Do anthropogenic aerosols enhance or suppress the surface cloud forcing in the Arctic? K. 1 Alterskjær , 1Department J. E. 1 Kristjánsson 1 Hoose and C. of Geosciences, University of Oslo, Norway Problem W/m2 • We investigate a proposed increase in the warming effect of Arctic clouds due to anthropogenic aerosols. • Observational studies from Alaska suggest that the longwave (LW) radiative flux at the surface may increase by 3.3 to 8.2 W/m2 due to aerosol-cloud interactions (Garrett and Zhao, 2006 and Lubin and Vogelmann, 2006). • We use a global climate model to study the annual and seasonal total radiative effect of interactions between anthropogenic aerosols and clouds over the entire Arctic region. • Over highly reflecting snow and ice surfaces, clouds have a small effect on the shortwave (SW) radiation budget. Hence, the longwave warming effect of the clouds dominates. Surface net cloud forcing, summer season (Jun-Sep) Model validation Indirect effect • Otherwise similar clouds are optically thicker in polluted than in clean environments due to reduced cloud droplet effective radius (re) and increased cloud liquid water path (LWP) (Twomey, 1977 and Albrecht, 1989). Clean (left) and polluted (right) environment. Figure from Lohmann (2005). Simulated re, surface albedo and surface concentrations of particulate sulfate are consistent with observations. LWP and surface cloud forcing (CFS) was validated against data from the SHEBA campaign (Zhang et al., 2002 and Intrieri et al., 2002): Results shown are from simulations in which the model LWP is reduced by a factor of five through reducing the cloud droplet number concentration (CDNC). This gave LWP and CFS closest to what was observed. Model tools and method Sensitivity experiments • CAM-Oslo global climate model • 5-year simulations • Off-line: Meteorological evolution the same in all model runs • Aerosols affect droplet nucleation only • Model modifications: Liquid water absorption coefficient depends on re: Manipulations of the simulated LWP and SO4 concentrations show that our conclusions are robust against changes in both cloud properties and aerosol concentrations. Change in annual cloud forcing [W/m2] LWP/5 ∆CDNC Min Max LW 0.55 0.10 0.85 SW -0.85 -1.29 -0.52 Net -0.30 -0.98 0.12 1 Results of sensitivity experiments • Averaged north of 71o N. • Experiments included reducing the LWP through reducing the CDNC, reducing the re, increasing the cloud ice fraction and increasing the auto-conversion so that more water is lost through precipitation. Also included were two experiments in which the SO4 concentrations were increased. • 1 Only one in ten experiments resulted in a positive annual change in CFS with anthropogenic aerosols. • Emission scenarios (AeroCom) • Pre-industrial • Present day Simulated changes in cloud properties Simulated anthropogenic change in re [µm] (left) and LWP ∆LWP [g/m2] (right), annual average. Simulated contribution to surface cloud forcing by anthropogenic aerosols [W/m2] ∆LW Observed (green) and simulated (blue) SO4 concentrations [µg S/m3]. Observational data from AMAP (2006) have been averaged over five years and error bars show the maximum and the minimum observed monthly means. The median of 10 AeroCom A models (Textor et al., 2006) for Spitsbergen (including Zeppelin) and Janiskoski for the year 2000 is shown in red. Conclusion Anthropogenic aerosols lead to: • A simulated decrease in re • A simulated increase in LWP ∆re Black dots: Observed surface albedos obtained from aircraft measurements in the vicinity of the SHEBA ship (Curry et al., 2000). Blue lines: CAM-Oslo simulated albedo. ∆SW • Stronger simulated surface cloud forcing both in the LW and the SW • An overall negative radiative effect (cooling) due to SW dominance. ∆ Total • Anthropogenic aerosols in the Arctic are likely to lead to an overall decrease in the radiative flux at the surface under cloudy skies. Change in surface cloud forcing [W/m2] Winter (Oct –May) Summer (Jun – Sep) Annual average LW 0.33 0.99 0.55 SW -0.19 -2.17 -0.85 Net 0.14 -1.18 -0.30 Averaged north of 71o N. • In the future, sulfate concentrations are expected to decrease, reducing the cooling effect, causing an enhanced melting of snow and ice. Contact information • Kari Alterskjær, [email protected] – 228 55775 References ∆LW ∆SW Top: Annual averages. Bottom: Summer/Melt season averages (June to September). Note that color bars differ. ∆ Total • • • • • • Alterskjær et al. (2010), J. Geophys. Res., 115(D22204), doi: 10.1029/2010JD014015 Albrecht, B. A., (1989) Science, 245, 1227–1230. Curry et al. (2000), Bull. Am. Meteorol. Soc., 81, 5–29. Garrett, T. J., and C. Zhao (2006), Nature, 440, 787–789, doi:10.1038/nature04636. Intrieri et al. (2002), J. Geophys. Res., 107(C10), 8039, doi:10.1029/2000JC000439. Lohmann, U. (2005). Indirect Effects: Aerosol and Cloud Microphysics. IPCC Expert Meeting on Aerosols, Geneva. • Lubin, D., and A. M. Vogelmann (2006), Nature, 439, 453–456, doi:10.1038/nature04449. • Twomey, S. (1977), J. Atmos. Sci., 34, 11491152. • Zhang et al. (2002), J. Geophys. Res., 107(D24), 4750, doi:10.1029/2001JD001484. This study was partly funded by the Norwegian Research Council through the projects POLARCAT (grant No. 175916) and NorClim (grant No. 178246).