Human-induced greening of the northern extratropical land surface

• Published in: Nature climate change 2016-10, Vol.6 (10), p.959-963
• Main Authors: Mao, Jiafu, Ribes, Aurélien, Yan, Binyan, Shi, Xiaoying, Thornton, Peter E., Séférian, Roland, Ciais, Philippe, Myneni, Ranga B., Douville, Hervé, Piao, Shilong, Zhu, Zaichun, Dickinson, Robert E., Dai, Yongjiu, Ricciuto, Daniel M., Jin, Mingzhou, Hoffman, Forrest M., Wang, Bin, Huang, Mengtian, Lian, Xu
• Format: Article
• Language: English
• Subjects: 631/158; 704/106; 704/172; Algorithms; Anthropogenic factors; Archives & records; Climate Change; Climate Change/Climate Change Impacts; Climate system; Climate variability; Continental interfaces, environment; Ecosystem services; Environment; Environmental Law/Policy/Ecojustice; ENVIRONMENTAL SCIENCES; Greenhouse gases; Laboratories; letter; Ocean, Atmosphere; Physiology; Remote sensing; Science; Sciences of the Universe; Simulation; Trends; Vegetation; Vegetation changes
• ISSN: 1758-678X; 1758-6798
• DOI: 10.1038/nclimate3056

Observed northern extratropical land greening is consistent with anthropogenic forcings, where greenhouse gases play a dominant role, but not with simulations that include only natural forcings and internal climate variability. Significant land greening in the northern extratropical latitudes (NEL) has been documented through satellite observations during the past three decades 1 , 2 , 3 , 4 , 5 . This enhanced vegetation growth has broad implications for surface energy, water and carbon budgets, and ecosystem services across multiple scales 6 , 7 , 8 . Discernible human impacts on the Earth’s climate system have been revealed by using statistical frameworks of detection–attribution 9 , 10 , 11 . These impacts, however, were not previously identified on the NEL greening signal, owing to the lack of long-term observational records, possible bias of satellite data, different algorithms used to calculate vegetation greenness, and the lack of suitable simulations from coupled Earth system models (ESMs). Here we have overcome these challenges to attribute recent changes in NEL vegetation activity. We used two 30-year-long remote-sensing-based leaf area index (LAI) data sets 12 , 13 , simulations from 19 coupled ESMs with interactive vegetation, and a formal detection and attribution algorithm 14 , 15 . Our findings reveal that the observed greening record is consistent with an assumption of anthropogenic forcings, where greenhouse gases play a dominant role, but is not consistent with simulations that include only natural forcings and internal climate variability. These results provide the first clear evidence of a discernible human fingerprint on physiological vegetation changes other than phenology and range shifts 11 .