Date of Award

Spring 1-1-2018

Document Type


Degree Name

Doctor of Philosophy (PhD)

First Advisor

Jennifer E. Kay

Second Advisor

Kristopher B. Karnauskas

Third Advisor

Ben Livneh

Fourth Advisor

Brian Medeiros

Fifth Advisor

Isla R. Simpson


The Southern Ocean, a unique region where clouds, ocean dynamics and sea ice interact to influence climate, has historically been poorly modeled and observed. Here, we improve a global climate model and use newly-available surface-independent cloud observations to establish how Southern Ocean cloud feedbacks impact global climate change. We modify the Community Earth System Model (CESM) by increasing supercooled cloud liquid to better match observations over the Southern Ocean. In the modified model, two extratropical cloud feedbacks cause equilibrium climate sensitivity (ECS, the equilibrium warming in response to doubled CO2) to increase from 4.1 K in the control to 5.6 K. First, reduced conversion of cloud ice to liquid at high southern latitudes decreases the magnitude of a negative cloud phase feedback. Second, warming is amplified in the mid-latitudes by a larger positive shortwave cloud cover feedback. Despite the 1.5 K ECS increase, transient 21st century warming hardly increases in the modified model over the control because ocean heat uptake moves heat input by extratropical cloud feedbacks to depth. Persistent extratropical ocean heat uptake implies that extratropical cloud biases may not be as important to 21st century warming as biases in other regions. Next, we determine how interactions with sea ice impact Southern Ocean cloud feedbacks. We use surface-independent cloud observations to diagnose how present-day sea ice–cloud interactions during spring and summer impact top-of-atmosphere albedo. Observed low cloud cover and opacity are larger over open water compared to over sea ice. The cloud opacity increase is due to an ice-toward-liquid cloud phase shift with no change in air-sea coupling. Even with the cloud response, top-of-atmosphere albedo decreases as sea ice retreats. In CESM, the cloud and albedo responses to sea ice variability are of the same sign but larger in magnitude than the observed responses. The modeled cloud opacity increase is linked to strengthened air-sea coupling rather than a cloud phase shift. Strengthened air-sea coupling with decreasing sea ice could impact model-predicted cloud feedbacks in a way inconsistent with observations. Our results highlight how Southern Ocean shortwave cloud feedbacks influence climate change in the coupled climate system.