A Combined Tree Ring and Vegetation Model Assessment of European Forest Growth Sensitivity to Interannual Climate Variability
A Combined Tree Ring and Vegetation Model Assessment of European Forest Growth Sensitivity to Interannual Climate Variability
复制标题
DOI：
10.1029/2017gb005856
复制
发表时间：
2018-08
影响因子：
5.2
通讯作者：
S. Klesse;F. Babst;Sebastian Lienert;R. Spahni;Fortunat Joos;O. Bouriaud;Marco Carrer;A. D. Filippo;Ben Poulter;V. Trotsiuk;Rob Wilson;David Frank
中科院分区：
文献类型：
--
作者：
S. Klesse;F. Babst;Sebastian Lienert;R. Spahni;Fortunat Joos;O. Bouriaud;Marco Carrer;A. D. Filippo;Ben Poulter;V. Trotsiuk;Rob Wilson;David Frank
The response of forest growth to climate variability varies along environmental gradients. A growth increase and decrease with warming is usually observed in cold‐humid and warm‐dry regions, respectively. However, it remains poorly known where the sign of these temperature effects switches. Here we introduce a newly developed European tree ring network that has been specifically collected to reconstruct forest aboveground biomass increment (ABI). We quantify, how the long‐term (1910–2009) interannual variability of ABI depends on local mean May–August temperature and test, if a dynamic global vegetation model ensemble reflects the resulting patterns. We find that sites at 8 °C mean May–August temperature increase ABI on average by 5.7 ± 1.3%, whereas sites at 20 °C decrease ABI by 3.0 ± 1.8% m−2 year−1 Δ°C−1. A threshold temperature between beneficial and detrimental effects of warming and the associated increase in water demand on tree growth emerged at 15.9 ± 1.4 °C mean May–August temperature. Because interannual variability increases proportionally with mean growth rate—that is, the coefficient of variation stays constant—we were able to validate these findings with a much larger tree ring data set that had been established following classic dendrochronological sampling schemes. While the observed climate sensitivity pattern is well reflected in the dynamic global vegetation model ensemble, there is a large spread of threshold temperatures between the individual models. Also, individual models disagree strongly on the magnitude of climate impact at the coldest and warmest locations, suggesting where model improvement is most needed to more accurately predict forest growth and effectively guide silvicultural practices.
