Robustness of anthropogenically forced decadal precipitation changes projected for the 21st century

• Published in: Nature communications Vol. 9; no. 1; pp. 1150 - 10
• Main Authors: Zhang, Honghai, Delworth, Thomas L.
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 20.03.2018; Nature Publishing Group; Nature Portfolio
• Subjects: 704/106/242; 704/106/35/823; 704/106/694/2786; 704/106/694/674; 704/172/4081; Anthropogenic factors; Climate change; Humanities and Social Sciences; multidisciplinary; Oceans; Precipitation; Rainfall; Regional analysis; Science; Science (multidisciplinary); Tropical environments; Variability
• ISSN: 2041-1723; 2041-1723
• DOI: 10.1038/s41467-018-03611-3

Precipitation is characterized by substantial natural variability, including on regional and decadal scales. This relatively large variability poses a grand challenge in assessing the significance of anthropogenically forced precipitation changes. Here we use multiple large ensembles of climate change experiments to evaluate whether, on regional scales, anthropogenic changes in decadal precipitation mean state are distinguishable. Here, distinguishable means the anthropogenic change is outside the range expected from natural variability. Relative to the 1950–1999 period, simulated anthropogenic shifts in precipitation mean state for the 2000–2009 period are already distinguishable over 36–41% of the globe—primarily in high latitudes, eastern subtropical oceans, and the tropics. Anthropogenic forcing in future medium-to-high emission scenarios is projected to cause distinguishable shifts over 68–75% of the globe by 2050 and 86–88% by 2100. Our findings imply anthropogenic shifts in decadal-mean precipitation will exceed the bounds of natural variability over most of the planet within several decades. Decadal precipitation changes are dominated by random natural variability, posing a challenge for projecting anthropogenic impacts. Here the authors use large suites of model simulations to show that human-induced future decadal shifts in regional precipitation can be distinguished from natural variability.