University of Edinburgh logo British Antarctic Survey logo National Oceanographic Centre logo University of Leeds logo


Using satellite data to understand the influence of marine biogenic activity on high latitude clouds and climate

This project aims to use satellite data to advance our understanding of how biogenic marine activity affects clouds and climate in remote high latitude regions.

The influence of aerosol particles that act as seeds to form cloud droplets and ice represents the largest uncertainty in our predictions of future climate change. This is particularly true for the pre-industrial period that is used as a baseline for determining the influence of anthropogenic aerosol. In the pre-industrial period natural aerosol sources dominated and so there is a great need to understand them in order to make accurate climate predictions.

There is good evidence that phytoplankton in the oceans strongly influence aerosols and cloud drop number concentrations (CDNC) in remote regions, which in turn affects how much of the Sun’s energy clouds reflect back to space. However, many of the details of the processes involved need to be clarified. Recent satellite studies suggest that the latest Earth System climate models (as used for the upcoming CMIP6 climate assessment) severely underestimate CDNC at high latitudes pointing towards exotic sources of biogenic aerosol potentially related to the presence and/or the melt of sea-ice. Our knowledge of what determines ice formation in high-latitude clouds is even more uncertain, but yet also has important implications for cloud reflectivity and climate model predictions. Our lack of knowledge partly stems from the inaccessibility of remote locations such as Antarctica, the Southern Ocean and the Arctic. Satellite instruments thus represent an ideal way to study these regions. The aims of the project are as follows :-

1) Validate the latest retrievals of CDNC and cloud glaciation using field campaign data: There is uncertainty in current CDNC and ice concentration retrievals that use reflected light from the Sun and a need to determine whether the high CDNC values around Antarctica observed by satellite are due to retrieval biases.

2) Use satellite data to determine the factors that lead to high CDNC values near Antarctica and other high latitude regions, and to determine the degree and potentially the causes of glaciation of high latitude clouds: Satellite retrievals of CDNC, ice concentrations, cloud phase (liquid or ice) and other cloud properties, as well as retrievals of sea-ice coverage and markers of phytoplankton activity, will be used. This will require the analysis of vast amounts of data with the opportunity to use machine learning techniques to determine correlations between CDNC/cloud glaciation and potential causative factors.

3) Use the satellite data to test and improve the representation of high latitude aerosols, CDNC, ice processes and clouds in the latest Met Office Earth system climate model (the UKESM1) : The model evaluation will involve the analysis of millions of variants of the UKESM1 model that sample uncertain parameter combinations. The aim is to use the satellite data to identify plausible model variants and reduce uncertainty in biogenic aerosol formation mechanisms. This will require the use of supercomputing facilities and processing techniques for “big data”.

The British Antarctic Survey is one project partner for this work and will bring great experience in collecting and analysing high latitude data, and will provide the opportunity to work with data from past and upcoming field campaigns, and potentially the chance to take part in an upcoming Antarctic campaign. Met Office partners will provide access to the latest cutting edge UKESM1 earth system model and supercomputing facilities with the opportunity to help develop the next generation of climate models.

Candidate: We are looking for a highly motivated candidate with strong computing skills and a background education in atmospheric science, remote sensing, physics, or other physical/computational/mathematical discipline. All the students will receive intensive training (12 weeks) in advanced Earth Observation and data analytical techniques for the first year (