Reconstructing past climate variability in the Iberian Peninsula using carbon and oxygen stable isotopes in tree rings

Improving the understanding of past climate in the Mediterranean basin is still a challenge due to the very distinct seasonality and high climatic variability inherent to this region. Studying the climate of the Iberian Peninsula is particularly complicated because of the complex orography, as well as the atmospheric circulation patterns composed by the influence of three climatic regimes: Atlantic, continental and Mediterranean. In this investigation, we seek climatic information recorded in several Iberian tree-ring chronologies. We found that stable carbon (δ¹³C) and oxygen (δ¹⁸O) isotope ratios measured in tree rings from pine forests located at the northern part of the Iberian Peninsula are very sensitive proxies to moisture variations during the summer period. While the width of the tree rings, the dendrochronological classical proxy most commonly and extensively used, is very dependent on local site conditions, the δ¹³C and δ¹⁸O series were able to capture a large-scale climatic signal of summer aridity. In the studied environments, the isotopic signatures seem to be mainly dominated by variations in stomatal conductance driven by changes in air relative humidity. As we were interested in the response of trees to changes in moisture or aridity conditions, we removed all the other non-climatic variability shown by the isotopic series. First, the raw δ¹³C series were corrected for a decreasing trend attributed to the rise of ¹³C depleted atmospheric CO₂ due to fossil fuel burning and deforestation since industrialization (Suess effect). A second correction was also applied to remove the low-frequency variability apparently caused by tree physiological responses to changes in CO₂ concentration. In contrast, no treatment was needed for δ¹⁸O. We found that both stable isotopic records were able to properly track July and August temperatures and precipitation fluctuations in the high-frequency domain. However, the climatic significance of their low-frequency variability remained more uncertain. Several statistical approaches were explored in order to discard non-climatic variability or noise. Aside from comparisons with paleoclimatic reconstructions based on tree rings and other natural archives, we also investigated how to validate our findings with a completely independent source of climatic information such as documentary and/or historical data.