Data.csv

Significant increases in tree mortality due to drought-induced physiological stress have been documented worldwide. This trend is likely to continue with increased frequency and severity of extreme drought events in the future. Therefore, understanding the factors that influence variability in drought responses among trees will be critical to predicting ecosystem responses to climate change and developing effective management actions. In this study, we used hierarchical mixed-effects models to analyze drought responses of Pseudotsuga menziesii in 20 unmanaged forests stands across a broad range of environmental conditions in northeastern Washington, USA. We aimed to 1) identify the biotic and abiotic attributes most closely associated with the responses of individual trees to drought and 2) quantify the variability in drought responses at different spatial scales. We found that growth rates and competition for resources significantly affected resistance to a severe drought event in 2001: slow-growing trees and trees growing in subordinate canopy positions and/or with more neighbors suffered greater declines in radial growth during the drought event. In contrast, the ability of a tree to return to normal growth when climatic conditions improved (resilience) was unaffected by competition or relative growth rates. Drought responses were significantly influenced by tree age: older trees were more resistant but less resilient than younger trees. Finally, we found differences between resistance and resilience in spatial scale: a significant proportion (approximately 50%) of the variability in drought resistance across the study area was at broad spatial scales (i.e. among different forest types), most likely due to differences in the total amount of precipitation received at different elevations; in contrast, variation in resilience was overwhelmingly (82%) at the level of individual trees within stands and there was no difference in drought resilience among forest types. Our results suggest that for Pseudotsuga menziesii resistance and resilience to drought are driven by different factors and vary at different spatial scales.

全球范围内已有多项研究记录到，干旱诱导的生理胁迫导致树木死亡率显著升高。随着未来极端干旱事件发生频率提升、强度加剧，这一趋势或将持续。因此，明确影响不同树木间干旱响应差异的驱动因子，对于预测生态系统对气候变化的响应及制定高效的经营管理措施至关重要。本研究以美国华盛顿州东北部20个未经营林分为研究对象，覆盖了广泛的环境梯度，采用层级混合效应模型（hierarchical mixed-effects models）分析花旗松（Pseudotsuga menziesii）的干旱响应特征。本研究的目标为：1）甄别与单株树木干旱响应最为密切的生物与非生物属性；2）量化不同空间尺度下树木干旱响应的变异程度。研究发现，树木生长速率与资源竞争程度显著影响其对2001年极端干旱事件的抗性：生长缓慢的树木、处于劣势冠层位置的树木，以及周边邻株较多的树木，在干旱期间的径向生长降幅更大。与之相反，当气候条件改善后树木恢复至正常生长水平的能力（恢复力，resilience），则不受资源竞争或相对生长速率的影响。树木年龄同样显著影响其干旱响应：相较于幼树，成年树木的干旱抗性更强，但恢复力更弱。最后，本研究发现干旱抗性与恢复力的空间尺度效应存在显著差异：研究区域内约50%的干旱抗性变异来源于大空间尺度（即不同林分类型之间），这大概率与不同海拔的总降水量差异相关；与之相反，恢复力的变异有82%集中于林分内单株树木尺度，且不同林分类型间的干旱恢复力并无显著差异。本研究结果表明，花旗松的干旱抗性与恢复力由不同的驱动因子调控，且在不同的空间尺度上呈现出差异化的变异特征。