We investigate trends in atmospheric CO₂ during three recent interglacials – the Holocene, the Eemian and MIS 11 – using an earth system model of intermediate complexity. Our model experiments show a considerable improvement in the modelled CO₂ trends for all three interglacials if peat accumulation and shallow water CaCO₃ sedimentation are included, forcing the model only with orbital and sea level changes. The Holocene CO₂ trend requires anthropogenic emissions of CO₂ only after 3 ka BP.

Within the framework of the PAGES NAm2k project, we estimated regional trends in the ground surface temperature change for the past 500 years in North America. The mean North American ground surface temperature history suggests a warming of 1.8 °C between preindustrial times and 2000. A regional analysis of mean temperature changes over the last 5 centuries shows that all regions experienced warming, but this warming displays large spatial variability and is more marked in high-latitude regions.

We simulate the climate, ice sheet, and sea-level evolution during the Last Interglacial (~ 130 to 115 kyr BP), the most recent warm period in Earth’s history. Our Earth system model includes components representing the atmosphere, the ocean and sea ice, the terrestrial biosphere, and the Greenland and Antarctic ice sheets. Our simulation is in good agreement with available data reconstructions and gives important insights into the dominant mechanisms that caused ice sheet changes in the past.

The ground surface temperature histories of the past 500 years were reconstructed at 10 sites in northern Ontario and Quebec. The regions experienced a warming of ~1–2 K for the past 150 years, agreeing with borehole reconstructions for southern Ontario and Quebec and proxy data. Permafrost maps locate the sites in a region of discontinuous permafrost but our reconstructions suggest that the potential for permafrost was minimal to absent over the past 500 years.

The fossilized seeds of common millet are suited to the production of quantitative Holocene precipitation reconstructions. Our reconstructed results showed that summer precipitation from 7.7–3.4 ka BP was ~ 50 mm, or 17 % higher than present levels. Maximal mean summer precipitation peaked at 414 mm during 6.1–5.5 ka BP, ~ 109 mm, or 36 % higher than today, indicating the EASM peaked at this time. This work can provide a new proxy for further research into continuous paleoprecipitation sequences.

Latitudinal shifts in the Southern Ocean westerly wind jet could explain large observed changes in the glacial to interglacial ocean CO₂ inventory. However there is considerable disagreement in modelled deglacial-warming jet shifts. Here multi-model output is used to show that expansion of sea ice during the glacial period likely caused a slight poleward shift and intensification in the westerly wind jet. Issues with model representation of the winds caused much of the previous disagreement.

Indirect evidence on past climate comes from the former composition of ecological communities such as plants, preserved as pollen grains in sediments of lakes. Transfer functions convert relative counts of species to a climatologically meaningful scale (e.g. annual mean temperature in degrees C). We show that the fundamental assumptions in the algorithms impact the reconstruction results in he idealized model world, in particular if the reconstructed variables were not ecologically relevant.

We quantify the contributions of physical and biogeochemical changes in the ocean to enhancing ocean carbon storage at the Last Glacial Maximum. We find that simulated circulation and surface conditions cannot explain changes in carbon storage or other major biogeochemical fields that existed during the glacial climate. Key modifications to the functioning of the biological pump are therefore required to explain the glacial climate and improve model–proxy agreement for all fields.