Department of Environmental Science, Policy and Management, University of California, Berkeley, California, USA.
Climate and Ecosystem Sciences Division, Lawrence Berkeley National Laboratory, Berkeley, California, USA.
Glob Chang Biol. 2022 May;28(9):3083-3093. doi: 10.1111/gcb.16104. Epub 2022 Feb 17.
The length of the growing season has a large influence on the carbon, water, and energy fluxes of global terrestrial ecosystems. While there has been mounting evidence of an advanced start of the growing season mostly due to elevated spring air temperatures, the mechanisms that control the end of the growing season (EOS) in most ecosystems remain relatively less well understood. Recently, a strong lagged control of EOS by growing season photosynthesis has been proposed, suggesting that more productive growing seasons lead to an earlier EOS. However, this relationship has not been extensively tested with in-situ observations across a variety of ecosystems. Here, we use observations from 40 eddy-covariance flux tower sites in temperate and boreal ecosystems in the northern hemisphere with more than 10 years of observations (594 site-years), ground observations of phenology, satellite observations from the Moderate Resolution Imaging Spectroradiometer (MODIS), and three leaf senescence models to test the extent of a relationship between growing season photosynthesis and end of season senescence. The results suggest that there is no significant negative relationship between growing season photosynthesis and observed leaf senescence, flux-inferred EOS estimates, or remotely sensed phenological metrics, in most ecosystems. On the contrary, while we found negative effects of summer air temperatures and autumn vapor pressure deficit on EOS, more productive growing seasons were typically related to a later, not earlier, EOS. Our results challenge recent reports of carry-over effects of photosynthesis on EOS timing, and suggest those results may not hold over a large range of ecosystems.
生长季节的长度对全球陆地生态系统的碳、水和能量通量有很大影响。尽管越来越多的证据表明，由于春季气温升高，生长季节的开始时间提前，但大多数生态系统中控制生长季节结束（EOS）的机制仍相对了解较少。最近，有人提出 EOS 受生长季光合作用的强烈滞后控制，这表明更具生产力的生长季节会导致 EOS 更早到来。然而，这种关系尚未通过各种生态系统的现场观测得到广泛验证。在这里，我们使用了北半球温带和北方森林生态系统中 40 个涡度协方差通量塔站点的观测数据，这些站点的观测时间超过 10 年（594 个站点年），包括物候地面观测、中分辨率成像光谱仪（MODIS）卫星观测以及三种叶片衰老模型，以检验生长季光合作用与季节末衰老之间的关系程度。结果表明，在大多数生态系统中，生长季光合作用与观测到的叶片衰老、通量推断的 EOS 估计值或遥感物候指标之间没有显著的负相关关系。相反，虽然我们发现夏季气温和秋季蒸气压亏缺对 EOS 有负面影响，但生产力较高的生长季节通常与 EOS 较晚而不是较早有关。我们的结果对光合作用对 EOS 时间的滞后效应的最新报告提出了挑战，并表明这些结果可能不适用于大范围的生态系统。
