Global tree-ring response and inferred climate variation following the mid-thirteenth century Samalas eruption

• Main Authors: Büntgen, U, Smith, SH, Wagner, S, Krusic, P, Esper, J, Piermattei, A, Crivellaro, A, Reinig, F, Tegel, W, Kirdyanov, A, Trnka, M, Oppenheimer, C
• Format: Article in Journal/Newspaper
• Language: English
• Published: Springer Science and Business Media LLC 2022
• Subjects: Climate models; Climate reconstructions; Dendrochronology; Growth response; Hydroclimate; Paleoclimate; Temperature change; Tree rings; Volcanic eruptions; Canada; Antarc*; Antarctica; Siberia
• Online Access: https://www.repository.cam.ac.uk/handle/1810/338454; https://doi.org/10.17863/CAM.85867

<jats:title>Abstract</jats:title><jats:p>The largest explosive volcanic eruption of the Common Era in terms of estimated sulphur yield to the stratosphere was identified in glaciochemical records 40 years ago, and dates to the mid-thirteenth century. Despite eventual attribution to the Samalas (Rinjani) volcano in Indonesia, the eruption date remains uncertain, and the climate response only partially understood. Seeking a more global perspective on summer surface temperature and hydroclimate change following the eruption, we present an analysis of 249 tree-ring chronologies spanning the thirteenth century and representing all continents except Antarctica. Of the 170 predominantly temperature sensitive high-frequency chronologies, the earliest hints of boreal summer cooling are the growth depressions found at sites in the western US and Canada in 1257 CE. If this response is a result of Samalas, it would be consistent with an eruption window of circa May–July 1257 CE. More widespread summer cooling across the mid-latitudes of North America and Eurasia is pronounced in 1258, while records from Scandinavia and Siberia reveal peak cooling in 1259. In contrast to the marked post-Samalas temperature response at high-elevation sites in the Northern Hemisphere, no strong hydroclimatic anomalies emerge from the 79 precipitation-sensitive chronologies. Although our findings remain spatially biased towards the western US and central Europe, and growth-climate response patterns are not always dominated by a single meteorological factor, this study offers a global proxy framework for the evaluation of paleoclimate model simulations.</jats:p>