Unnatural trend of global land long-term surface air temperature change

摘要

In the context of global warming, determining to what extent natural variability and external forcing have an impact on climate change is crucial for formulating mitigation policies and adaptation plans. We quantifiably and effectively separate unnatural trends from the global land surface air temperature (SAT) changes during 1951-2015 by calculating the long-term correlation exponent through detrended fluctuation analysis and applying mainly probability statistics methods. We provide effective discriminated evidence for whether the observed SAT change exceeds its natural internal variability. Furthermore, we search the range of SAT changes caused by unnatural trends, and identify their corresponding minimum contributions to warming relative trends in different regions. The unnatural warming of the mid- to high latitude inland areas of Eurasia and North America is basically high, and the warming responses to unnatural forcing are most intense in these areas. Compared to earlier detection, we find the minimum unnatural trend contributes more than 80% to the observed SAT change in North America, northern Greenland, eastern Africa, and the mid- to high latitudes of Eurasia. In addition, the temporal evolutions of unnatural trends in key areas are determined. In North Africa, Western Australia, the United States and Mexico, the unnatural impact became significant at the earliest. Meanwhile, the overall unnatural trend in North Africa is the most significant, followed by western Brazil, and grows most rapidly over time in the two key areas.