UK Met Office Informatics Lab: Data platforms, causality and physics driven machine learning seminar - Kevin Donkers

As the volume and speed of data availability increases the demand for data platforms to ease access increases. This bore Pangeo, an elastically scalable data platform which can be deployed on any compute resource. The UK Met Office Informatics Lab has been exploring the use of such platforms for operational meteorology and its integration with scientific research. Moreover, due to the availability of data this platform can facilitate machine learning projects and their pull through to operations. This talk will also look at where the Informatics Lab is exploring machine learning for precipitation nowcasting, the challenges of physics driven machine learning, and where causality can be exploited for hypothesis testing and generation.

Kevin works for the UK Met Office Informatics Lab, who are a research and innovation team looking at novel ways to view, use and access UK Met Office data. After a degree in chemical physics and some time in chemical robotics research he joined the UK Met Office as a scientific software developer on the Iris and Cartopy projects, before joining the Informatics Lab in 2019. His current areas of research are data provenance and workflow tracking.


Room Number: 

Type of event:

Announcement Timing: 
Tuesday, October 8, 2019 to Monday, October 14, 2019
Calendar Timing: 
Monday, October 14, 2019 - 2:00pm to 3:00pm

Posted by Jenny Bolton at ext. 2852,

Quick Facts

NCAR is managed by the nonprofit University Corporation for Atmospheric Research on behalf of NSF and the UCAR university community.

  • NCAR is not a federal agency and its employees are not part of the federal personnel system.
  • Our activities complement those of the federal agencies and we work closely with them.

More about us