Reducing uncertainty in the effect of shipping on climate change
Background
The influence of aerosol particles (e.g., emitted by transport and energy generation) on Earth’s energy balance is a major uncertainty in climate science. Aerosols mask warming from greenhouse gases and so aerosol uncertainty limits our understanding of the influence of greenhouses gases in the past, which affects how precisely we can simulate the future. The uncertainty also has implications for geoengineering schemes that seek to modify the climate through the deliberate release of aerosols.
Shipping is a major source of aerosols, but in 2020 legislation limited aerosol emissions from shipping. A record temperature rise followed in 2023 raising the question of whether the shipping regulations were partly responsible and bringing the aerosol impacts from shipping into the spotlight. The effect of shipping aerosols on clouds is considered the largest contributor to shipping-climate interactions and aerosol-cloud interactions are one of the largest sources of climate uncertainty.
PhD Opportunity
This project will reduce uncertainty in shipping aerosol climate effects using data science/machine learning (ML), modelling and observations with the student playing a leading role in steering the research questions and methods. Uncertainties in the representation of shipping aerosol in models and its interaction with clouds could be explored using a Perturbed Parameter Ensemble (PPE) ML method to characterize the effect of varying model parameter settings and functions/relationships. This tool has been used extensively at the U. Leeds. It quantifies the feasible range of aerosol impacts (i.e., model uncertainty) and the importance of the different input parameters and functions, allowing the targeting of specific processes for improvement. Observations (satellite and field campaign data) will play a vital role in both improving model process relationships and ruling out unrealistic model variants (constraining the PPEs) and therefore reducing the model uncertainty. Process relationships between clouds and aerosols have been characterized at U. Leeds using other ML methods applied to satellite observations to isolate the influence of aerosol within noisy data; these relationships could be used to constrain the PPEs. This approach could be expanded and applied to shipping; e.g., by incorporating shipping emissions from real ship data and determining its influence upon clouds.
Modelling tools bridging scales from tens of metres to the ~100km resolution of global climate models are available to investigate challenges such as determining the importance of multiple local individual ship tracks vs the wider-scale effect of aerosols from ships once it is diluted and spread over a larger area further downwind.
As well as reducing uncertainty in shipping-aerosol interactions and hence in the recent accelerated warming, the project will reduce uncertainty in aerosol-cloud interacations in general, which will feed through to reduced uncertainty in climate projections.
Applicant Profile
Students with a background in physics, maths or computer/data science with coding skills who want to apply their skills to atmospheric or climate science are particularly encouraged.
Other information
https://www.nature.com/articles/s43247-024-01442-3
https://environment.leeds.ac.uk/homepage/176/atmospheric-chemistry-and-aerosols
https://www.youtube.com/watch?v=hg6tSM4BoHg
