled runs as well as the corresponding AMIP runs from the ongoing IPCC AR5 data sets.
The purpose of proposed research is to provide a better understanding of how the simulation of ENSO—the asymmetry between its two phases in particular—in global climate models is affected by increases in model resolution and changes in convection scheme, in support of the development of next-generation climate models involving both higher resolution and improved physical representations.