The European Geosciences Union (EGU) General Assembly 2017 was held in Vienna on 23-28 April (see http://www.egu2017.eu/). Lee de Mora and I attended from the UKESM core group; several other colleagues from the Met Office and NERC were also at the meeting.

The EGU is a learned society for the study of earth, planetary and space sciences. Tens of thousands of scientists attend its annual general assembly; around 14,500 of them, from more than a hundred countries, were at the 2017 meeting. The programme consisted of multiple parallel streams covering a very wide range of subjects; individual sessions had attendance ranging from a handful of people to several hundred. Some of the sessions were targeted at journalists, describing or discussing work which was potentially newsworthy. For example, one of these examined whether a two degree global warming limit was possible without relying on carbon capture and storage. Others addressed the subject of science communication; two particularly stimulating sessions were entitled “Make Facts Great Again: How can scientists stand up for science?”and “Science communication in the age of Brexit and Trump: How to reach the hard to reach”.

In addition, I attended an interesting technical session on the development of global coupled climate models at higher resolutions (i.e. towards the resolving of clouds and eddies) on exascale machines. This session included a presentation by Thomas Melvin of the Met Office about the LFRic project, as well as similar talks from other centres including ECMWF; this group also presented work in the poster session about climate modelling using reduced precision which can lower both execution time and output storage requirements.
Other sessions focussed on enabling technologies such as visualization – for example, DKRZ presented an animation of the CMIP5 multi-model ensemble climate projection which displayed uncertainties in an improved (compared to that used in the IPCC AR5 report) fashion.

Lee and I both spoke in the session on “Towards CMIP6 internationally coordinated climate modelling experiments”, which was well-attended and provoked stimulating discussions. I described our data generation and processing pipeline in “Responding to complexity: Building an improved modelling infrastructure for CMIP6”, while Lee described the functionality of BGC-val in “The UK Earth System Model Marine Biogeochemical Evaluation Toolkit”. The same session also featured a useful talk about how the ES-DOC project is aiming to deliver a robust documentation infrastructure for CMIP6.