The mechanisms underlying the tropical Pacific surface warming

– A keynote by Sarah Kang

In the spirit of furthering collaboration and exchange, the program of the 4th km-scale hackathon hosted at the Max-Planck-Institute for Meteorology (MPI-M) earlier this year centered on intensifying interactions within and between different working groups. To expand upon these critical periods of mutual exchange, several keynote speeches were presented on overarching topics that string together the diverse themes convoluted in nextGEMS, WarmWorld, and EERIE. On Wednesday – the halfway point of the hackathon – the MPI-M’s newest member on the board of directors, Sarah Kang, gave an insightful talk, providing an outlook on her prospective work and its links to ongoing research at the institute.

From the outset, Sarah Kang noted that the research underpinning her talk, “Possible shifting mechanism for tropical Pacific surface warming pattern”, was motivated by a “passion for understanding why things do what they do and how they work”. With this indicative motto in mind, Kang’s presentation indexed the central problem tackled by her research: the discrepancy between model and observed sea surface temperatures in the Southern Pacific Ocean (SPO). This divergence was highlighted by contrasting simulations based on observational records with different model projections from 1979-2014. However, since the large observed trends cannot be explained by natural variability alone, there may be an external factor influencing the temperature of the SPO that is presently not captured by the applied models – What are the mechanisms underlying this forced response?

The real difficulty in identifying such mechanisms lies in isolating them from the many convoluted processes that induce warming and cooling patterns over the greater Pacific area. One factor impacting the cooling in the South Pacific is the relative warming trend in the Indian and Atlantic Oceans. For the former, this teleconnection can be explained by a strengthening of the so-called trade winds, the prevailing easterly winds at the equator, caused by a warming of the Indian Ocean. Another process tthought to influence the South Pacific cooling is a decrease in sea surface temperature (SST) in the Southern Ocean (SO). Here, increasing amounts of Antarctic meltwater impact the internal variability associated with deep ocean convection, thereby amplify a milder cooling effect associated natural variability. This latter point may offer an inroad to study the model-observation discrepancy as commonly used GCMs do not represent the SO cooling.

The experimental setup to test these links between the SO and the Southern Pacific is based on this insight. By using historical simulations from CMIP5 and CMIP6 runs, and comparing them to simulation scenarios adapted by Sarah Kang and her collaborators, the team could discern a substantial difference in outcome. In the latter simulations runs, termed Southern Ocean Pacemaker (SOPACE) simulations, historical radiative forcing was included and the sea surface temperature anomalies poleward of 40°S were restored to observations from 1970-2014. While global warming trends again dominated historical simulations, the SOPACE runs clearly linked the SO SST declines to the Southern Pacific cooling. 

Although the experiments have detected SO SST as an external forcing of Southern Pacific cooling, trends in multi-decadal cooling are expected to be transient features that eventually subside into the dominant global warming trend. However, open questions remain attached to this line of research, particularly about the role of mesoscale processes in modulating the mechanisms at fast time scales. Therefore, one of the next steps in bringing this research to the MPI-M will be to adopt the globally coupled version of ICON (5km ocean and 10km atmosphere) in experimental setups.