Black carbon (BC), a dark component of carbonaceous particles emitted from incomplete combustion processes (Masiello, 2004), is thought to be the second-largest contributing agent to observed global warming from anthropogenic activities after carbon dioxide (Jacobson, 2010). BC heats the air by absorbing shortwave radiation and re-emitting long-wave radiation back into the atmosphere (Jacobson, 2010). Studies conducted with global general circulation models show that heating induced by BC can modify atmospheric temperature gradients, which perturbs stability over a region and thus augments large-scale circulation systems, particularly those over Asian domains. As such, BC exerts a significant impact on the Asian monsoon climate (Menon et al., 2002; Ramanathan et al., 2005; Zhang et al., 2009; Mahmood and Li, 2011, 2012).

Although the radiative impacts of BC aerosols are strong over isolated source regions due to their short residence time in the atmosphere (Shindell and Faluvegi, 2009), the significant component of these responses can extend great distances through changes in circulation and thermodynamic processes (Kim et al., 2006; Wang, 2007). Kim et al. (2006) investigated the observed dipole-like Eurasian temperature anomaly occurring in the spring season relative to global aerosol concentrations, which they attributed to remote influences of absorbing aerosols including BC and dust. Moreover, they determined that such anomalous patterns away from major aerosol source regions can be detected through global teleconnections excited as climate responses to aerosol direct radiative forcing. Similarly, Wang (2009) examined such remote influences on tropical convective precipitation by using modeled responses to isolated BC concentrations over various locations and attributed the anomalous responses to altered positioning of the Inter-tropical Convergence Zone (ITCZ) toward the warming region. Moreover, Shindell and Fluvegi (2009) concluded that forcing from middle-and high-latitude aerosols can be significantly stronger than those from low-latitudes aerosols, which highlights the importance of aerosol concentrations over high-latitudinal regions such as Europe and North America. Recently, Cowan and Cai (2011) compared the influences of Asian and non-Asian aerosols on the Asian summer monsoon by using a global general circulation model and determined that non-Asian aerosols can significantly intensify the Asian monsoon response to local aerosol forcing. This sensitivity of the Asian climate to aerosols in upstream non-Asian regions further illustrates their strong influence on climate in downstream regions. Of these aerosols, BC is of particular importance.

Europe, distantly located upstream from East Asia, contains a high concentration of aerosols including BC. The amount of influence European black carbon (EUBC) exerts on the East Asian climate is an intriguing and important topic for study. However, the answer remains unclear due to a lack of relevant research. In particular, reductions in EUBC emission have occurred since 1992 (Schaap et al., 2004). Coincidentally, the East Asian summer monsoon (EASM) weakened substantially around the early 1990s (Ding et al., 2007). Analysis of a possible connection between the two is also an interesting topic. Such considerations have motived the present study, which uses an ensemble of atmospheric general circulation model (AGCM) simulations to investigate the impact of EUBC on the East Asian summer climate.

The model used in this study, the Geophysical Fluid Dynamics Laboratory Atmospheric Model version 2.1 (GFDL AM2.1), has a horizontal resolution of 2.5° longitude × 2° latitude with 24 vertical levels in a hybrid coordinate grid from approximately 30 m above the surface to approximately 3 hPa. Aerosol climatologies were obtained from the Model for Ozone And Related chemical Tracers (MOZART) (Horowitz, 2006; Horowitz et al., 2003), which is driven by meteorological inputs in which simulated aerosol climatologies are in general agreement with observations (Horowitz, 2006; Tie et al., 2005; Ginoux et al., 2006).

We performed two sets of ensemble experiments, each consisting of five members. The first set was a control group (CTL) with historical evolutions of aerosol burdens that included BC. Each of the five runs in this group was simulated by integrating the model for 21 years for the first five days in January from 1980 to December 2000. The second ensemble simulations, referred to as noEUBC, were performed in a manner similar to that of CTL except that the BC maintained the level of that recorded in 1860 over the European region (35-65° N, 0-60° E). All other model settings including greenhouse gases and other aerosol species were identical in both ensemble simulations. The sea surface temperature (SST) was prescribed with climatological seasonal cycle.

Only the outputs from the summer season (June-August, JJA) are described. The responses for EUBC were calculated as the difference between the 20-year CTL average, which excluded the model spin-up year of 1980, and the noEUBC ensemble simulations. We applied a two-sided Student’ s t-test to evaluate the significance of the modeled responses.

