Aerosol–cloud interactions over the Southern Ocean

• Main Author: Hartery, Sean
• Format: Article in Journal/Newspaper
• Language: English
• Published: University of Canterbury 2021
• Subjects: New Zealand; Ross Sea; Southern Ocean
• Online Access: https://dx.doi.org/10.26021/10766; https://ir.canterbury.ac.nz/handle/10092/101713

In this work, connections between the abundance of sub-micrometer particulate within the marine boundary layer and optical properties of low-level marine stratus over the Southern Ocean are explored. Global climate models (GCMs) currently predict that much more shortwave radiation is entering the Earth system over the Southern Ocean than satellite- borne radiometers observe. This undermines long-term climate projections in the wider region, leading to greater uncertainty about our climate future. The central hypothesis of this thesis is thus: as the Southern Ocean is a region with near-total cloud cover, and as clouds are opaque to shortwave radiation, and as the abundance of boundary layer particulate available to nascent clouds is known to strongly influence the opacity of those clouds, then radiative biases within GCMs are fundamentally related to the abundance of boundary layer particulate. The goals of this thesis are thus two-fold. First, to quantify whether GCMs accurately represent the abundance and types of particles present within the Southern Ocean boundary layer; and if not, use new observations to constrain existing parameterizations predicting their generation. Second, to quantify how sensitive Southern Ocean clouds are to variations in the abundance of such particles. Central to this examination is a new record of measurements collected within the Southern Ocean boundary layer. In February of 2018, scientific instruments were installed on the R/V Tangaroa for a voyage to the Ross Sea. The voyage departed and returned to Wellington, New Zealand, providing 40 days of continuous in situ observations within the marine boundary layer. This included measurements of the abundance of suspended particulate by an optical particle counter, a differential mobility analyzer and a cloud condensation nuclei counter; discrete samples of ambient particulate collected on filters; attenuated back-scattered light from the boundary layer measured by a ceilometer, and several meteorological variables measured by radiosondes and an automated weather station. Measurements collected by the optical particle counter throughout this voyage were used to estimate the abundance of sea spray particles within the boundary layer. These particles were prevalent throughout the observational record. As they are highly soluble in water, these particles are ideal cloud condensation nuclei and form cloud droplets when the air sat- urates with water vapour during cloud formation. However, as the particles themselves are formed at the sea surface from crashing waves, they also contain insoluble organic detritus from bacteria which can act as nucleation points for ice. Since waves break as a result of the continual stress the wind exerts on the ocean surface, the generation of sea spray particles is primarily a function of near-surface wind speed. As the Southern Ocean is one of the windiest oceans on Earth, the measurements recorded throughout the 2018 voyage on the R/V Tangaroa provide a valuable data set to test the extremes of parameterizations for the flux of such particles. The results of this study indicate that existing parameterizations pro- duce too many sea spray particles at all wind speeds, leading to significant biases. Based on the observational record, an existing parameterization for the flux of sea spray particles is constrained. This leads to an improved representation of sea spray particles within GCMs. While this initial study provided a better characterization of the number of particles that may be available to nascent Southern Ocean cloud, low-level clouds can decouple from the boundary layer, which limits the number of particles that would normally be available to a new cloud. In a second study, in situ observations of the abundance of particle surface area are correlated to measurements of the total attenuated backscatter measured by a ceilometer to better understand the conditions in which low-level clouds have access to the reservoir of particulate available in the boundary layer. This study finds that the strong winds over the Southern Ocean provided sufficient turbulent kinetic energy to evenly mix particulate from the sea surface to cloud base, 80% of the time. As a result, in situ measurements of boundary layer particles measured near sea-level provide valuable information about the number of cloud condensation and ice nuclei available to cloud. Finally, the sensitivity of low-level Southern Ocean cloud to variations in the abundance of sea spray particles is explored. This analysis leveraged recent model results from the Unified Model, a well-known GCM, which showed that the amount of shortwave radiation reflected by a low-level Southern Ocean cloud is very sensitive to the number of ice-nucleating particles available to it. This study demonstrated that a substantial fraction of the shortwave radiation bias over the Southern Ocean could be explained by improved representation of primary ice nucleation within mixed-phase clouds. These results are distilled into a simple parameterization for the amount of primary ice formed when a cloud freezes, which should be appropriate in climate models with a single-moment representation of cloud phase. Thus, while many other physical phenomena influence the abundance and optical properties of Southern Ocean cloud, this thesis establishes that the abundance of boundary layer particulate is a fundamental physical quantity which governs the extraterrestrial input of energy into the Earth system. As such, future iterations of GCMs should give more careful attention to how these particles are generated within the simulated environment and how they interact with clouds. In this regard, the parameterizations developed within this thesis should provide a fruitful starting point.