[A-CC26] Ice cores and paleoenvironmental modeling

[ACC26-09] Possible albedo reduction of snow surface due to black carbon over the past 350 years on the Greenland Ice Sheet estimated from the SIGMA-D ice core

*Teruo Aoki^1,2, Kumiko Goto-Azuma^1,2, Sumito Matoba³, Rigen Shimada⁴, Yutaka Kondo¹, Nobuhiro Moteki⁵, Makoto Koike⁵, Koji Fujita⁶, Hideaki Motoyama^1,2, Masahiro Hori⁷ (1.National Institute of Polar Research, 2.SOKENDAI, 3.Institute of Low Temperature Science, Hokkaido University, 4.Earth Observation Research Center, Japan Aerospace Exploration Agency, 5.School of Science, University of Tokyo, 6.Graduate School of Environmental Studies, Nagoya University, 7.School of Sustainable Design, University of Toyama)

Keywords:albedo, ice core, Greenland Ice Sheet, black carbon, snow grain size, SIGMA-D

Snow surface albedo strongly depends on snow grain size and impurity concentration (Wiscombe and Warren, 1980; Warren and Wiscombe, 1980). If these snow parameters are known, the albedo can be calculated with a radiative transfer model. This study presents the possible albedo reduction due to black carbon (BC) in snow over the past 350 years on the Greenland Ice Sheet (GrIS) based on BC concentration from an ice core and satellite-derived snow grain size. The albedo is calculated with a physically based snow albedo model (PBSAM, Aoki et al., 2011) for two temporal variations of higher and lower cases of BC concentrations obtained from the SIGMA-D ice core (Matoba et al., 2015, Goto-Azuma et al., 2019) and snow grain radius R_s1 derived from Moderate Resolution Imaging Spectroradiometer (MODIS) data from 2000 to 2019 over the GrIS (Aoki et al., 2020). We employed three R_s1 averaged over accumulation areas in the GrIS at an elevation h < 1 km (R_s1 = 478 µm), all range of h (R_s1 = 185 µm), and h > 3 km (R_s1 = 94 µm) based on the monthly median of R_s1 in June when the maximum insolation is expected. The higher and lower BC concentrations are the annual maximum of monthly mean values and annual mean value, respectively. From those two BC concentrations and BC free case, and three values of R_s1, a total of 9 kinds of temporal albedo variations are calculated for solar zenith angle at local solar noon on June 21. The albedo reductions are calculated by the albedo difference between the BC-contained case and the BC free case. Assuming the BC concentration from SIGMA-D to be the same all over the GrIS, possible albedo reduction due to BC for lower, overall, higher elevation areas of GrIS can be estimated. The maximum albedo reduction is estimated to be 0.025 in 1925 for higher BC and the largest R_s1 = 478 µm (h < 1 km), and the second largest reduction is 0.019 in 1710, and the average albedo reduction over the 350 years is 0.0054. In the lower BC case, the albedo reductions all over the period are almost less than 0.01 for any case of R_s1. For the area at h > 3 km (R_s1 = 94 µm), the maximum albedo reductions are estimated to be 0.011 and 0.0045 in 1925 for the higher and lower BC cases, respectively. These results indicate that BC on GrIS usually has not a large effect on albedo reduction but sometimes cannot be ignored (albedo reduction > 0.01) under the high concentration events.