Figure 1a shows the difference in BC column burden over Europe in the summer. The maximum center was located in western Europe at value of 3 mg m^-2. The annual mean BC burdens over Europe determined by CTL (not shown here) were in agreement with those reported by Ginoux et al. (2006) and Bond et al. (2013). The corresponding net shortwave radiative flux absorbed by BC in the atmosphere for clear-sky conditions, presented in Fig. 1b, was calculated by subtracting the net shortwave radiative flux at surface from that at top of air (TOA). Coinciding with the BC burden maximum center, the maximum value of the EUBC-induced shortwave radiative flux was + 14 W m^-2, which is larger in magnitude than that reported by Chung et al. (2005). However, their result was based on all global anthropogenic aerosols, while the focus of the present study is EUBC. Therefore, the difference is likely attributed to the scattering and reflecting of other aerosol species in addition to model differences. Ramanathan et al. (2007) determined that strong heating within the lowest three kilometers of the troposphere absorbs aerosols over South Asia. Similarly, Meehl et al. (2008) demonstrated that solar radiation is absorbed by the BC-enhanced shortwave heating rate in the lower troposphere of the subcontinent. Thus, strong positive radiative forcing over Europe likely caused enhanced shortwave heating in the BC-forcing region. Figure 1c illustrates the clear-sky shortwave heating rate anomaly averaged within the lower troposphere at 1000-500 hPa. Our results are consistent with those reported in previous studies; BC induced a significantly enhanced shortwave heating rate over the forcing region at a maximum of approximately 0.2 K d^-1. Figure 1d shows the heating rate averaged over 45-55° N, which covers the latitude range of the BC burden maximum center. While heating at surface was essentially confined to the BC forcing region, the shortwave heating area in the mid-upper troposphere expanded downstream, with a secondary heating center situated over East Asia and the northwestern Pacific. Such expansion could strongly influence the circulation over Eurasia and affect climate conditions in East Asia. The heating center over Europe at 300-400 hPa is attributed to the shortwave radiative absorption of BC; the secondary center could be attributed to a change in surface albedo due change in circulation through dynamical processes.

Figure 1

Figure 1 (a) Black carbon (BC) column burden differences over Europe observed between the model’ s control (CTL) and noEUBC (European BC simulation) ensembles (units: mg m-2); (b) simulated responses to EUBC for net shortwave radiative flux absorbed by BC in the atmosphere under clear-sky conditions (units: W m-2); (c) shortwave radiative heating rate averaged within the lower troposphere at 1000-500 hPa under clear-sky conditions (units: K d-1); and (d) shortwave radiative heating rate averaged at 45-55° N under clear-sky conditions (units: K d-1). Black dots in (b), (c), and (d) represent at the 99% confidence level.

Figure 2a illustrates summer precipitation anomalies over China in response to EUBC forcing. The strongest anomalies appeared over the Tibetan Plateau and the Yangtze River valley with maximum values of 1 mm d^-1. Similarly, the rainfall responses over northeastern China and the eastern part of the Yellow River valley were positive with maximum values of 0.6 mm d^-1; most regions of northwestern and southern coastal China showed slightly negative responses. The 500 hPa geopotential height response is shown in Fig. 2b. EUBC induced positive responses over most of China and the western Pacific, with the appearance of positive anomalous values over northern and southern China and relatively weaker positive values over the Yangtze River valley. In comparison with the climatology, the West Pacific Subtropical High (WPSH) strengthened and exhibited northward and westward extension, which lead to an anomalous southwesterly to the west of the subtropical high. This result is manifested in the strong southwesterly anomalies at 850 hPa, as shown in Fig. 2c. EUBC triggered an anticyclone anomaly over the northern South China Sea. Strong anomalous winds up to 1 m s^-1 appear to have entered China from the South China Sea and the Bay of Bengal, bringing more moisture supplied by the ocean, thus accounting for the increase in rainfall over the Yangtze River valley. Consistent with the response at 500 hPa, broadly positive geopotential height anomalies appeared at 100 hPa (Fig. 2d) over most of China and South Asia, with the appearance of positive anomalies over northern and western China, particularly over the Tibetan Plateau, and relatively weaker positive responses over the middle of Asia. Thus, the South Asian high (SAH) tended to intensify and move eastward, which is also favorable for EASM enhancement.

Figure 2

Figure 2 Simulated responses to EUBC for (a) precipitation over China (units: mm d-1), (b) geopotential height at 500 hPa (units: gpm), (c) horizontal wind at 850 hPa (units: m s-1), and (d) geopotential height at 100 hPa (units: gpm). Black dots in (a) and shading in (b) and (d) represent at the 95% confidence level. Dashed lines in (b) and (d) represent 5880 and 16700 climatological contours, calculated from CTL ensemble, which indicate the climatological subtropical high and the climatological South Asian high, respectively. In (c), only regions with at least one component of the wind at the 95% confidence level are shown.

