Open AccessPartitioning the Uncertainty of Ensemble Projections of Global Glacier Mass Change
Author(s) -
Marzeion Ben,
Hock Regine,
Anderson Brian,
Bliss Andrew,
Champollion Nicolas,
Fujita Koji,
Huss Matthias,
Immerzeel Walter W.,
Kraaijenbrink Philip,
Malles JanHendrik,
Maussion Fabien,
Radić Valentina,
Rounce David R.,
Sakai Akiko,
Shan Sarah,
Wal Roderik,
Zekollari Harry
Publication year - 2020
Publication title -
earth's future
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.641
H-Index - 39
ISSN - 2328-4277
DOI - 10.1029/2019ef001470
Subject(s) - glacier , representative concentration pathways , climatology , future sea level , climate change , environmental science , glacier mass balance , sea level , sea level rise , scale (ratio) , general circulation model , physical geography , atmospheric sciences , geology , geography , sea ice , oceanography , cryosphere , ice stream , cartography
Glacier mass loss is recognized as a major contributor to current sea level rise. However, large uncertainties remain in projections of glacier mass loss on global and regional scales. We present an ensemble of 288 glacier mass and area change projections for the 21st century based on 11 glacier models using up to 10 general circulation models and four Representative Concentration Pathways (RCPs) as boundary conditions. We partition the total uncertainty into the individual contributions caused by glacier models, general circulation models, RCPs, and natural variability. We find that emission scenario uncertainty is growing throughout the 21st century and is the largest source of uncertainty by 2100. The relative importance of glacier model uncertainty decreases over time, but it is the greatest source of uncertainty until the middle of this century. The projection uncertainty associated with natural variability is small on the global scale but can be large on regional scales. The projected global mass loss by 2100 relative to 2015 (79 ± 56 mm sea level equivalent for RCP2.6, 159 ± 86 mm sea level equivalent for RCP8.5) is lower than, but well within, the uncertainty range of previous projections.