EGU session: CR2.5 Data Science and Machine Learning for Cryosphere and Climate
Calling all polar scientists who are working on data science and machine learning projects… At this year’s EGU we are convening an exciting session which aims to bring together the increasing number of us
working in this area to disseminate our research and stimulate discussion around how we can use more advanced data science techniques to answer important research questions in cryosphere and climate studies. Session details are below, the abstract deadline
is Jan 15th – we hope you can join us online in April!
Amber Leeson, James Lea, Michel Tsamados, Celia Baumhoer
Understanding and predicting climate variability is vital if we are to properly prepare for the impact of climate change in an increasingly warmer world, including rising sea level as a result of melting ice
and iceberg discharge. Fortunately, technological developments mean that 1) our numerical models of the cryospheric and climate systems are increasingly able to capture their inherent complexity, and 2) we are able to acquire much more detailed observations
of our polar regions by satellite than ever before. This also brings an important challenge however: how can we extract the maximum possible meaning from these data while minimizing the increase in uncertainty that added volume/complexity/heterogeneity brings?
In this session we invite submissions on research that applies Data Science techniques to answer research questions in glaciology and polar climate studies. This includes, but is not limited to, studies using
machine learning and AI, advanced statistics (e.g. extreme value analysis or changepoint methods), surrogate modelling (emulators), network analysis and innovative software/computing solutions. These could be applied to any, or any combination of, data sources
including remote sensing, numerical model output and field/ground/lab observations. We are particularly interested in contributors interested in a wider discussion about Data Science and its application in climate and cryospheric research and in contributions
which reveal new insight that would not be possible using traditional methods.