To investigate the mechanism of the remote influence of EUBC on the EASM, we calculated the mid-upper tropospheric (500-200 hPa) mean temperature response to examine heating at that level (Fig. 3a). In contrast with shortwave heating contained within the forcing region, the temperature responses were not confined to Europe but expanded downstream. Stronger responses were observed with distance, particularly over eastern Siberia and the Tibetan Plateau. The BC forcing region in our simulations included only Europe; therefore, these responses may have propagated eastward through dynamical processes such as the Eurasian wavetrain (Ambrizzi et al., 1995). Such results were expected; previous studies also detected remote responses for local BC (Menon et al., 2002; Wang, 2009; Teng et al., 2012; Mahmood and Li, 2013). Menon et al. (2002) determined that remote temperature changes over distant regions could be attributed to initial heating of the tropospheric air over the forcing region, which is then propagated away through dynamical processes. The large area of warming responses over most of the Eurasian continent coupled with the cooling response over the northern South China Sea and western Pacific intensified the temperature gradient between land and ocean, resulting in EASM enhancement.

The probable mechanism by which the EUBC induces the downstream heating anomaly is shown in Fig. 3b, which illustrates wave activity flux (Takaya and Nakamura, 2001) at the 200 hPa level. A discernible Rossby wave train first propagated northeastward from the northern Mediterranean before moving southeastward across Siberia to China, which likely caused EUBC-induced local signals to propagate downstream to East Asia through dynamical processes. The enhanced heating induced by this process overlapped with the downstream heat advection transportation induced by the mid-latitude westerly.

The impact of EUBC on the EASM was investigated by performing an ensemble of sensitivity experiments with GFDL AM2.1. We determined that BC over Europe tends to intensify the EASM. Correspondently, the SAH tends to intensify and move eastward. The WPSH also strengthens and exhibits a northward extension, which leads to anomalous southwesterlies in Southwest China. The wind anomaly brings more moisture from the South China Sea and the Bay of Bengal, accounting for an increase in rainfall over the Yangtze River valley.

Figure 3

Figure 3 Simulated responses to EUBC for (a) air temperature averaged between 500 hPa and 200 hPa (units: K), and (b) Takaya-Nakamura wave activity flux at 200 hPa (units: m2 s-2). Black dots in (a) and shadings in (b) represent at the 90% confidence level.

The strong shortwave radiative forcing induced by EUBC amplifies the shortwave heating rate over the forcing region and induces tropospheric warming that expands to a wide elongated region of Russia and extends to the northern Pacific and North America through dynamical processes. This radiative forcing also induces a cooling response over the northern South China Sea, which increases the land-ocean temperature gradient. These factors combine to enhance the EASM.

The present result implies that EUBC may have played an important role in modulating decadal variation of the EASM. In the early 1990s, the East Asian rainfall anomaly pattern changed from tripolar to dipolar, with substantially increased precipitation occurring in South China (Ding et al., 2007; Liu et al., 2011). Significant anomalous northerlies and easterlies appeared in North and northeastern China (Fig. 4), which implies a significant weakening of the EASM, as noted by Ding et al. (2007). Moreover, a re-evaluation study demonstrated that European BC emission increased from 1.222 Tg C yr^-1 in 1984 to 1.393 Tg C yr^-1 in 1992 before decreasing to 1.283 Tg C yr^-1 in 1995 (Schaap et al., 2004). The modeled intensified rainfall responses between the Huang River and the Yangtze River to EUBC are opposite the observed reduction in central China after the early 1990s. The present result indicates that the decreased EUBC may have contributed to the weakened EASM. Thus, modulation of the decadal variation in the East Asian summer climate can be attributed to local BC combined with EUBC, in addition to other natural factors such as SST (Huang, 2001; Gong and Ho, 2002; Deng et al., 2009; Fu et al., 2009; Fu and Li, 2013), Arctic sea ice concentration (Wu et al., 2009; Li and Leung, 2013), snow cover over the Tibetan Plateau (Zhao and Chen, 2001; Zhang and Tao, 2001; Wu et al., 2012), and forcing by external sulfate aerosols (Wu et al., 2011; Xu, 2001; Zhao et al., 2006).

Figure 4

Figure 4 (a) Observed 850 hPa summer wind differences determined by subtracting the 1979-92 average from the 1993-2006 average (units: m s-1). (b) 850 hPa wind summer climatology observed in 1979-2010 (units: m s-1). The ERA-Interim dataset was used for analysis. Shading in (a) represents at the 95% confidence level.