Ice, mud, and models: how paleo data assimilation is helping illuminate our geologic past

Montag 23.01.2023, 11:00 → 12:00 Europe/Berlin

Speaker: Prof. Dr. Matthew B. Osman, University of Cambridge & University of Arizona, USA

- online colloquium -

Where?    ZOOM meeting room https://geomar-de.zoom.us/j/85043306987?pwd=VHQ2NmpQWmNvTTU0N2R0OGpMSENaQT09

Meeting ID: 850 4330 6987
Code: 438543

 

Short abstract:
Earth’s geologic past entrains a vast, variable, and highly informative climatic archive.  Incidentally, it is also often elusive to ascertain. Whereas models and climatic proxies can both be used to inform our geologic past, each comes with its own challenges: Models rely on imperfect physics and underlying assumptions, whereas proxies are sparsely situated and often uncertain. In recent years, ‘paleo data assimilation’ has arisen as a means of leveraging and combining these two archives.  In this talk, I discuss the paleoDA method including new approaches, advances, and future challenges (or, pitfalls), illustrated through recent and ongoing climate reconstruction efforts that span three million years ago up to the recent Holocene.
 

Quick Bio:
Matt Osman is an Assistant Professor in the Department of Geography at Cambridge University.  Matt’s work uses statistical methods that combine diverse datasets — climate model simulations, paleoclimate proxy databases, simple process models, and modern observations — to improve understanding of climate processes and change.  His recent efforts have focused on understanding past temperature, jet stream, sea ice, marine productivity, and ice cap dynamics. Matt received his PhD in climate science in 2019 from MIT/WHOI.