环境科学
水分
土壤碳
估计
含水量
碳纤维
土壤科学
陆地生态系统
土壤水分
生态学
地理
数学
生态系统
地质学
气象学
生物
经济
岩土工程
管理
算法
复合数
作者
Heng Yan,Zhenghui Xie,Binghao Jia,Peihua Qin,Zhang Xia,Qunlong Dai,Jinbo Xie,Longhuan Wang,Ruichao Li,Yuhang Tian,Yanbin You
摘要
Abstract Most ecosystems have resistance to soil moisture (SM) deficit, which is termed drought resistance. Drought resistance can be invalid and global terrestrial carbon uptake losses can be aggravated when SM deficit exceeds a critical threshold. However, soil moisture thresholds (SMTs) that detrimentally impact global terrestrial carbon uptake are still unclear. We performed numerical simulations using the Community Earth System Model, and estimated the SMTs by the back propagation neural network method for the years 2004–2014. The SMTs represent the inflection point for vegetation changes from high to low drought resistance phase, and terrestrial carbon uptake losses from low to high rate. Soil moisture‐limited ecoregions have higher SMTs than energy‐limited ecoregions, indicating the increased vulnerability and sensitivity of SM‐limited ecoregions to SM deficit and more easily aggravated terrestrial carbon uptake losses during drought. SMTs varied in different vegetation types and broadleaf deciduous trees displayed the highest SMTs and C3 arctic grasses have lowest thresholds. Humid and high vegetation coverage rate regions have lower thresholds. The SMTs increase with the increase of clay content and the decrease of sand content. In addition, land‐atmosphere feedback caused by SM deficit has a large impact on terrestrial carbon uptake and may be one of the main reasons for the aggravation of vegetation carbon uptake losses. Our results provide a unique perspective for investigating the impact of drought on vegetation.
科研通智能强力驱动
Strongly Powered by AbleSci AI