Understanding the Role of the Ocean in a Changing Climate through Model Hierarchies

Due to the growing complexity of climate models, model hierarchies are becoming important in understanding climate dynamics. In this dissertation, I build two hierarchies of ocean models to study the ocean's role in a changing climate. In chapter 2, I develop a hierarchy of simplified ocean models for the coupled ocean, atmosphere, and sea ice climate simulations using the Community Earth System Model version 1 (CESM1). The hierarchy has four members: a slab ocean model (SOM), a mixed-layer model (MLM) with entrainment and detrainment, an Ekman mixed-layer model (EMOM), and an ocean general circulation model (OGCM). Flux corrections of heat and salt are applied to the simplified models ensuring that all hierarchy members have the same climatology. I diagnose the needed flux corrections from auxiliary simulations in which I restore the temperature and salinity to the daily climatology obtained from a target CESM1 simulation. The resulting three-dimensional corrections contain the interannual variability fluxes that maintain the correct vertical gradients of temperature and salinity in the tropics. I find that the inclusion of mixed-layer entrainment and Ekman flow produces sea surface temperature and surface air temperature fields whose means and variances are progressively more similar to those produced by the target CESM1 simulation.To illustrate the usefulness of the hierarchy, I revisit the ITCZ problem. The position of ITCZ is controlled by the interhemispheric energy imbalance. The ITCZ problem involves determining mechanisms responsible for reducing the shift of ITCZ in fully coupled climate models compared to those that couple the atmosphere to a simple SOM. In this dissertation, I tackle this problem in the context of the loss of Arctic sea ice. In the absence of dynamic ocean circulation, energy-balance considerations require the ITCZ to shift northward so that the expansion of the southern Hadley cell produces a southward energy transport. This transport eliminates the interhemispheric energy imbalance caused by the reduced northern hemisphere albedo. Two processes present in OGCMs but absent in the SOM have been proposed to explain this difference --- the Atlantic meridional overturning circulation (AMOC) and the wind-driven Ekman flow. Using my hierarchy, I confirm that AMOC strongly inhibits the shift in the position of the ITCZ by producing a strong southward anomalous heat transport in the ocean. This heat transport eliminates the need for the atmosphere to respond to the interhemispheric energy imbalance. I also show that Ekman flow does not damp the ITCZ shift as previously suggested. The explanation for this surprising result is that at the equator, the vanishing of the Coriolis parameter leads to a frictionally driven overturning cell with northward surface flow. This cell produces a northward heat transport that reinforces the interhemispheric energy imbalance, thus contradicting the original hypothesis that a northward shift of the ITCZ would drive a southward surface Ekman flow across the equator. Overall, with my hierarchy I find that it is the change in the AMOC in response to Arctic sea ice loss that is responsible for preventing the ITCZ from shifting northward.In chapter 3, I use my hierarchy to elucidate the differing influences of Ekman and frictional flows on the deep-tropical contraction phenomenon. I study this issue in the context of the equilibrium climate response to an abrupt quadrupling of atmospheric CO2 concentration in CESM1. I find that the atmospheric model coupled to the ocean model that includes the dynamic Ekman flow, i.e., EMOM, can replicate the deep-tropical contraction simulated in comprehensive models. I demonstrate that the differences between atmospheric models coupled to EMOM and OGCM are due to the different penetration depth of the oceanic vertical motion and the differences in the subsurface horizontal diffusivity of temperature near the equator. Furthermore, I demonstrate that the frictional Ekman flow contributes at least 50\% of the increased ocean heat convergence in the equatorial Pacific and Atlantic Oceans. This shows that the frictional flow must be included in any explanation of the deep-tropical contraction.In chapter 4, I develop a hierarchy of intermediate meridional overturning circulation models to explore the zonal asymmetry of freshwater forcing as a stability threshold that controls the existence of multiple equilibria of the overturning circulation. I developed two new models to build a hierarchy. The first one is the zonally averaged two-slabs ocean model (ZATOM) that explicitly resolves zonal buoyancy gradients. The second one is an extension of the Stommel two-box model to include the effect zonal asymmetry of freshwater forcing. I found that the zonal asymmetry of freshwater forcing strongly determines the regimes with the multiple equilibria of the overturning circulation. I show that changing the strength of hydrological forcing has a different effect on the overturning circulation depending on zonal asymmetry of freshwater forcing. Thus, the zonal asymmetry of freshwater forcing controls the subsequent strengthening or weakening of the overturning circulation. I also use this result to explain the emergence of multiple equilibria in an oceanic general circulation model in Dijkstra and Weijer (2003).In conclusion, I have built two hierarchies of ocean models to study the ocean's role in a changing climate. I use these hierarchies to advance our understanding of the tropical air-sea coupling and the control of zonal asymmetry of freshwater forcing on the existence of AMOC's multiple equilibria in the context of climate change