1NASA Goddard Institute for Space Studies and Columbia Earth Institute, New York, NY, USA
2National Center for Atmospheric Research (NCAR), Boulder, CO, USA
3Department of Atmospheric, Oceanic and Space Sciences, University of Michigan, Ann Arbor, MI, USA
4Center for International Climate and Environmental Research Oslo (CICERO) and Department of Geosciences, University of Oslo, Oslo, Norway
5Desert Research Institute, Nevada System of Higher Education, Reno, NV, USA
6State Key Laboratory of Loess and Quaternary Geology, Institute of Earth Environment, Chinese Academy of Sciences, Xi'an, China
7Met Office, Hadley Centre, Exeter, UK
8Department of the Environment and Heritage, Australian Antarctic Division, Antarctic Climate and Ecosystem Cooperative Research Centre, Tasmania, Australia
9Department of Imaging and Applied Physics, Curtin University, Bentley, WA, Australia
10Pacific Northwest National Laboratory, Richland, WA, USA
11NOAA Geophysical Fluid Dynamics Laboratory, Princeton, NJ, USA
12Center for International Climate and Environmental Research Oslo (CICERO), Oslo, Norway
13National Institute for Environmental Studies, Tsukuba-shi, Ibaraki, Japan
14UCAR/NOAA Geophysical Fluid Dynamics Laboratory, Princeton, NJ, USA
15Department of Earth and Environmental Science, Graduate School of Environmental Studies, Nagoya University, Nagoya, Japan
16Research Institute for Applied Mechanics, Kyushu University, Fukuoka, Japan
17F.-A. Forel Institute, University of Geneva, Versoix, Switzerland
Abstract. As part of the Atmospheric Chemistry and Climate Model Intercomparison Project (ACCMIP), we evaluate the historical black carbon (BC) aerosols simulated by 8 ACCMIP models against observations including 12 ice core records, long-term surface mass concentrations and recent Arctic BC snowpack measurements. We also estimate BC albedo forcing by performing additional simulations using offline models with prescribed meteorology from 1996–2000. We evaluated the vertical profile of BC snow concentrations from these offline simulations using the recent BC snowpack measurements.
Despite using the same BC emissions, the global BC burden differs by approximately a factor of 3 among models due to differences in aerosol removal parameterizations and simulated meteorology: 34 Gg to 103 Gg in 1850 and 82 Gg to 315 Gg in 2000. However, the global BC burden from preindustrial to present-day increases by 2.5–3 times with little variation among models, roughly matching the 2.5-fold increase in total BC emissions during the same period. We find a large divergence among models at both Northern Hemisphere (NH) and Southern Hemisphere (SH) high latitude regions for BC burden and at SH high latitude regions for deposition fluxes. The ACCMIP simulations match the observed BC surface mass concentrations well in Europe and North America except at Jungfraujoch and Ispra. However, the models fail to predict the Arctic BC seasonality due to severe underestimations during winter and spring. The simulated vertically resolved BC snow concentrations are, on average, within a factor of 2–3 of the BC snowpack measurements except for Greenland and the Arctic Ocean.
For the ice core evaluation, models tend to capture both the observed temporal trends and the magnitudes well at Greenland sites. However, models fail to predict the decreasing trend of BC depositions/ice-core concentrations from the 1950s to the 1970s in most Tibetan Plateau ice cores. The distinct temporal trend at the Tibetan Plateau ice cores indicates a strong influence from Western Europe, but the modeled BC increases in that period are consistent with the emission changes in Eastern Europe, the Middle East, South and East Asia. At the Alps site, the simulated BCsuggests a strong influence from Europe, which agrees with the Alps ice core observations. Models successfully simulate higher BC concentrations observed at Zuoqiupu during the non-monsoon season than monsoon season, but models underpredict BC in both seasons. Despite a large divergence in BC deposition at two Antarctic ice core sites, models are able to capture the relative increase from preindustrial to present-day seen in the ice cores.
In 2000 relative to 1850, globally annually averaged BC surface albedo forcing from the offline simulations ranges from 0.014 to 0.019 W m−2 among the ACCMIP models. Comparing offline and online BC albedo forcings computed by some of the same models, we find that the global annual mean can vary by up to a factor of two because of different aerosol models or different BC-snow parameterizations and snow cover. The spatial distributions of the offline BC albedo forcing in 2000 show especially high BC forcing (i.e. over 0.1 W m−2) over Manchuria, Karakoram, and most of the Former USSR. Models predict the highest global annual mean BC forcing in 1980 rather than 2000, mostly driven by the high fossil fuel and biofuel emissions in the Former USSR in 1980.