The 2012–2016 California Drought severely impacted natural vegetation across a wide range of environmental gradient. Although several studies have reported an increase in plant water stress and mortality, the spatiotemporal variations of ecosystem productivity responses and the associated environmental and biological drivers remain unclear. Here, using Enhanced Vegetation Index from the Moderate resolution imaging spectrometer, we found that 45% of the natural ecosystems showed an abrupt change (breakpoint [BP]) in productivity during 2012–2016. There were three major contrasting temporal patterns of productivity responses: (i) a steady increase under higher temperature followed by a decline due to accumulated moisture depletion (high elevation forest) or temperature decrease (high elevation nonforest), (ii) gradual decline during the drought followed by a rapid recovery within 1 year after drought stress was partially relieved, and (iii) both a gradual decline and an abrupt decline. The magnitude of abrupt changes was negatively correlated (r = −0.80, p < 0.001) with initial gradual changes. Overall, changes during BP offset, on average, 57% of the preceding gradual responses. The spatial variability in ecosystem response patterns is driven by both environmental and biological factors. Particularly, for forests, positive BP was driven by increasing rainfall and decreasing temperature, while negative BP was mainly driven by the precipitation anomaly. By 2019, 33% of the natural vegetation have recovered to the level of EVI in 2010. Ecosystem responses to multiyear droughts can influence ecosystem dynamics in a complex pattern. Multiple ecohydrological factors should be considered to understand and predict the long‐term drought impacts on ecosystems.

中文翻译：

---

恢复：快速和缓慢— 2012-2016年加利福尼亚干旱期间的植被响应

2012-2016年加州干旱在广泛的环境梯度中严重影响了自然植被。尽管几项研究报告了植物水分胁迫和死亡率的增加，但生态系统生产力响应的时空变化以及相关的环境和生物驱动因素仍不清楚。在这里，使用中分辨率成像光谱仪的增强植被指数，我们发现2012-2016年期间，有45％的自然生态系统显示出生产力的突然变化（断点[BP]）。生产力响应有三种主要的时间对比模式：（i）在较高温度下稳步上升，然后由于累积的水分消耗（高海拔森林）或温度下降（高海拔非森林）而下降，（ii）在干旱期间逐渐下降，随后在干旱压力得到部分缓解后的1年内迅速恢复，并且（iii）逐渐下降和突然下降。突变的幅度呈负相关（r  = −0.80，p  <0.001），且有初始的逐渐变化。总体而言，BP期间的变化平均抵消了之前的渐进响应的57％。生态系统响应模式的空间变异性是由环境和生物因素共同驱动的。特别是，对于森林，正BP受到降雨增加和温度降低的驱动，而负BP主要受降水异常的驱动。到2019年，33％的天然植被已恢复到2010年的EVI水平。生态系统对多年干旱的响应可能以复杂的模式影响生态系统的动态。应该考虑多种生态水文因素，以了解和预测长期干旱对生态系统的影响。