Land and Sea Surface Temperatures (LST and SST respectively) are essential climate variables, and are typically measured by infrared sounders. It is particularly important when studying Urban Heat Islands, a typical phenomenon that occurs in cities where the urban surface is hotter than the surrounding.

The Geostationary Interferometric Infrared Sounder (GIIRS) is an infrared hyper-spectral space-borne interferometer that flies in geostationary orbit to take measurements of three-dimensional atmospheric structure over East Asia, and in particular China (figure 1).

Over China, there are measurements every hour on average, at an 8km resolution.

At LATMOS, we work on the IASI instrument, an infrared sounder, similar to GIIRS but in polar orbit (crosses the same location two times per day). We have developed for IASI a dedicated Land Surface Temperature product (https://iasi-ft.eu/products/skt/), that we would like to adapt and enhance to GIIRS data.

The objective of this internship is to explore existing AI methods and develop new ones to calculate land and sea surface temperatures from the raw data of GIIRS (radiances), in preparation for a similar mission/instrument with similar technical abilities over Europe and Africa (the Infrared Sounder IRS on board MeteoSat third generation).

Details of the work:

·        Examine and explore the differences between the IASI and GIIRS spectra

·        Investigate the existing LST and SST retrieval codes and adapt them to GIIRS over China.

·        Compare and validate with existing products

·        Adapt to IRS over Europe and Africa

Prerequisites and skills required

•       M2 in meteorology, math, AI tools, or M2 in environmental sciences, atmospheric physics/chemistry

•       Skills in computer programming (Python, Matlab, other) and the linux environment

•       Knowledge in Artificial Intelligence/deep learning/machine learning tools and methods

•       English proficiency