CISL Seminar Series: New Uses for Old Weather

06/26/2019 - 10:00am


Speaker: Philip Brohan, PhD


We are very confident that the world is warmer now than it was 100 years ago, but that's not all we need to know: is it wetter? windier? stormier? In particular, much of the impact of weather and climate change comes in extreme events - so if we forecast a future severe storm we'd like to be able to say if it's different from storms we've seen before, or just more of the same. This means we need to reconstruct the extreme weather of the last 100+ years, in detail.  Observing the weather of 100 years ago presents some unusual practical challenges: We collaborate with historians and archivists all over the world to find surviving records of past observations. We then digitise the observations from paper records, standardise them and add them to climate datasets, and assimilate them into long reanalyses. By doing this we can make large improvements in the reconstructions of historic storms, but the process is labour intensive, expensive, and slow. We are working to speed the process up dramatically, by using thousands of volunteer citizen scientists, machine learning/computer vision tools for data digitisation, new data formats and databases, and new approaches to data assimilation. 


Philip completed a PhD in theoretical physics more years ago than he cares to remember. He then worked for a while as a nuclear engineer, before switching to climate science in 2003, joining the Met Office Hadley Centre in the UK to work on marine observations. In his first week as a climate scientist, Philip worked out where the bathrooms were, went to a completely incomprehensible seminar on something called 'ENSO', and wrote to Steve Worley - starting a collaboration with NCAR that has continued ever since. Philip builds datasets of past weather and climate, and to that end he spends most of his time mounting search and rescue missions for historical weather observations.


Refreshments will be served at 9:45 a.m.!


Date: Wednesday, June 26, 2019

Time: 10:00 a.m. - 11:00 a.m.

Location: Mesa Lab, Main Seminar Room (MSR)