NOAA/WDS Paleoclimatology - Pumijumnong - Ban Rai - TEGR - ITRDB THA019

Interannual variability in the global land carbon sink is strongly related to variations in tropical temperature and rainfall. This association suggests an important role for moisture-driven fluctuations in tropical vegetation productivity, but empirical evidence to quantify the responsible ecological processes is missing. Such evidence can be obtained from tree-ring data that quantify variability in a major vegetation productivity component: woody biomass growth. Here we compile a pantropical tree-ring network to show that annual woody biomass growth increases primarily with dry-season precipitation and decreases with dry-season maximum temperature. The strength of these dry-season climate responses varies among sites, as reflected in four robust and distinct climate response groups of tropical tree growth derived from clustering. Using cluster and regression analyses, we find that dry-season climate responses are amplified in regions that are drier, hotter and more climatically variable. These amplification patterns suggest that projected global warming will probably aggravate drought-induced declines in annual tropical vegetation productivity. Our study reveals a previously underappreciated role of dry-season climate variability in driving the dynamics of tropical vegetation productivity and consequently in influencing the land carbon sink.

全球陆地碳汇（global land carbon sink）的年际变异性与热带温度及降雨量的变化显著相关。这一关联表明，水分驱动的热带植被生产力波动发挥着关键作用，但目前仍缺乏可量化相关生态过程的实证证据。此类证据可通过树木年轮（tree-ring）数据获取，这类数据可量化植被生产力的核心组成部分——木质生物量生长的变异性。本研究构建了泛热带树木年轮网络，结果显示，年度木质生物量生长主要随干季降水增加而提升，随干季最高温升高而下降。上述干季气候响应的强度因站点而异，通过聚类（clustering）分析得到的四类稳健且独立的热带树木生长气候响应分组，即可体现这一差异。借助聚类与回归（regression）分析，我们发现，在更为干旱、炎热且气候变异性更强的区域，干季气候响应会被显著放大。这类放大模式表明，未来预估的全球变暖或将加剧干旱诱导的热带植被生产力年度下降。本研究揭示了此前未被充分重视的干季气候变异性在驱动热带植被生产力动态、进而影响全球陆地碳汇过程中的重要作用。