地球系统科学
气候变化
环境科学
温室气体
土地利用
土地利用、土地利用的变化和林业
气候模式
化石燃料
生产力
生态系统
自然资源经济学
碳循环
土地覆盖
农业
气候承诺
气候学
环境资源管理
全球变暖
地理
生态学
全球变暖的影响
经济
地质学
宏观经济学
考古
生物
作者
P. E. Thornton,Katherine Calvin,Andrew D. Jones,Alan Di Vittorio,Ben Bond‐Lamberty,Louise Chini,Xiaoying Shi,Jiafu Mao,W. D. Collins,Jae Edmonds,Allison M. Thomson,John Truesdale,Anthony P Craig,M. L. Branstetter,G. C. Hurtt
摘要
Significant feedbacks in energy, agriculture, land use and the carbon cycle are identified for the twenty-first century when climate impacts on land are factored into climate projections so as to allow for two-way interactions between human and Earth systems. Fossil fuel combustion and land-use change are the two largest contributors to industrial-era increases in atmospheric CO2 concentration1. Projections of these are thus fundamental inputs for coupled Earth system models (ESMs) used to estimate the physical and biological consequences of future climate system forcing2,3. While historical data sets are available to inform past and current climate analyses4,5, assessments of future climate change have relied on projections of energy and land use from energy–economic models, constrained by assumptions about future policy, land-use patterns and socio-economic development trajectories6. Here we show that the climatic impacts on land ecosystems drive significant feedbacks in energy, agriculture, land use and carbon cycle projections for the twenty-first century. We find that exposure of human-appropriated land ecosystem productivity to biospheric change results in reductions of land area used for crops; increases in managed forest area and carbon stocks; decreases in global crop prices; and reduction in fossil fuel emissions for a low–mid-range forcing scenario7. The feedbacks between climate-induced biospheric change and human system forcings to the climate system—demonstrated here—are handled inconsistently, or excluded altogether, in the one-way asynchronous coupling of energy–economic models to ESMs used to date1,8,9.
科研通智能强力驱动
Strongly Powered by AbleSci AI