Matthew Osman

Climate dynamics across timescales; integrating climate proxies, observations, and models to refine future projections; paleoclimate data assimilation; cryosphere–climate change

Research Area

My research focuses on understanding climate dynamics across timescales by integrating geochemical proxy records, modern observations, and climate model simulations using physics-informed statistics. My group develops quantitative tools to reconstruct past climates and constrain future projections – our work explicitly links paleoclimate and historical evidence with ongoing / projected climate changes.

We are particularly interested in:

1. Using data assimilation and modelling methods to reconstruct global climate and ice sheets during key intervals (e.g., the mid-Pliocene, last Ice Age, past interglacials)
2. Creating probabilistic frameworks that combine proxies, climate simulations, and modern observations to refine model-based projections of future climate change
3. Developing statistical proxy system models to better interpret proxy signals and constrain models
4. Investigating cryosphere-climate feedbacks – especially the role of ice sheets and sea ice in modulating sea level and climate sensitivity.

I lead the Cambridge Computational Climate and PaleOceanography (C3PO) group and work closely with the international PMIP/CMIP community. Our collaborative projects span Arctic sea ice sensitivity, AMOC weakening, global carbon cycle feedbacks, and applications of machine learning to past and future Earth system changes.

Project Interests

I’m especially enthusiastic to co-develop PhD projects with students interested in quantitative paleoclimate methods and future climate risk. Possible areas include (and may combine):

• Using paleo and historical climate data to constrain future projections via emergent constraints and probabilistic weighting
• Developing fingerprinting techniques to link proxy records with spatiotemporal climate signals
• Developing (sometimes sassily named) probabilistic proxy system models (for, e.g., ice core geochemistry, marine faunal assemblages)
• Applying paleoclimate data assimilation techniques during key intervals of ice sheet collapse
• Building a quantitative foundation for climate risk modelling (e.g., floods, heat wave frequency, droughts) using physics-informed statistics and paleo constraints