Bright clouds and artificial intelligence help illuminate climate science

25 October 2022

Ship tracks are the clouds that form around ship exhaust plumes in still air. As ships travel they emit pollution particles into the atmosphere, such as 13% of the global emissions of sulphur dioxide, that can act as seeding for clouds. Water condenses around these ‘seeds’ creating long, narrow, vivid clouds that follow the path taken by the vessel.

These bright cloud lines can be seen in satellite imagery and they are usually made up of smaller water droplets, causing increased light scattering and therefore, making the tracks look brighter and more reflective than standard cloud formations over sea.

Ship tracks reflect sunlight helping to cool the planet, offsetting some of the warming effects of greenhouse gases. However, the magnitude of this effect is the largest uncertainty in determining changes in the Earth’s radiative balance caused by human activity.

By studying ship tracks scientists are able to gain a better understanding of the impact of sulphate aerosols and the radiative effect of clouds in remote locations, far away from other human influences. However, over the years the clouds have been arduous to catalogue and typically studied only in particular regions for short periods of time.

This study used state-of-the-art satellite machine-learning to automate ship track detection and the team. Led by the University of Oxford in partnership with Plymouth Marine Laboratory (via the NEODAAS service) and Imperial College London, catalogued more than one million ship tracks over a 20 year period to provide a global climatology.

This was then used to investigate the effect of stringent fuel regulations introduced by the International Maritime Organization in 2020 on the ship track’s global prevalence since then, whilst accounting for the disruption in global commerce caused by COVID-19.

The analysis showed a marked but clearly nonlinear decline in ship tracks globally; an 80% reduction in emissions of sulphur oxides causes a 25% reduction in the number of tracks detected. In real terms, about 40k ship tracks were formed every year until 2020, when the fuel regulations were introduced, then the number dropped to 30k. This research has highlighted the need to consider the wider impacts of environmental regulation that could affect climate change mitigation and global climate models.

Dr Duncan Watson-Parris of the University of Oxford and lead author of the study commented: “Air pollution must be reduced for human health reasons. But our analysis demonstrates that such policies must be accompanied by determined action against global warming, to compensate for the loss of the cooling effect from human aerosols.”

The ship track dataset generated for this study is the largest dataset to date produced by Plymouth Marine Laboratory’s MAssive GPU Cluster for Earth Observation (MAGEO) system and was required to process over 250TB of raw data from the MODIS sensor, flown on the NASA Aqua satellite. The dataset is being made available via CEDA for use by other researchers.

Angus Laurenson, NEODAAS Data Analyst at Plymouth Marine Laboratory, said: “It was great to work with scientists at the University of Oxford and Imperial College London to deploy this model at scale. We put MAGEO through its paces on a genuinely ‘big’ dataset and I think it is a neat demonstration of how AI and machine-learning can be utilized to conduct research that is otherwise infeasible.”
 

Related information

Sulphur dioxide is a colourless gas with an irritating pungent odour and is a common air pollutant. It is produced naturally by active volcanoes and forest fires, however, the majority of sulphur dioxide present in the environment is due to the burning of coal and oil, such as in traditional power stations. Shipping, motor vehicles, domestic boilers and fires also release sulphur dioxide into the environment. Breathing in sulphur dioxide causes irritation of the nose and throat. Exposure to higher concentrations can cause nausea, vomiting, stomach pain and corrosive damage to the airways and lungs.

‘Shipping regulations lead to large reduction in cloud perturbations’ is published in PNAS

The paper was published as part of the Atmospheric Composition and Radiative forcing changes due to UN International Ship Emissions regulations (ACRUISE) project, read more about the project here

The dataset is available here

Computing resources for running the inference and funding supports were provided through the NERC Earth Observation Data Analysis and AI Service (NEODAAS), which is overseen by the National Centre for Earth Observation (NCEO)

More information on applying for AI support through NEODAAS is available here.