How the ocean affects climate on land

The study of past climate is crucial in understanding how climate changes naturally through time and thus in defining the background context against which anthropogenic global warming takes place. To study long-term temperature variability before thermometers were invented, indirect recorders of past climate variations, so called ‘climate proxies’ are used. Plants leave behind in lake sediments traces of their presence in the form of pollen grains, which once compiled provide snapshots of past ecological composition, thus tracking vegetation changes along with climate changes. 

“Thousands of pollen records have been compiled over the world and we have gathered them in a single database,” explains Prof. Ulrike Herzschuh, of the Alfred Wegener Institute, Helmholtz Centre for Polar and Marine Research (AWI). “If you want to study long-term variability over land, pollen records are essential. They provide extensive spatial coverage of past terrestrial vegetation and climate changes like no other archives since they are found anywhere where vegetation grows, or was growing in the past.” Lead author Dr. Raphaël Hébert and his colleagues analyzed this database together with temperature measurements and measured retrospectively how much the regional temperature temporally varies over years to millennia.

“If we take global climate models as a baseline for natural variability, those which are generally used for future projections, then regional temperatures are expected to be very stable, within half a degree variation for any given century, and within 0.1 to 0.3 degrees for millennia,” says Hébert regarding their study area comprising locations over land in the northern hemisphere. “While in climate models temperature shows smaller and smaller variations for longer and longer periods, our pollen-based temperature reconstructions show temperature fluctuations of about 1 degree Celsius independent of the span of the averaging period such that a typical century scale fluctuation has a similar magnitude as a millennial scale one.” This could hint at the existence of mechanisms driving long-term variability that are missing or not adequately represented in the climate models. 

The study did not only characterize the mean behavior, but also provided detailed maps of the millennial scale variability. This revealed patterns intrinsically linked to the propagation of oceanic influence inland. “When we found that the spatial patterns of millennial variability were not only spatially coherent, but also related to those from the completely independent modern temperature measurements, we suspected we found something significant for a deeper understanding of long-term variability,” explains Hébert.

The relationship the researchers identified supports a dual role of oceans in influencing temperature variability. „We know that oceanic climates are generally more stable, at least on annual to decadal timescales. What was surprising however, is that the same regions became the most variable on millennial timescales. Therefore, while oceanic influence stabilizes short term climate, it appears to be the main driver of long-term variability.“ 

The study extends on previous results on ocean temperature variations from Prof. Thomas Laepple who is heading an interdisciplinary research group funded by a European Research Council (ERC) Grant specifically on this topic. He adds “We knew already that large regional scale variability was expected over the ocean from our previous study using corals and marine sediments, but this was never confirmed over land which was believed to be more stable. Consequently, this means we could see unpredicted shifts in local climates, and they may already be hidden in the ocean. This adds to the uncertainty of climate change for the upcoming century as this layers on top of the expected global warming.”

Original publication

Hébert, R., Herzschuh, U. & Laepple, T. Millennial-scale climate variability over land overprinted by ocean temperature fluctuations. Nature Geoscience. (2022). https://doi.org/10.1038/s41561-022-01056-4.