Quantification of the climate sensitivity of terrestrial ecosystems through the analysis of ultra-trace gases in ice cores over the last millennium

We present a project aiming to provide a new estimate of the parameter known as "climate sensitivity" (symbol γ_L in the models) which is essential to constrain models of future climate change. This parameter describes how the amount of carbon sequestered by terrestrial ecosystems depends on temperature. Predictions of future climate by models show significant uncertainties associated with the estimates of carbon sequestration by terrestrial ecosystems with future temperature increases. Quantifying γ_L with data measured in the industrial era is very complicated because the terrestrial part of the carbon cycle is dominated by the effect of the increase in atmospheric CO₂ (the so-called anthropogenic “fertilization” or CO₂ concentration feedback, symbol β_L in the models), while the effect of temperature is smaller. We will derive γ_L from measurements of ultra-trace gases trapped in polar ice cores in pre-industrial times.

The Little Ice Age (i.e. the period that roughly covers the centuries 1400-1800 AD) was characterized by temperatures lower than the average of the last millennium, due to intense volcanic activity and reduced solar activity. The global decrease in temperature has coincided with a decrease in the atmospheric concentration of CO₂, mainly caused by sequestration from terrestrial ecosystems. Low CO₂ concentrations contributed negligibly to the decrease in temperature, making the Little Ice Age a suitable time to derive γ_L.

Why CO2 decreased during the Little Ice Age is debated. On the one hand, considerations deriving from models that simulate the amount of carbon present in terrestrial ecosystems suggest that primary productivity increased during the Little Ice Age because of an anthropogenic effect. This increase would have been caused by pandemics and colonial conquests in America which led to a depopulation of cultivated lands and a regrowth of tree species. On the other hand, measurements of carbonyl sulphate (COS) and numerical calculations capable of closing the COS budget suggest that primary productivity naturally decreased during the Little Ice Age. In this second case, the decrease in CO₂ would be caused by the fact that the respiration of terrestrial ecosystems decreased to a greater extent than the decrease in primary productivity. Therefore, if this second hypothesis is correct, it would be possible to derive γ_L from COS data covering the Little Ice Age.

Unfortunately, COS measurements covering the Little Ice Age have great uncertainty. It is therefore necessary to carry out new measurements of COS concentration during the Little Ice Age. The COS measurements will be accompanied by CO₂ and δ¹³C-CO₂ measurements, necessary to confirm, on the one hand, the working hypothesis, and, on the other, the quality of the ice samples used. Finally, future developments could build on measurements of COS isotopes in ice samples.

Rubino M., et al. Terrestrial uptake due to cooling responsible for low atmospheric CO₂ during the Little Ice Age, Nature Geoscience, 9, 691-694 (2016)